Danforth's Obstetrics and Gynecology [9 ed.]

Citation preview

Danforth's Obstetrics and Gynecology, 9th Ed: James R., Md. Scott, Ronald S., Md. Gibbs, Beth Y., Md. Karlan, Arthur F., Md. Haney, David N. Danforth By Lippincott Williams & Wilkins Publishers; 9th edition (August 2003)

Danforth's Obstetrics and Gynecology CONTENTS Editors Contributors Preface Farewell

Chapter 1 Prenatal Care Vern L. Katz Chapter 2 Normal Labor, Delivery, Newborn Care, and Puerperium Dwight J. Rouse and Elaine St. John Chapter 3 Obstetric Analgesia and Anesthesia Joy L. Hawkins Chapter 4 Early Pregnancy Loss D. Ware Branch and James R. Scott Chapter 5 Ectopic Pregnancy Michael J. Heard and John E. Buster Chapter 6 Genetics and Prenatal Diagnosis Kenneth Ward Chapter 7 Drugs in Pregnancy Jerome Yankowitz Chapter 8 Ultrasound in Obstetrics Keith H. Nelson and Lewis H. Nelson, III Chapter 9 Fetal Monitoring Catherine Y. Spong Chapter 10 Preterm Labor and Delivery Debra A. Guinn and Ronald S. Gibbs Chapter 11 Premature Rupture of the Membranes Ronald S. Gibbs Chapter 12 Intrauterine Growth Restriction Henry L. Galan and John C. Hobbins Chapter 13 Prolonged Pregnancy Lisa Moore and James N. Martin, Jr. Chapter 14 Multiple Gestation Roger B. Newman Chapter 15 Diabetes Mellitus and Pregnancy E. Albert Reece and Carol J. Homko Chapter 16 Hypertensive Disorders of Pregnancy Julie S. Moldenhauer and Baha M. Sibai Chapter 17 Medical and Surgical Complications of Pregnancy Maureen P. Malee Chapter 18 Immunologic Disorders in Pregnancy T. Flint Porter, Morgan Peltier, and D. Ware Branch Chapter 19 Obstetric and Perinatal Infections Howard L. Minkoff and Ronald S. Gibbs Chapter 20 Placenta Previa and Abruption Helen H. Kay Chapter 21 Breech, Other Malpresentations, and Umbilical Cord Complications Dwight P. Cruikshank Chapter 22 Complications of Labor Donald J. Dudley Chapter 23 Operative Vaginal Delivery Michael A. Belfort Chapter 24 Cesarean Delivery T. Flint Porter and James R. Scott Chapter 25 Critical Care Obstetrics Steven L. Clark Chapter 26 Office Gynecology William R. Keye, Jr. and Marian D. Damewood Chapter 27 Office Surgical Procedures Deborah Manzi-Smith and Charles C. Coddington, III Chapter 28 Gynecologic Ultrasound Frederick Larsen Chapter 29 Pediatric and Adolescent Gynecology Ann J. Davis Chapter 30 Contraception Herbert B. Peterson, Kathryn M. Curtis, Olav Meirik, and Catherine d'Arcangues Chapter 31 Induced Abortion Suzanne R. Trupin Chapter 32 Pelvic Infections and Sexually Transmitted Diseases David A. Eschenbach Chapter 33 Benign Vulvovaginal Disorders

Edward J. Wilkinson Chapter 34 Amenorrhea Marc A. Fritz Chapter 35 Abnormal Uterine Bleeding Steve N. London Chapter 36 Premenstrual Syndrome Robert L. Reid Chapter 37 Androgen Excess Disorders Richard S. Legro and Ricardo Azziz Chapter 38 Infertility Julia Johnson Chapter 39 Assisted Reproductive Technology Spencer S. Richlin, Aida Shanti, and Ana A. Murphy Chapter 40 Endometriosis Robert S. Schenken Chapter 41 Menopause Marcelle I. Cedars and Michele Evans Chapter 42 Human Sexuality and Female Sexual Dysfunction Laurie Jane McKenzie and Sandra Ann Carson Chapter 43 Chronic Pelvic Pain Howard T. Sharp Chapter 44 Perioperative Evaluation F. Joseph Kelly, Phillip Y. Roland, and James W. Orr, Jr. Chapter 45 Pelvic Organ Prolapse John O.L. DeLancey and Kris Strohbehn Chapter 46 Laparoscopic Surgery Joseph S. Sanfilippo and Lisa M. Roberts Chapter 47 Hysteroscopic Surgery R. Stanford Williams Chapter 48 Incontinence Dee E. Fenner Chapter 49 Leiomyomata Arthur F. Haney Chapter 50 Disorders of the Breast Mary L. Gemignani Chapter 51 Vulvar and Vaginal Cancer Natalie S. Gould, Joan L. Walker, and Robert S. Mannel Chapter 52 Cervical Cancer Robert E. Bristow Chapter 53 Management of the Abnormal Pap Smear Ellen E. Sheets Chapter 54 Uterine Cancer David G. Mutch Chapter 55 Neoplasms of the Ovary and Fallopian Tube Ilana Cass and Beth Y. Karlan Chapter 56 Management of the Adnexal Mass Linda Van Le Chapter 57 Gestational Trophoblastic Neoplasms Andrew J. Li and Beth Y. Karlan Chapter 58 Medical Ethics Frank A. Chervenak and Laurence B. McCullough Chapter 59 Evidence-Based Approach to Obstetrics and Gynecology Herbert B. Peterson, Susan F. Meikle, and R. Brian Haynes Appendix 1 Color Plate

EDITORS Edited by JAMES R. SCOTT, MD Professor, Department of Obstetrics and Gynecology University of Utah School of Medicine Salt Lake City, Utah RONALD S. GIBBS, MD Professor and Chair Department of Obstetrics and Gynecology; E. Stewart Taylor Chair in Obstetrics and Gynecology University of Colorado Health Sciences Center Denver, Colorado BETH Y. KARLAN, M.D. Professor, Obstetrics and Gynecology Geffen School of Medicine at UCLA Los Angeles, California; Director Division of Gynecologic Oncology Cedars-Sinai Medical Center ARTHUR F. HANEY, MD The Catherine Lindsay Dobson Professor and Chairman Department of Obstetrics and Gynecology Division of Biologic Sciences and the Pritzker School of Medicine The University of Chicago, Chicago Illinois LISA MCALLISTER Acquisitions Editor JENNY KIM Developmental Editor PATRICK CARR Production Editor COLIN WARNOCK Manufacturing Manager DAVID LEVY Cover Designer SANDI FRANK Indexer

Contributors Ricardo Azziz, MD, MPH, MBA Chairman, Department of Obstetrics and Gynecology, Cedars-Sinai Medical Center; Professor and Vice-Chair, Department of Obstetrics and Gynecology, University of California, Los Angeles, Los Angeles, California Michael A. Belfort, MD, PhD Professor, Department of Obstetrics and Gynecology, University of Utah, Salt Lake City, Utah; Director, Maternal–Fetal Medicine, Utah Valley Regional Medical Center, Provo, Utah D. Ware Branch, MD Professor, Department of Obstetrics and Gynecology, University of Utah Medical Center, Salt Lake City, Utah Robert E. Bristow, MD Associate Professor, The Kelly Gynecologic Oncology Service, Department of Gynecology and Obstetrics, The Johns Hopkins Medical Institutions, Baltimore, Maryland John E. Buster, MD Professor, Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas Sandra Ann Carson, MD Professor, Department of Obstetrics and Gynecology, Baylor College of Medicine; Director, Baylor Assisted Reproductive Technology, The Methodist Hospital, St. Luke's Hospital, Houston, Texas Ilana Cass, MD Division of Gynecologic Oncology, Cedars-Sinai Medical Center; Assistant Professor, David Geffen School of Medicine at UCLA, Los Angeles, California Marcelle I. Cedars, MD Professor, Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California at San Francisco; Director, Center for Reproductive Health, University of California at San Francisco, San Francisco, California Frank A. Chervenak, MD Professor and Chairman, Department of Obstetrics and Gynecology, Joan and Sanford I. Weill Medical College; Obstetrician and Gynecologist-in-Chief, Department of Obstetrics and Gynecology, New York Presbyterian Hospital, New York, New York Steven L. Clark, MD Professor, Department of Obstetrics and Gynecology, Unviversity of Utah School of Medicine, Department of Perinatology, LDS Hospital, Salt Lake City, Utah Charles C. Coddington, III, MD Director, Department of Obstetrics and Gynecology, Denver Health Medical Center; Professor, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Denver, Colorado Dwight P. Cruikshank, MD Chairman and Jack and Elaine D. Klieger Professor, Department of Obstetrics and Gynecology, Medical College of Wisconsin; Director, Froedtert Memorial Lutheran Hospital, Milwaukee, Wisconsin Kathryn M. Curtis, PhD Division of Reproductive Health, Centers for Disease Control and Prevention, Atlanta, Georgia Marian D. Damewood, MD Clinical Professor of Obstetrics & Gynecology, Johns Hopkins University School of Medicine; Chairman, Department of Obstetrics & Gynecology; Director, Women and Children's Services, York Hospital/WellSpan Health System, York, Pennsylvania Catherine d'Arcangues, PhD, MD Department of Reproductive Health and Research, World Health Organization, Geneva, Switzerland Ann J. Davis, MD Beth Israel Deaconess Hospital, Boston, Massachusetts John O.L. DeLancey, MD Norman F. Miller Professor of Gynecology; Director of Pelvic Floor Research, The University of Michigan, Ann Arbor, Michigan Donald J. Dudley, MD Professor, Department of Obstetrics and Gynecology, University of Texas Health Science Center at San Antonio; Teaching Faculty, Department of Obstetrics and Gynecology, University Hospital, San Antonio, Texas David A. Eschenbach, MD Professor, Department of OB & GYN, University of Washington School of Medicine, Seattle, Washington Michele Evans Clinical Fellow, Division of Reproductive Endocrinology and Infertility, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California—WHCRC, San Francisco, California Dee E. Fenner, MD Associate Professor, Department of Obstetrics and Gynecology, University of Michigan Medical Center; Director of Gynecology, Department of Obstetrics and Gynecology, University of Michigan, Ann Arbor, Michigan Marc A. Fritz, MD Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina Henry L. Galan, MD Assistant Professor, Department of Obstetrics & Gynecology, University Hospital, University of Colorado School of Medicine, Denver, Colorado Mary L. Gemignani, MD Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York Ronald S. Gibbs, MD Professor and Chair, Department of Obstetrics and Gynecology; E. Stewart Taylor Chair in Obstetrics and Gynecology, University of Colorado School of Medicine, Denver, Colorado

Natalie S. Gould, MD Fellow and Clinical Instructor, Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma Debra A. Guinn, MD Assistant Professor, Department of Obstetrics & Gynecology, University of Colorado School of Medicine, Denver Health Medical Center, Denver, Colorado Arthur F. Haney, MD Professor and Chairman, Department of Obstetrics and Gynecology, The Pritzker School of Medicine, University of Chicago, Chicago, Illinois Joy L. Hawkins, MD Professor, Department of Anesthesiology, University of Colorado School of Medicine; Director of Obstetric Anesthesia, University of Colorado Hospital, Denver, Colorado R. Brian Haynes, MD, Ph.D. Professor and Chair, Department of Clinical Epidemiology and Biostatistics, McMaster University; Attending Staff, Department of Medicine, Hamilton Health Sciences, Hamilton, Ontario, Canada Michael J. Heard, MD, FACOG Clinical Assistant Professor, Department of Obstetrics & Gynecology, University of Texas Health Science Center; Medical Director, Houston Reproductive Endocrinology and Infertility, Houston, Texas John C. Hobbins, MD Professor and Chief of Obstetrics, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Denver, Colorado Carol J. Homko, RN, PhD, CDE Research Assistant Professor, Temple University School of Medicine; Nurse Manager, General Clinical Research Center, Temple University Hospital, Philadelphia, Pennsylvania Julia V. Johnson, MD University of Vermont, Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, Burlington, Vermont Beth Y. Karlan, MD Professor of Obstetrics and Gynecology, David Geffen School of Medicine at UCLA; Director, Women's Cancer Research Institute and Division of Gynecologic Oncology, Cedars-Sinai Medical Center, Los Angeles, California Vern L. Katz, MD Clinical Associate Professor, Department of Obstetrics & Gynecology, Oregon Health Sciences University, Portland, Oregon; Medical Director, Perinatal Services, Sacred Heart Medical Center, Eugene, Oregon Helen H. Kay, MD Professor and Chair, Department of Obstetrics and Gynecology, University of Arkansas for Medical Sciences, Little Rock, Arkansas F. Joseph Kelly, MD Clinical Assistant Professor, Department of Obstetrics and Gynecology, University of South Florida, Tampa, Florida, Florida Gynecologic Oncology, Lee Cancer Care, Lee Memorial Health Systems, Fort Myers, Florida William R. Keye, Jr., MD Director, Division of Reproductive Endocrinology and Infertility, Beaumont Medical Services, Royal Oak, Michigan Frederick W. Larsen, MD Associate, Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Duke University Medical Center, Durham, North Carolina Richard S. Legro, MD Department of Obstetrics and Gynecology, Pennsylvania State University, College of Medicine, Hershey, Pennsylvania Andrew John Li, MD Division of Gynecologic Oncology, Cedars-Sinai Medical Center, Los Angeles, California; Assistant Professor, Department of Obstetrics and Gynecology, David Geffen School of Medicine at UCLA Steve N. London, MD Chairman, Department of Obstetrics and Gynecology, Louisiana State University Health Sciences Center at Shreveport, Shreveport, Louisiana Maureen P. Malee, MD, PhD Associate Professor, Department of Obstetrics & Gynecology, Brown University School of Medicine, Maternal–Fetal Medicine Specialist, Department of Obstetrics & Gynecology, Women & Infants Hospital, Providence, Rhode Island Robert S. Mannel, MD Professor and Chairman, Department of Obstetrics and Gynecology, University of Oklahoma Health Sciences Center; Chairman, Department of Obstetrics and Gynecology, Oklahoma University Medical Center, Oklahoma City, Oklahoma Deborah L. Manzi-Smith, MD Associate Professor, Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, University of Colorado Health Sciences Center; Associate Professor, Department of Obstetrics and Gynecology, Advanced Reproductive Medicine, University of Colorado Hospital, Denver, Colorado James N. Martin, Jr., MD Professor, Department of Obstetrics and Gynecology, University of Mississippi School of Medicine; Director, Maternal-Fetal Medicine, University of Mississippi Medical Center, Jackson, Mississippi Laurence B. McCullough, Ph.D. Professor of Medicine and Medical Ethics, Center for Medical Ethics and Health Policy, Baylor College of Medicine, Houston, Texas Laurie J. McKenzie, MD Fellow, Department of Obstetrics and Gynecology, Baylor College of Medicine; Faculty, Department of Obstetrics and Gynecology, Ben Taub Hospital, Houston, Texas Susan Meikle, MD, MSPH Medical Officer, Agency for Healthcare Research and Quality, Rockville, Maryland Olav Meirik, MD, PhD Instituto Chileno de Medicina Reproductiva, Santiago, Chile

Howard L. Minkoff, MD Distinguished Professor, Department of Obstetrics & Gynecology, State University of New York Health Science Center at Downstate; Chairman, Department of Obstetrics & Gynecology, Maimonides Medical Center, Brooklyn, New York Julie S. Moldenhauer, MD Division of Reproductive Genetics, Detroit Medical Center/Wayne State University, Detroit, Michigan Lisa E. Moore, MD Fellow, Department of Maternal–Fetal Medicine, University of Mississippi; Fellow, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, Jackson, Mississippi Ana A. Murphy, MD Anne Bates Winship Leach Professor; Division Director, Reproductive Endocrinology and Infertility, Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, Georgia David G. Mutch, MD Ira and Judith Gall Professor of Obstetrics and Gynecology; Director, Division of Gynecologic Oncology, Washington University School of Medicine, St. Louis, Missouri Keith H. Nelson, MD Clinical Assistant Professor, Department of Obstetrics & Gynecology, East Carolina University, Brody School of Medicine, Greenville, North Carolina Lewis H. Nelson, III, MD, RDMS Professor of Obstetrics and Gynecology; Associate Dean of Student Admissions, Wake Forest University School of Medicine, Department of Obstetrics & Gynecology, Wake Forest University Baptist Medical Center, Winston-Salem, North Carolina Roger B. Newman, MD Professor and Vice Chair for Academic Affairs, Department of Obstetrics & Gynecology, Medical University of South Carolina, Charleston, South Carolina James W. Orr, Jr., MD Clinical Professor, Department of Obstetrics and Gynecology, University of South Florida, Tampa, Florida; Director, Florida Gynecologic Oncology, Lee Cancer Care, Lee Memorial Health Systems, Fort Myers, Florida Morgan Peltier, MS, PhD Instructor, Department of Obstetrics and Gynecology, University of Utah Health Sciences, Salt Lake City, Utah Herbert B. Peterson, MD Coordinator, Promoting Family Planning, Department of Reproductive Health and Research, World Health Organization, Geneva, Switzerland; Clinical Professor, Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina T. Flint Porter, MD Assistant Professor, Department of Obstetrics & Gynecology, University of Utah; Director, Department of Maternal–Fetal Medicine, LDS Hospital, Salt Lake City, Utah E. Albert Reece, MD Professor and Chair, Department of Obstetrics & Gynecology, Temple University School of Medicine, Philadelphia, Philadelphia Robert L. Reid, MD Department of Obstetrics and Gynecology, Kingston General Hospital, Kingston, Ontario, Canada Spencer S. Richlin, MD Assistant Professor, Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Loma Linda University School of Medicine, Loma Linda, California Lisa M. Roberts, MD Gynecology & Laparoscopy Surgeons, PC, Raleigh, North Carolina Phillip Y. Roland, MD Assistant Clinical Professor, Department of Obstetrics and Gynecology, University of South Florida, Tampa, Florida, Florida Gynecologic Oncology, Lee Cancer Care, Lee Memorial Health Systems, Fort Myers, Florida Dwight J. Rouse, MD Associate Professor, Department of Obstetrics and Gynecology, University of Alabama at Birmingham; Medical Director, Center for Research in Women's Health, University of Alabama Hospital, Birmingham, Alabama Joseph S. Sanfilippo, MD, MBA Professor, Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh; Vice-Chairman, Department of Reproductive Sciences, Magee Women's Hospital, Pittsburgh, Pennsylvania Robert S. Schenken, MD Professor and Chairman, Department of Obstetrics and Gynecology, University of Texas Health Sciences Center at San Antonio, San Antonio, Texas James R. Scott, MD Professor, Department of Obstetrics & Gynecology, University of Utah Medical Center, Salt Lake City, Utah Aida Shanti, MD Associate Professor; Director of In Vitro Fertilization, Division of Reproductive Endocrinology and Infertility, Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, Georgia Howard T. Sharp, MD Associate Professor, Department of Obstetrics and Gynecology, University of Utah School of Medicine; Associate Professor, Department of Obstetrics and Gynecology, University of Utah Medical Center, Salt Lake City, Utah Ellen E. Sheets, MD Associate Professor, Department of Obstetrics and Gynecology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts Baha M. Sibai, MD Professor and Chairman, Department of Obstetrics and Gynecology, University of Cincinnati College of Medicine, Cincinnati, Ohio Catherine Y. Spong, MD Department of Obstetrics & Gynecology, Georgetwon University Medical Center, Washington, DC; Chief, Pregnancy and Perinatology Branch, National Institute of Child Health and Human Development, Bethesda, Maryland

Elaine Barefield St. John, MD Associate Professor, Department of Pediatrics, University of Alabama at Birmingham; Attending Neonatologist, Department of Pediatrics, University of Alabama Hospital, Birmingham, Alabama Kris Strohbehn, MD Associate Professor, Dartmouth College Medical School; Director, Division of Urogynecology/Reconstructive Pelvic Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire Suzanne R. Trupin, MD Department of Obstetrics and Gynecology, University of Illinois–Urbana–Champaign, Champaign, Illinois Linda Van Le, MD Professor, Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina Joan L. Walker, MD Associate Professor and Chief, Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, Oklahoma City, Oklahoma Kenneth Ward, MD Professor and Chair, Department of Obstetrics, Gynecology, and Women's Health, John A. Burns School of Medicine, University of Hawaii, Honolulu, Hawaii Edward J. Wilkinson, MD, FACOG, FCAP Professor and Vice Chairman, Department of Pathology and Laboratory Medicine; Adjunct Professor of Obstetrics and Gynecology; Director and Chief, Division of Anatomic Pathology, University of Florida College of Medicine, Gainesville, Florida R. Stanford Williams, MD Professor and Associate Chairman, Department of Obstetrics and Gynecology, University of Florida; Medical Director, Department of Obstetrics and Gynecology, Shands Hospital at University of Florida, Gainesville, Florida Jerome Yankowitz, MD Professor, Department of Obstetrics and Gynecology, Roy J. and Lucille A. Carver College of Medicine; Director, Division of Maternal—Fetal Medicine, Department of Obstetrics & Gynecology, University of Iowa Hospitals and Clinics, Iowa City, Iowa

Preface Now in its ninth edition, Danforth's Obstetrics and Gynecology has been widely recognized as a standard textbook for practicing physicians, residents, medical students, and nurses. It is gratifying that previous editions have been translated into several languages and have enjoyed worldwide acceptance. For many practitioners, this text has been the basis of learning and an essential reference in our beloved specialty. The ninth edition has a new look. Three of the four editors of the eighth edition (Drs. DiSaia, Hammond, and Spellacy) have now turned over the reins to new editors. We are delighted that Jim Scott has continued to serve as an editor for the ninth edition to bring perspective, experience, and continuity. In planning this edition, we carefully reviewed the topic of each chapter and its contents. Consisting of 59 chapters, the ninth edition is now focused to provide a practical, clinically-useful text that covers all of obstetrics and gynecology in one volume. In previous editions, basic science chapters were free standing. In the ninth edition, the basic science concepts have been incorporated into pertinent, more clinically oriented chapters. We have strived to incorporate evidence-based medicine in each of the chapters and have strengthened each of the component chapters. We are also pleased to welcome new contributing authors who are all experts in their field and we thank, also, the contributing authors who have updated and revised their previous chapters. For ready learning, each chapter contains summary points, conveniently placed at the end of the chapter. In addition, rather than using a heavily referenced format, each author has included a concise list of up-to-date suggested readings at the end of each chapter. The editors are most grateful for the contributions of the administrative staff at our respective universities: Jane Cook and Barb Carpenter at the University of Colorado; Sheryl Martin at the University of Utah; and Phyllis Lopez at Cedars Sinai Medical Center. We also wish to thank Lisa McAllister, Sonya Seigafuse, and Jenny Kim at Lippincott Williams & Wilkins for their great editorial assistance, support, encouragement—and patience during the gestation of this work. James R. Scott, MD Ronald S. Gibbs, MD Beth Y. Karlan, MD Arthur F. Haney, MD

Farewell With this ninth edition, I end my editorship of Danforth's Obstetrics and Gynecology. When Dr. Danforth first asked me almost 20 years ago to serve as co-editor for the fifth edition, it was only after careful thought and preparation that I was willing to take on such a monumental task. We also added three co-editors I considered to be the leading authorities in each subspecialty: Drs. Philip DiSaia (Gynecologic Oncology), Charles Hammond (Reproductive Endocrinology), and William Spellacy (Maternal-Fetal Medicine). It was a privilege and pleasure to work with them as we edited the sixth, seventh and eighth editions. Now it is time for a transition. As Sparky Anderson once said as he stepped down as manager of the Detroit Tigers after 17 seasons, “It's just time.” Dr. Ronald Gibbs has served as the lead editor for this edition, and Drs. Beth Karlan and Arthur Haney are the other new editors. Each was carefully chosen as a highly respected and nationally recognized expert in his/her subspecialty. They have done a superb job, and I am confident that the book is in good hands for future editions. Dr. Danforth would have agreed. James R. Scott, MD

Chapter 1 Prenatal Care Danforth’s Obstetrics and Gynecology

Chapter 1 Vern L. Katz

Prenatal Care

PRIMARY AND PRECONCEPTION CARE Philosophy Content of the Preconception Visit Risk Assessment Interventions INITIAL PRENATAL VISIT Diagnosis of Pregnancy Gestational Age Physical Examination Laboratory Evaluation Discussion with the Patient ROUTINE ANTEPARTUM SURVEILLANCE Rationale for Routine Prenatal Care Timing and Frequency of Visits Content of Subsequent Prenatal Visits CONCERNS AND QUESTIONS PARTICULAR TO PREGNANCY Psychological Nutrition Exercise Nausea and Vomiting Ptyalism Heartburn Constipation Fatigue Varicosities and Hemorrhoids Leg Cramps Backache Round Ligament Pain Headache Emotional Changes Sexual Relations Employment Urinary Frequency Skin Changes Leukorrhea Syncope X-rays/Ionizing Radiation Travel Immunizations PREPARING FOR CHILDBIRTH Prenatal Education Classes Signs of Labor CONCLUSION SUMMARY POINTS SUGGESTED READINGS Appendix I Appendix II Appendix III

This chapter reviews the principles and specifics of prenatal care. The time period from the periconceptual visit through labor and delivery is, for most women, the single greatest psychological transition that a woman undergoes. During these months, the obstetrician, family physician, or midwife serves a much larger role than just as health care provider. Our role is not only to assess the health of the mother and fetus, prescribe interventions, and try to influence behaviors, but also to advise and help our patients as they undergo this challenging psychological passage. Over the three trimesters of pregnancy a woman must develop new aspects of identity. Her self-image develops an additional sense of femininity beyond what was developed at puberty, and a maternal self-concept must develop as well. Reba Rubin, in her works on the maternal experience, describes a new mother's psychological tasks as the woman grows into her new role. These tasks include: accepting a new body image, which is often in conflict with accepted societal views of attractiveness accepting the child that is growing inside her reordering her identity with her mother, her friends, and the father of the pregnancy symbolically finding acceptance and safety for her child (i.e., making a new home). For many women with good social support these tasks are anticipated and desired roles, which bring a sense of fulfillment. For other women, some or all of these tasks are unanticipated and difficult. The obstetrician, in multiple ways, helps the mother through these transitions, while at the same time ensures the physical health of both patients (mother and fetus). Many aspects of prenatal care have grown from their original role of health promotion to ritualized traditions that have acquired symbolic value in helping women and their families adapt to these psychological transitions. For example, studies have found that for women of average weight, the practice of weighing a woman during each visit has minimal medical value. Yet, if the nurse forgets to weigh a patient, that woman usually remarks quite quickly about having her weight taken. Another example is the routine ultrasound. This is now a demand ritual. At this visit, a mother will usually bring several female family members or friends, to see the sonogram. The new mother not only uses the sonogram to bond with her child, but also shows the baby to the other women around her for their acceptance. Throughout the world, cultures and subcultures view prenatal care differently, but most all hold it with respect. A woman might miss her annual Pap smear, but she rarely misses her prenatal visit. Many authors have commented on the special relationship of obstetrician and mother, and the window of opportunity to affect healthy behaviors, such as smoking cessation or nutrition. Compared to other physician–patient relationships, the time period of pregnancy allows much greater involvement of the physician in the patient's life. It is with this developmental background that prenatal care has evolved in the last century. In the United States, the first organized prenatal care programs began in 1901 with home nurse visits. The first prenatal clinic was established in 1911. It is not surprising that this focus on maternal and infant health occurred as a direct outgrowth of the women's suffrage movement. The emergence of the physician-obstetrician as the primary caregiver for women of reproductive age also took place during the last century. This role has become one of partnership with allied health care providers, including nurses, nurse midwives, nurse practitioners, nutritionists, and social workers. This team not only may provide advice and medical treatment, but also serves as the patient's advocate. Our current emphasis on prenatal care stems from historic pronouncements and retrospective analyses that concluded that women who receive prenatal care have less fetal, infant, and maternal morbidity and mortality. However, a conclusive scientific foundation is lacking for the content of prenatal care and the relationship of its components to good outcomes. As technology flourishes and resources dwindle, it has become increasingly important to obtain scientifically based evidence demonstrating which components of prenatal care are clinically appropriate, cost-effective, and deserving of preferential funding. At this time, the optimal content and delivery of prenatal care remain the subject of discussion and debate. Given the increasing number of tools of prenatal assessment, the current consensus is that the best prenatal care is individualized for the specific needs of the mother.

PRIMARY AND PRECONCEPTION CARE Philosophy Care of the woman of reproductive age for both preconception and pregnancy is integrated and accessible, focuses on the majority of personal health care needs, represents a sustained partnership between patient and provider, and occurs within the context of family and community. For many women, pregnancy care occurs as a part of the continuum in a long-term relationship with the health care provider. The first visit may be a preconception visit or may occur after the woman is pregnant. The content of the visits is similar. If a woman is seen for a preconceptual visit, many issues need not be readdressed when she becomes pregnant. This next section focuses on the preconception visit. Content of the Preconception Visit The preconception visit is a focused visit for the woman who is planning to or is considering becoming pregnant in the near future. The content of this interval visit includes a history, physical examination, risk assessment and intervention, selected laboratory testing based on the patient's age and the results of the foregoing evaluation, ongoing management of medical conditions, and a plan of care. A purposeful discussion of contraception, sexually transmitted disease prevention, and timing of conception is appropriate. Timely administration of routine immunizations, educational counseling, and advice complete the visit. Risk Assessment A goal specific to the preconception interval visit is the systematic identification of potential risks to pregnancy and the implementation of early intervention as necessary. These risks fall into several categories, described in the following sections. Unalterable Factors These are preexisting factors and cannot at present be altered in any medical way by clinical intervention. These include the patient's height, age, reproductive history, ethnicity, educational level, socioeconomic status, and genetic composition. Factors Benefiting from Early Intervention Conditions that should or could be modified before pregnancy is attempted include poor nutrition, an underweight or obese body mass index (BMI), and poorly controlled medical diseases such as diabetes mellitus, asthma, epilepsy, phenylketonuria, hypertension, and thyroid disease. Some prescription medications that are known teratogens should be discontinued and appropriate substitutions made. These include medications such as isotretinoin (Accutane), warfarin sodium (Coumadin), certain anticonvulsants, and angiotensin-converting enzyme inhibitors. Many medications are safe and patients may be assured of medications for asthma, most antihistamines, and antidepressants. Determining the status of a patient's immunity to rubella, varicella, and hepatitis is appropriate during the preconception visit. In high-risk populations or endemic geographic areas, patients should be assessed for active tuberculosis with skin testing and chest x-ray. Social Risk Factors Inquiry should be made regarding occupational hazards involving exposure to toxins such as lead, mercury and other atom heavy metals, and organic solvents (both liquid and vapors). Hazards in the home, such as exposure to toxoplasmosis or toxic chemicals (asbestos, pesticides), are important to identify. If a woman uses well water, it may be assessed for acidity, lead, and copper. Family violence is a particularly important household hazard. Nonjudgmental, open-ended evaluation should be applied. Judith MacFarland has recommended questions such as “Are you in a relationship in which you are being hit, kicked, slapped, or threatened?” “Do you feel threatened?” “Have you been forced to do things against your will?” These questions should be asked again at the first prenatal visit. Some studies have suggested that a written questionnaire, in addition to oral questions, will allow for greater identification of domestic abuse. Risky Health Habits The use of illicit drugs or abuse of alcohol represents a significant health hazard to pregnancy. Alcohol is a known teratogen. There is no consensus on the correlation between the quantity of alcohol consumed and the manifestation of adverse fetal effects. Therefore, the best advice to women who wish to become pregnant is to stop drinking. The T-A-C-E screen for alcohol abuse has been well studied. The letters stand for four questions asked in a nonjudgmental manner: 1. 2. 3. 4.

T—“How much you drink to feel drunk?” ( tolerance) A—“Does your drinking annoy anyone?” C—“Has anyone told you to cut down?” E—“Do you drink in the morning to feel better?” ( eye-opener).

Smoking cigarettes is associated with adverse pregnancy outcomes including low birth weight, premature birth, and perinatal death. Smoking by both the pregnant woman and members of the household should be avoided during pregnancy and, preferably, not resumed postpartum. The relative risk of intrauterine growth restriction (IUGR) among pregnant smokers has been calculated at 2.2 to 4.2. Because of the morbidity associated with smoking, various methods to assist women to quit smoking should be encouraged prior to pregnancy. Numerous interventions are available ( Table 1.1). Use of the transdermal nicotine patch in pregnancy is thought to be preferable to smoking. One benefit of using a nicotine patch is the elimination of exposure to other toxins such as carbon monoxide inhaled in cigarette smoke. Its theoretical risk is that it creates a constant blood level of nicotine, as opposed to the vacillations that occur with smoking. Similarly, all illicit drugs have the potential of harming the pregnancy.

TABLE 1.1. Smoking cessation in pregnancy Other behaviors that should be avoided are those that promote exposure to sexually transmitted and other infectious diseases. These include unprotected sexual intercourse in a nonmonogamous relationship and the sharing of needles between addicts. Interventions The final phase of the preconception visit involves specific interventions derived from the information obtained during the history, physical examination, and risk assessment phases. The specific interventions may include immunization against rubella, varicella, or hepatitis, changes in prescribed medications, behavior modification, genetic screening for such conditions as Tay–Sachs disease, cystic fibrosis and sickle cell anemia, and nutritional and physical activity recommendations. During the physical examination, evaluation of thyroid and breast examinations is important. If a woman is 35 years of age or older, a screening mammogram should be ordered, since as much as two and a half years may pass before she will be able to have one (mammograms have significantly decreased sensitivity during pregnancy and for up to six months after lactation). If a woman has a family history of premenopausal breast cancer, a mammogram may be considered at younger ages. If a Pap smear has not been done within a year, this test should be repeated at this time. Additionally, it is valuable at this visit to examine the patient's skin. The incidence of melanoma is increasing faster than any other malignancy in the United States. The obstetrician has the unique opportunity to assess and teach at this visit regarding this cancer. Folic acid as a supplement can reduce the occurrence and recurrence of neural tube defects, and may reduce the risk of other birth defects as well. Women who have had a previous pregnancy affected by neural tube defects should take 4 mg of folic acid per day, starting 4 weeks prior to conception through the first trimester. For all other women of reproductive age who have the potential to become pregnant, 1 mg of folic acid should be prescribed. Approximately 20% of all pregnant women are battered during their pregnancy. About one half of women who are physically abused prior to pregnancy continue to be battered during pregnancy. For some women, the violence begins with pregnancy. All such patients require information regarding their immediate safety and referrals for counseling and support. Some patients purposely initiate a preconception visit to determine whether or not a preexisting medical condition is an absolute contraindication to pregnancy. Pulmonary hypertension, for example, although rare, is associated with up to a 50% maternal mortality and a greater than 40% fetal mortality. It is possible to obtain epidemiologic studies that provide statistics on the morbidity and mortality for mother and fetus for most disease states. These cannot, however, provide specific data for any one patient with her own unique set of medical, demographic, and social variables. Many patients who make these inquiries will benefit by reading the relevant medical materials themselves and by obtaining more than one opinion. Consultation with other medical specialists may be necessary. For example, women with orthopedic problems often inquire about vaginal delivery. Another common concern is advanced maternal age. Specific risks of increased rates of aneuploidy and

miscarriage should be discussed. Women over 40 have been found to have higher rates of low birth weight, preterm birth, and operative delivery. It is also important to discuss how and when to discontinue contraceptive measures. Patients using medroxy-progesterone acetate (Depo-Provera) injections may experience a delay of several months in the return of regular ovulatory menstrual cycles. An intrauterine device (IUD) may be removed at any time in the cycle. It should be removed as soon as conception is considered, since removal during pregnancy (although much better than leaving in place) is associated with a higher rate of pregnancy loss. Likewise, birth control pills may be discontinued at any time prior to attempting conception. Many physicians believe that discontinuing the use of oral contraceptives for one cycle allows better growth of the endometrium. Although definitive evidence is lacking, the thought is that there may be better implantation. The patient should be advised to seek early prenatal care by making an appointment after missed menses or on confirmation of pregnancy by a home pregnancy test. Unfortunately, in the United States, only 75% of pregnant women receive prenatal care beginning in the first trimester. Ongoing barriers to prenatal care access include lack of money or insurance to pay for care, system under-capacity for appointments, and inadequate transportation ( Table 1.2).

TABLE 1.2. Probability of pregnancy based on presumptive signs

INITIAL PRENATAL VISIT This visit represents the first detailed assessment of the pregnant patient. The optimal timing of this visit may vary. For women who have not undergone the comprehensive preconception visit, prenatal visits should begin as soon as pregnancy is recognized. For these women, much of the content of the preconception visit will need to be addressed at this time—for example, screening for domestic abuse and alcohol use. All other women should be seen by about 8 menstrual weeks (6 weeks after conception) of gestation. For all patients, the appropriate content of prenatal care and the first prenatal visit is contained in the antepartum record published by the American College of Obstetrics and Gynecology (ACOG) (see Appendix I). Identifying data, a menstrual history, and a pregnancy history are obtained. Past medical, surgical, and social history are recorded, along with symptoms of pregnancy. A focused genetic screen, infection history, and risk status evaluation are performed. Diagnosis of Pregnancy Evaluation of the signs and symptoms associated with the presumptive diagnosis of pregnancy (see Table 1.2), while a useful adjunct, has been largely superseded by the widely available urine pregnancy test. The detection of greater than 35 mIU of human chorionic gonadotropin (hCG) in the first morning void has a very high specificity for pregnancy. Other tests for confirming the presence of pregnancy include a positive serum ß-hCG and demonstration of the fetal heart by either auscultation or ultrasound. Using a transvaginal probe, fetal cardiac activity should be seen by postconception week 3. Ultrasound imaging is not routinely indicated to diagnose pregnancy but may be used in the evaluation of a patient who is at increased risk for ectopic pregnancy or threatened abortion. In conjunction with early quantitative serum ß-hCG assessments, these conditions can be clearly differentiated from a normal intrauterine pregnancy and timely therapy initiated. Gestational Age The Nägele rule is commonly applied in calculating an estimated date of confinement (EDC). Using the date of the patient's last menstrual period minus 3 months, plus 1 week and 1 year, it is based on the assumptions that a normal gestation is 280 days and that all patients have 28-day menstrual cycles. Although several studies have found the average length of gestation for primiparous women to be 282 to 283 days, for convention 280 days is the currently accepted average gestation. After adjustment for a patient's actual cycle length, natality statistics indicate that the majority of pregnancies deliver within 2 weeks before or after this estimated date. During prenatal care, the week of gestation can be obtained based on the calculated EDC. When the last menstrual period is unknown or the cycle is irregular, ultrasound measurements between the 14th and 20th week of gestation provide an accurate determination of gestational age (see Chapter 8). Physical Examination A targeted physical examination during the first prenatal visit includes special attention to the patient's BMI, blood pressure, thyroid, skin, breasts, and pelvis. On pelvic examination, the cervix is inspected for anomalies and for the presence of condylomata, neoplasia, or infection. A Pap smear is performed, and cultures for gonorrhea and chlamydia are taken, if indicated. A small amount of bright red bleeding may occur after these manipulations and the patient can be assured that this is normal. On bimanual examination, the cervix is palpated to assess consistency and length, and to detect the presence of cervical motion tenderness. Size, position, and contour of the uterus are noted. The adnexa are palpated to assess for masses. The pelvic examination may include evaluation of the bony pelvis, specifically, the diagonal conjugate, the ischial spines, the sacral hollow, and the arch of the symphysis pubis. This evaluation need only be performed once during the pregnancy. Laboratory Evaluation Several laboratory tests are routinely done at the first prenatal visit. Blood Tests Hematologic testing includes a white blood cell count, hemoglobin, hematocrit, platelet count for women of Asian descent, a serologic test for syphilis (rapid plasma reagin or VDRL), a rubella titer, a hepatitis B surface antigen, a blood group (ABO), and Rh type and antibody screen. Human immunodeficiency virus (HIV) testing should be offered to all pregnant patients and refusal documented in the chart. Routine assessment for toxoplasmosis, cytomegalovirus, and varicella immunity is not necessary, but may be obtained if indicated. The National Institutes of Health and ACOG recommend offering all Caucasian women testing for cystic fibrosis status. Women with histories suggestive of thrombophilia, or a personal or family history for blood clots, should be screened at this time. Women with a history suggestive of thyroid disease should also be evaluated. Appropriate screening for genetic carrier status, if not performed at the preconception visit, includes, but is not limited to, Tay–Sachs disease, Canavan disease in women of Jewish ancestry, a- and ß-thalassemia in women of Asian and Mediterranean descent, and sickle cell disease in women of African descent. Urine Tests All women should have a clean-catch urine sent for culture. Asymptotic bacteriuria occurs in 5% to 8% of pregnant women. Urinary stasis is present during pregnancy secondary to physiologic changes in the urinary system, including decreased ureteral peristalsis and mechanical uterine compression of the ureter at the pelvic brim as pregnancy progresses. Bacteriuria combined with urinary stasis predisposes the patient to pyelonephritis, the most common nonobstetric cause for hospitalization during pregnancy. Asymptomatic bacteriuria is identified using microscopic urine analysis, urine culture (>100,000 colonies per mL), or a leukocyte esterase–nitrite dipstick on a clean-catch voided urine. Cultures and Infections The use of routine genital tract cultures in pregnancy is controversial. While it is clear that chlamydiosis, gonorrhea, group B streptococcal disease, herpes infection, and potentially bacterial vaginosis can be detrimental to the ultimate health of the fetus or newborn, the indications for, and timing of, cultures for these infections are not clear. The ACOG recommends assessment for chlamydiosis and gonorrhea at the first prenatal visit for high-risk patients. The high-risk patient is defined as less than 25 years of age with a past history or current evidence of any sexually transmitted disease, a new sexual partner within the preceding 3 months, or multiple sexual partners. Any abnormal discharge should be assessed with a wet prep or Gram stain. Symptomatic patients should be treated. Tuberculosis skin testing in high-risk populations or in certain geographic areas should be done if the patient has not been vaccinated with BCG vaccine. BCG vaccinations are not given in the United States. Discussion with the Patient The first prenatal visit is a time for the caregiver and patient to exchange expectations, to answer questions, and to set the stage for what will occur throughout the rest of normal prenatal care. The timing and content of future visits, and the timing and rationale behind further laboratory testing should be explained. The patient should be given authoritative educational resources and materials that are written at the appropriate reading level. She and her partner are encouraged to ask questions about what they will read and to share the concerns they have about the pregnancy. It is important to reinforce that there is no such thing as a meaningless, “dumb”, or trivial question. Emergency and routine phone numbers should be given to the patient in writing. Social services and community resources, such as Women, Infants, and Children (WIC) programs, may be identified for the patient on an as-needed basis. Discussion regarding sexual activities, physical activities, and nutrition are usually initiated at this time. Instructions on safe and unsafe over-the-counter medications (i.e., acetaminophen vs. ibuprofen) are also initiated. Instruction on the use of seat belts and domestic abuse is also recommended here. Finally, the patient should be made aware of the warning signs and symptoms of infection (fevers, chills, dysuria/hematuria) or threatened pregnancy loss (bleeding,

cramping, passage of tissue). Should any of these occur, the patient should seek immediate medical attention. At the completion of the first visit, the next prenatal appointment is made.

ROUTINE ANTEPARTUM SURVEILLANCE Rationale for Routine Prenatal Care Prenatal care involves the following goals for pregnant women: to provide continuing, ongoing primary preventive health care to maintain or increase maternal health and the capability for self-care and to improve self-image before, during, and after pregnancy to reduce the risk of maternal mortality and morbidity, as well as unnecessary pregnancy intervention to reduce the risks to health before subsequent pregnancies and beyond the childbearing years to promote the development of parenting skills. The goals of prenatal care for the fetus are as follows: to reduce the risk of preterm birth, IUGR, retardation, and congenital anomalies to enhance fetal health and reduce the need for extended hospitalization after birth to promote healthy growth and development, immunization, and health supervision of the infant to reduce the risk of neurologic, developmental, and other morbidities to reduce the risk of child abuse and neglect, injuries, and preventable acute and chronic illness. The goals of prenatal care for the family during pregnancy and the first year of an infant's life are the following: to promote family development and positive parent–infant interaction to reduce the number of unintended pregnancies to identify and treat behavioral disorders that can lead to child neglect and family violence. Timing and Frequency of Visits The traditional timing and number of prenatal visits is summarized as follows:

Preconception:Up to 1 year before conception First prenatal: 6 to 8 weeks after missed menses Monthly: Up to 28 weeks Bimonthly: Up to 36 weeks Weekly: Until delivery

For the past decade prenatal care has been modified for the number and timing of visits in low-risk patients. Visits may be based on necessary assessments and interventions. A U.S. Public Health Service report delineated the interventions and tests deemed minimally necessary in a normal pregnancy and the suggested the timing for each ( Table 1.3). Randomized, controlled prospective trials have evaluated and confirmed the safety of fewer prenatal visits that are modeled around specific times of intervention, such as 16- and 28-week visits. Studies have looked at 6 to 7 visits instead of the usual 13 to 14 and have found similar outcomes. Interestingly, many women reported that they felt dissatisfied with fewer visits.

TABLE 1.3. Timing (in weeks) of prenatal care based on specific interventions

Content of Subsequent Prenatal Visits The components of each prenatal visit are the taking of an interval history, assessment of fetal growth and maternal health, risk assessment and identification, intervention as necessary, collection and recording of an ongoing database (see Appendix I), and education, advice and support of the patient and her family. Interval History Each prenatal visit begins with information gathering. Patients should be asked questions about their general health (see “ Concerns and Questions Particular to Pregnancy” later in the chapter), their diet, sleeping patterns, and fetal movement. Questions regarding warning signs such as bleeding, contractions, leaking of fluid, headache, or visual disturbances are also appropriate. Patients should be given the opportunity to raise their own questions and concerns at each visit, with open-ended inquiries. Physical Examination The patient is weighed, and total weight gain and trends are evaluated (see “ Nutrition” later in the chapter). The blood pressure is taken and trends are assessed for possible pregnancy-induced hypertension. As blood pressure tends to decrease during the second trimester, increases of 30 mm Hg systolic or 15 mm Hg diastolic over first trimester pressures are considered abnormal. The fundal height is measured with a tape from the top of the symphysis pubis, over the uterine curve, to the top of the fundus ( Fig. 1.1 and Fig. 1.2). This technique places an emphasis on change in growth patterns rather than the absolute measurement in centimeters, which can vary between patients. In women who are obese, periodic ultrasound assessments of fetal growth may be necessary. Gestational age is approximately equal to fundal height in centimeters from 16 to 36 weeks gestation. Measurements that are more than 2 cm smaller than expected for week of gestation are suspicious for oligohydramnios, IUGR, fetal anomaly, abnormal fetal lie, or premature fetal descent into the pelvis. Conversely, larger than expected measurements may indicate multiple gestation, polyhydramnios, fetal macrosomia, or leiomyomata. These concerns can be resolved with ultrasound examination.

FIG. 1.1. The height of the fundus at comparable gestational dates varies among patients. Those shown are the most common. A convenient rule of thumb is that at 20 weeks gestation, the fundus is at or slightly above the umbilicus.

FIG. 1.2. Fundal heights versus gestational age. Fetal heart rate is auscultated, with care taken to differentiate fetal from maternal rates. The normal fetal heart rate throughout pregnancy is between 120 and 160 beats per minute. Fetal position has been traditionally evaluated with the use of Leopold maneuvers. These are initiated at midpregnancy, when fetal body parts are more clearly identified. The maneuvers consist of four parts; the first three are performed with the examiner standing to one side of the patient and facing her head and the last with the examiner facing the patient's feet. The first maneuver answers the question, “What fetal part occupies the fundus?” ( Fig. 1.3). The examiner palpates the fundal area and differentiates between the irregular, firm breech and the round, hard head.

FIG. 1.3. The first Leopold maneuver reveals what fetal part occupies the fundus. The second maneuver answers the question, “On which side is the fetal back?” ( Fig. 1.4). The palms of the hands are placed on either side of the abdomen. On one side, the linear continuous ridge of the back is felt, while on the other side compressible areas and nodular parts are found.

FIG. 1.4. The second Leopold maneuver reveals the position of the fetal back. The third maneuver answers the question, “What fetal part lies over the pelvic inlet?” ( Fig. 1.5). A single examining hand is placed just above the symphysis. The fetal part that overrides the symphysis is grasped between the thumb and third finger. If the head is unengaged, it is readily recognized as a round, hard object that frequently can be displaced upward. After engagement, the back of the head or a shoulder is felt as a relatively fixed, knoblike part. In breech presentations, the irregular, nodular breech is felt in direct continuity with the fetal back.

FIG. 1.5. The third Leopold maneuver reveals what fetal part lies over the pelvic inlet. The fourth maneuver answers the question, “On which side is the cephalic prominence?” ( Fig. 1.6 and Fig. 1.7). This maneuver can be performed only when the head is engaged; if the head is floating, the maneuver is inapplicable. The examiner faces the patient's feet and places a hand on either side of the uterus, just above the pelvic inlet. When pressure is exerted in the direction of the inlet, one hand can descend farther than the other. The part of the fetus that prevents the deep descent of one hand is called the cephalic prominence.

FIG. 1.6. The fourth Leopold maneuver reveals the position of the cephalic prominence. In a flexion attitude, the cephalic prominence is on the same side as the small parts.

FIG. 1.7. In the fourth Leopold maneuver, in an extension attitude, the cephalic prominence is on the same side as the back.

The routine examination is completed by evaluating the patient for edema. A finding of new-onset edema of the face and hands in association with proteinuria and elevated blood pressure is consistent with preeclampsia. Dependent pitting edema of the ankles and legs in the absence of other findings is normal in late pregnancy. It responds well to resting, with the legs elevated, and therefore is usually absent on rising in the morning. Sudden weight gain in the third trimester to a large extent reflects an increase in edema. The amount of weight gain that is pathologic is not known. However, more than 5 lbs in a week is generally considered problematic. Routine examination of the cervix is not necessary unless the patient is at risk for cervical incompetence or is being evaluated for preterm labor. Laboratory Evaluation Several laboratory evaluations are offered to all patients. These include screening for neural tube defects and aneuploidy, with alpha-fetoprotein and other serum values (triple or quad screen), screening for gestational diabetes with a glucose challenge, as well as screening for maternal antibodies to fetal blood type. Maternal Serum Screening Test Screening with maternal serum alpha-fetoprotein (MSAFP) is performed to detect fetal neural tube defects and fetal ventral wall defects. During pregnancy, AFP is produced in sequence by the fetal yolk sac, the fetal gastrointestinal tract, and finally, the fetal liver. Its peak concentration in fetal serum occurs at the end of the first trimester. Excretion of AFP in fetal urine results in high levels of AFP in the amniotic fluid. Transfer of AFP to the maternal serum occurs via the placenta and transamniotically. MSAFP levels continue to rise until approximately 30 weeks gestation ( Fig. 1.8). The interpretation of the MSAFP screening test is dependent on gestational age and should be performed at the 15th to 20th menstrual weeks of pregnancy.

FIG. 1.8. Alpha-fetoprotein concentrations in fetal serum, amniotic fluid, and maternal serum throughout gestation. (From Seppala M, ed. Amniotic fluid, second ed. New York: Excerpta Medica, 1978; with permission.) MSAFP levels are reported as multiples of the median from the database of the individual laboratory. Elevated maternal serum and amniotic fluid levels of AFP detect 85% of open neural tube defects (open spina bifida and anencephaly). Other causes for elevated MSAFP levels include omphalocele, gastroschisis, multiple gestation, fetal demise, incorrect dates, and adverse pregnancy outcomes. Patients with abnormal MSAFP levels require evaluation with targeted fetal ultrasonography ( Fig. 1.9)

(see Chapter 6, Chapter 8).

FIG. 1.9. Screening for Down syndrome during pregnancy by measuring markers in maternal serum. (Adapted from Haddow JE, Palomaki GE, Knight, GJ, et al. Prenatal screening for fetal Down syndrome with the use of measurement of maternal serum markers. N Engl J Med 1992;327:588.) The risk of fetal trisomy 21 is related to gestational age/normalized concentrations of MSAFP, total ß-hCG, unconjugated estriol, and inhibin. Thus, these biochemical markers may be used in conjunction with the maternal age-related risk of fetal trisomy 21 to derive a more precise risk of fetal trisomy 21 in a given pregnancy. Called the triple screen or quad screen, this scheme assumes that computed risks for fetal trisomy 21 greater than 1 in 270 (, the mid-trimester risk of fetal trisomy 21 in a 35-year-old woman) should prompt consideration of genetic counseling, targeted ultrasound, and possibly amniocentesis. Using such a scheme, prospective studies show that approximately 60% to 70% of fetal trisomy 21 will be detected, with a false-positive rate of 5%. In women older than age 34 at delivery, the triple marker will detect 80% to 85% of trisomy 21. These women should be offered genetic counseling, as some will opt for amniocentesis or chorionic villus sampling. It is important to counsel a patient about the differences between a screening test and a diagnostic test. The possibility of false-positive and false-negative test results should also be explained. Approximately 60% of trisomy 18 may be detected with biochemical means (triple or quad screen). Many women will opt for first trimester screening, including nuchal thickening and biochemical markers. Early studies hold promise for this type of screening, though precise predictive values are still being estimated. Screening for Gestational Diabetes The 1-hour, 50-g oral glucose screen is used to detect glucose intolerance in pregnancy. Following an abnormal screen, a 3-hour glucose tolerance test, commencing with a fasting blood sugar, followed by a 100-g Glucola, is currently recommended. Two or more abnormal values on this test are considered diagnostic of gestational diabetes mellitus (GDM). Universal screening is controversial, though most clinicians have opted for a universal approach. Risk factors for GDM include: maternal age greater than 30 years previous macrosomic, malformed, or stillborn infant GDM in a previous pregnancy family history of diabetes maternal obesity persistent glucosuria chronic use of certain drugs such as ß-sympathomimetics or corticosteroids. Proponents of universal screening argue that screening only those patients with risk factors will detect no more than half of patients with glucose intolerance. Opponents argue that the inconvenience and expense of testing are not necessary in patients without these risk factors, because the incidence of frank GDM in this population is so low. Routine screening, if used, is performed on all patients between 26 and 28 weeks gestation. Selective screening based on risks may be performed earlier and repeated as needed, if negative at earlier gestations. A patient may be tested in the fasting or nonfasting state. One hour after administration of a 50-g glucose load, the patient's blood is drawn. A patient with a glucose value greater than 140 mg per dL of serum is a candidate for a 3-hour, 100-g glucose tolerance test. The significance of GDM lies not in an increased risk of fetal loss but in the risk of excessive fetal growth with its attendant birth-related morbidities. In addition, women with GDM have a 60% likelihood of developing overt diabetes mellitus within 16 years. Rescreening for Rh Antibodies All Rh-negative women who are unsensitized at the beginning of pregnancy should be retested at approximately 26 to 28 weeks gestation. If the antibody screen remains negative, the mother should receive Rh 0 (D) immune globulin 300 mcg at 28 weeks, to prevent isoimmunization in the third trimester. Approximately 1% of Rh-negative women will become sensitized if not given Rh immune globulins. Screening for Bacterial Vaginosis Bacterial vaginosis (BV) is a condition in which the normal flora of the vagina (specifically lactobacilli) are reduced in number and replaced by an overgrowth of anaerobic organisms. Studies have linked BV with an increased incidence of preterm labor, endometritis, and premature rupture of the membranes. A simple and effective screen, performed late in the second trimester, consists of a pelvic examination and wet mount to detect BV. A Gram stain is an alternative diagnostic tool. The treatment for women who are positive for BV includes either metronidazole or clindamycin (Cleocin). Because BV is often asymptomatic, a test of cure may be appropriate. Routine screening is not recommended, as studies have not shown that screening and treatment decreases preterm labor and delivery. However, symptomatic women, women with cerclage, or women with preterm dilated cervices should be screened and treated. Testing for Group B Streptococci Group B streptococci (GBS) are part of the normal vaginal, genitourinary, and gastrointestinal tract flora in up to 30% of healthy women. GBS have been implicated in amnionitis, endometritis, and wound infection in the mother. Vertical transmission during pregnancy, labor, and delivery may result in generalized sepsis in the newborn and related long-term morbidity or neonatal death. Prevention strategies have focused on detection of the bacteria in the mother and early onset of GBS disease in the newborn. The recommended strategy involves routine anogenital cultures of all pregnant women at 35 to 37 weeks gestation. Cultures are obtained from the lower third of the vagina and perianal area. Cervical cultures are not reliable and a speculum is not necessary to obtain an adequate culture sample. Culture-positive women are treated during labor with antibiotic prophylaxis to prevent fetal–neonatal GBS infection. Women with a positive urine culture for GBS should be given antibiotic prophylaxis in labor. These women do not need to be recultured. Testing Based on Symptoms or Clinical Risk Assessment A part of prenatal care of the normal patient consists of ongoing risk assessment and intervention or referral if a risk is identified. Several clinical signs or symptoms warrant further evaluation. Symptoms suggestive of urinary tract infections should prompt examination of a clean-catch urine specimen and cultures when appropriate. High-risk behaviors, identified during the course of a pregnancy, should prompt a test (or retest) for HIV infection and sexually transmitted diseases (STDs) or performance of a urinary drug screen. Repeated testing of hemoglobin should be done if the patient is symptomatic or at nutritional risk for anemia. Other testing, performed on an as-needed basis, includes ultrasound to detect abnormal fetal growth, antepartum fetal monitoring to assess fetal oxygenation status, or comprehensive targeted ultrasound examinations. A more thorough discussion of antepartum fetal monitoring can be found in Chapter 9. Discussion with Patients and Families: Answering Questions Patients need the opportunity to engage in dialogue with their health care provider and to feel confident that their concerns are heard. Patients and families will often interact with a nurse or triage person in a physician's office. These individuals need to be trained in careful assessment and evaluation. The value of information ancillary personnel can provide cannot be overemphasized. The prenatal visits are a time to stress the involvement of the entire family in the pregnancy process, including the role of the father and siblings. Therefore, an important part of the prenatal visit is discussion with the patient, her partner, or her family, both to exchange questions and answers, and to provide reassurance and education. The exact content of these discussions will vary from visit to visit. Reaffirming the importance of appropriate social behaviors, such as smoking cessation, is beneficial, as are periodic evaluations of the social support systems and help in the home, both now and after the birth of the infant. Ongoing risk assessment requires that the patient be educated about the signs and symptoms of preterm labor and preeclampsia. The list of warning signs for which an emergent telephone call is warranted includes the following: vaginal bleeding leaking of fluid from the vagina rhythmic cramping pains of more than six per hour abdominal pain of a prolonged or increasing nature fever or chills burning with urination prolonged vomiting with inability to hold down liquids or solids for more than 24 hours severe continuous headache, visual changes, or generalized edema a pronounced decrease in the frequency or intensity of fetal movements.

CONCERNS AND QUESTIONS PARTICULAR TO PREGNANCY Psychological Pregnancy is a time of change, expectation, and anticipation. It may also be a time of heightened anxiety, emotionality, concern, and uncertainty. Many symptoms that the nonpregnant patient might view as minor may indicate a cause for alarm during pregnancy. The provision of direct, concise, and accurate information in a

compassionate and reassuring manner will assuage many of these worries and provide direction for day-to-day activities. Nutrition The objectives of nutritional assessment and counseling are to develop, in concert with the patient, an analysis of maternal nutritional risk, a goal for total weight gain, and a diet plan that will fit the patient's lifestyle and is ethnically sensitive. The principle of good nutrition is that there is a positive linear relationship between maternal weight gain and newborn weight and that prepregnant maternal BMI can affect fetal weight independently of the amount gained by the mother during pregnancy. Together, initial weight and weight gain have an impact on IUGR and low birth weight. However, for a woman of normal weight and normal nutrition, the relationship between poor weight gain and fetal growth restriction may be an association not a cause and effect. The BMI is a calculation that relates the patient's weight to her height, thereby providing a more accurate indirect estimate of the patient's body fat distribution than can be obtained by weight alone. The BMI is calculated by dividing weight in kilograms by height in meters squared. If pounds and inches are used, the quotient is multiplied by 700. The BMI of a patient is categorized as underweight, normal weight, overweight, or obese (see Appendix II). Maternal Weight Gain The ideal weight gain for an individual patient during pregnancy depends on several factors. The most important of these are the prepregnant BMI and the type of gestation (single vs. multiple). It is important to note that weight gain for a normal BMI is 25 to 35 lbs (11 to 16 kg), but the optimal weight gain for an underweight teenager carrying a singleton pregnancy approaches 40 lbs (18.2 kg), or 5 lbs every 4 weeks in the second half of pregnancy ( Fig. 1.10). An obese woman, on the other hand, may need to gain no more than 15 lbs (6.8 kg). Figure 1.11 illustrates the components of weight gain in a normal pregnancy. During the first and second trimesters, most of the weight gained reflects maternal changes, primarily an increase in total body water, while fetal growth is most rapid in the last trimester, with the fetus more than tripling its weight. The optimal weight gain for women with twins is usually greater than 40 lbs. Poor weight gain is often a reflection of patients not expanding their intravascular volume. This is associated with low birth weight and greater complications in pregnancy. Having patients “eat more” does not usually help. In contrast, weight gain greater than 30 lbs will often remain as extra weight after delivery. Women should be instructed not to diet during pregnancy. Many women will not gain significant weight until the middle of the second trimester. Patients may be reassured that this is a normal pattern.

FIG. 1.10. Target weight gains for normal-weight women with body mass index (BMI) of 19.8 to 26, underweight women with a BMI of less than 19.8, and overweight women with a BMI of more than 26 to 29. A: A 1.6 kg (3.5 lb) gain in the first trimester and the remaining gain at a rate of 0.44 kg (0.97 lb) per week are assumed. B: A 2.3 kg (5 lb) gain in the first trimester and the remaining gain at a rate of 0.49 kg (1.1 lb) per week are assumed. C: A 0.9 kg (2 lb) gain in the first trimester and the remaining gain at a rate of 0.3 kg (0.67 lb) per week are assumed.

FIG. 1.11. Pattern and components of weight gain during pregnancy (LMP, last menstrual period). (From Pitkin RM. Nutritional support in obstetrics and gynecology. Clin Obstet Gynecol 1976;19:489, with permission.) Maternal Diet While weight gain is an important gauge of caloric intake, the quality of the diet and the frequency of meals may also affect patient and fetal well-being. A diet should be balanced by containing foods from all of the basic food groups. Specifics of a diet will vary considerably according to patient preference, family eating patterns, and cultural and ethnic background. Increased nutritional requirements during pregnancy reflect the needs of the fetus for growth, as well as maternal physiologic needs. To meet the overall increasing energy needs, the average woman must consume an additional 300 kcal per day beyond her baseline needs. The appropriate daily caloric content of a diet required to supply energy needs and achieve appropriate weight gain can be estimated by multiplying the patient's optimal body weight in kilograms by 35 kcal and adding 300 kcal to the total ( Table 1.4).

TABLE 1.4. Recommended dietary allowances for women of reproductive age and for pregnant and lactating women Vitamin and Mineral Supplementation Multivitamin supplements are not routinely necessary in a woman eating a well-balanced diet. However, 800 µg of supplemental folic acid daily is necessary because the requirement cannot be met with food alone. Additional folate may be necessary for women with a hemoglobinopathy or MTHFR mutation, for women on antiseizure medications, or for women with a history of neural tube defects. Vitamin D supplementation is not generally required, unless there is inadequate exposure to sunlight. Mineral supplementation is also not needed in healthy women. The exception is iron. The iron requirements of pregnancy total about 1 g. Due to the monthly menses, most women have less than optimal iron stores during their reproductive years. Therefore, supplementation with 30 mg of elemental iron is recommended in the second and third trimesters to prevent anemia and to meet this requirement. One tablet of iron salts per day, ingested between meals or at bedtime, is sufficient to meet this requirement. Women with iron deficiency anemia require 60 to 120 mg of elemental ferrous iron per day. Additional zinc (15 mg) and copper (2 mg) are then needed, as iron inhibits the absorption of these ions. Calcium supplementation is not necessary in women with a diet that includes adequate dairy foods. Absent this, calcium supplementation may be used on an as-needed basis to meet the recommended dietary allowance (RDA) of 1200 to 1500 mg per day during pregnancy and 2000 mg per day with lactation. Zinc is a trace mineral. A zinc deficiency may be teratogenic in humans, although this has not yet been conclusively demonstrated. Zinc levels in amniotic fluid correlate with antimicrobial activity, suggesting that zinc plays a role in protecting against intrauterine infection. Low dietary intake of zinc has been associated with IUGR, although it does not cause IUGR. The RDA for zinc during pregnancy is increased from 15 to 20 mg per day. Other Dietary Considerations Vegetarianism Lacto-ovo vegetarians should have no particular nutritional deficiency, with the possible exceptions of iron and zinc, which may be supplemented. The strict vegan, however, must design a diet of sufficient vegetable proteins to provide all of the essential amino acids normally found in animal protein. Due to the decreased protein density of most vegetables, this may cause an unusually high weight gain. Supplementation of zinc, vitamin B 12, and iron is necessary. Food Restriction Dieting and fasting on a chronic basis in an otherwise healthy woman can result in suboptimal fetal growth. Eating disorders such as bulimia and anorexia nervosa reflect extreme forms of food restriction and malnutrition. There are limited data about these disorders in pregnancy, but anorexics in particular place their fetus at risk. Bulimic women may suffer from electrolyte imbalance and a deficit of trace minerals. Many pregnant women in the United States are not eating an optimal diet due to poverty and inadequate resources to purchase food. It is appropriate to inquire about resources in impoverished women and to refer these patients to groups such as WIC and to appropriate agencies providing food stamps. Pica Pica is the compulsive ingestion of nonfood substances with little or no nutrient value. The practice most commonly involves ice, clay (geophagia), or starch (amylophagia). Although pica is most commonly recognized during pregnancy, it is not specific to the gravid state. Neither the cause nor the medical implications of pica are well understood. It is unusual for pica to cause significant harm if the diet is otherwise nutritionally adequate.

Phenylketonuria Women with phenylketonuria who are not on a phenylalanine-controlled diet are at increased risk of bearing fetuses with microcephaly, growth retardation, and mental retardation. The goal of dietary management is to minimize these adverse fetal outcomes by reducing the maternal serum phenylalanine levels to less than 20 mg per dL before and during the pregnancy. At the first prenatal visit, every pregnant woman should be asked if she was on a special diet as a child. Megadose Vitamins The misuse of megadose nutrients can be categorized as a fad type of dietary manipulation. Water-soluble vitamins such as vitamin C cannot be consumed in harmful quantities because they are readily excreted in the urine. However, a problem occurs with fat-soluble vitamin A. There is an association between high doses of supplemental vitamin A and birth defects similar to those seen with isotretinoin. Although the minimum teratogenic dose in humans has not been identified, it may be a little as 10,000 IU per day. Beta-carotene is a provitamin of vitamin A, but it does not produce similar toxicity. Most prenatal vitamins contain less than 5000 IU of vitamin A and, until further data are available, this should be considered the maximum safe supplemental dose. Caffeine Caffeine is contained in numerous foodstuffs such as coffee, tea, chocolate, and cola beverages. A naturally occurring substance, it is the most widely used psychoactive drug in the United States. It is a central nervous system stimulant and is physically and psychologically addictive. Withdrawal symptoms include nausea, lethargy, malaise, and headache. The only evidence for teratogenic effects of caffeine comes from animal studies using doses not compatible with human consumption. Several large human studies have failed to show that caffeine has deleterious effects on the fetus, when ingested in low amounts. However, it is associated with an increased risk of miscarriage when taken in greater than the equivalent of three cups of coffee. Caffeine intake of the equivalent of two to three cups is thus discouraged. Adverse maternal effects of caffeine include insomnia, acid indigestion, reflux, and urinary frequency. As these problems are already exaggerated in pregnancy, moderation in the consumption of caffeine is advisable. Exercise Exercise is a routine part of many women's daily activities. For a normal pregnancy, a low-impact exercise regimen may be continued throughout pregnancy. Additionally, studies also show that women may increase their levels of fitness during pregnancy without problems. There are no data to indicate that pregnant women must decrease the intensity of their exercise or lower their target heart rates. However, physiologic changes of pregnancy may alter the effect of various exercises on the body or may limit the body's ability to perform certain types of exercise. Body position as a modulator of cardiac output is particularly important in the third trimester, when either motionless standing or the supine position can result in decreased venous return and cardiac output. In some instances, this will result in hypotension or syncope. Both oxygen uptake and baseline oxygen consumption are increased during pregnancy. Deep breathing is more difficult, particularly in later pregnancy, due to uterine size and decreased diaphragmatic excursion. These changes combine to make less oxygen available for aerobic activity, thereby decreasing maximum exercise performance. Exercise is not a means of weight control in pregnancy. Women who seek to exercise to keep from gaining weight or losing their prepregnancy shape should be counseled regarding normal pregnancy body changes. Data regarding the fetal response to maternal exercise are reassuring. Moderate (submaximal) exercise has never been shown to increase maternal core body temperature and thus, fetal temperature. Studies of the fetus following submaximal maternal exercise (65% to 70% aerobic capacity) have not shown any associated changes in the fetal heart rate. Exercise-induced effects on the fetus, including malformations, increased miscarriage rates, retardation, or growth restriction, have not been demonstrated in human pregnancies. Women who exercise strenuously throughout pregnancy, such as elite athletes, may deliver infants as much as 300 to 400 g smaller in weight than women who do not exercise as strenuously. This, however, is not considered deleterious. Recommendations Women who exercise regularly before pregnancy may be encouraged to continue. They may be counseled that performance capacity tends to fall, but this is not a sign that they should forgo regular moderate exercise. Indeed, exercise may relieve stress, diminish anxiety, and increase self-esteem. Some studies have also shown that women who exercise regularly have shorter labors. For most women with gestational diabetes, regular exercise has been shown to be helpful for glucose control. Specific exercise regimens should be individualized and patients who have not been physically active prior to pregnancy are advised to proceed slowly. General recommendations include the following: Exercise should be regular rather than sporadic and intermittent. Exercise should be stopped if signs and symptoms of oxygen deprivation, such as extreme fatigue, dizziness, or extreme shortness of breath, occur. To avoid becoming overheated, pregnant women should exercise in a cool area, stay well hydrated, and wear appropriate clothing. Exercise that requires prolonged time in the supine position should be avoided during the second and third trimesters. The form of exercise chosen should not be one with significant risk of trauma (especially to the abdomen) or falls. Caloric intake should be increased in direct proportion to the additional energy requirements of exercise. Contraindications Relative contraindications to exercise during pregnancy include the following: evidence of IUGR persistent vaginal bleeding incompetent cervix or cervical cerclage placement risk factors for preterm labor rupture of membranes pregnancy-induced hypertension chronic medical conditions that might be adversely impacted by vigorous exercise. Nausea and Vomiting Recurrent nausea and vomiting during the first trimester occurs in over one half of pregnancies. While the term morning sickness is well known, it is a misnomer, as these symptoms can occur at any time throughout the day or night. Symptoms usually begin in weeks 6 to 8, peak during weeks 12 to 14, and are significantly resolved by week 22. The etiology of this problem is not clear. Hormonal, as well as emotional, factors have been investigated without consistent results. Symptoms can be mild or so severe that the patient becomes dehydrated and risks electrolyte imbalance and caloric malnutrition. Nonpharmacologic measures often suffice to alleviate, if not completely relieve, the symptoms. These include avoidance of fatty or spicy foods, eating small, more frequent meals, drinking ginger teas, inhaling peppermint oil vapors, wearing motion sickness bands on the wrists, and increasing rest periods each day. In severe cases of emesis, various pharmacologic agents have been used with varying success. These include pyridoxine, a variety of antihistamines, promethazine, metoclopramide, trimethobenzamide, methylprednisolone, and droperidol. Because supplemental vitamin and mineral preparations may exacerbate symptoms of nausea, they should be stopped until the symptoms have resolved. Women and their families may be reassured that minimal weight gain in the first 18 weeks is common. Ptyalism Ptyalism is the increased production of saliva, probably induced by the consumption of starch. There is no cure, although reducing carbohydrate intake may be helpful. The problem is often self-limiting. Heartburn Heartburn is usually caused by reflux esophagitis from both mechanical factors (enlarging uterus displacing the stomach above the esophageal sphincter) and hormonal factors (progesterone causing a relative relaxation of the esophageal sphincter). Treatment consists of eliminating acidic and spicy foods, decreasing the amount of food and liquid at each meal, limiting food and liquid intake before bedtime, sleeping in a semi-Fowler position or propped up on pillows, and use of antacids. Liquid forms of antacids and H 2 -receptor inhibitors provide the most consistent relief of symptoms. Patients should be cautioned that antacids containing aluminum may cause constipation, while diarrhea may be associated with use of those containing magnesium. Constipation Progesterone-induced relaxation of the intestinal smooth muscle slows peristalsis and increases bowel transit time. Dietary management of this common condition includes increased fluids and liberal intake of high-fiber foods. Iron salts may exacerbate the problem. Over-the-counter (OTC) products containing psyllium draw fluid into the intestine and promote a more rapid transit time. Enemas and strong cathartics should be avoided. Fatigue Pregnant women will usually have an increased sense of fatigue during pregnancy. This is a normal symptom. A sense of breathlessness is also normal. A significant

increase in fatigue or breathlessness however should alert the clinician to possible pathology. Varicosities and Hemorrhoids Varicosities most often occur in the lower extremities but may be seen in the vulva as well. Contributing factors include genetic predisposition, advanced maternal age, increased parity, and prolonged standing. Manifestations can range from mild cosmetic effects to chronic pain and superficial thrombophlebitis. Treatment includes avoidance of garments that constrict at the knee and upper leg, support stockings, and increased periods of rest with the legs elevated. Hemorrhoids, varicosities of the rectal veins, are due to mechanical compression by the enlarging uterus, as well as from constipation and straining at stool. Treatment includes OTC preparations, witch hazel, topical preparations, cool sitz baths, and stool softeners. If thrombosis of a hemorrhoid occurs, the clot can be excised to relieve pain and swelling. Leg Cramps Almost half of all pregnant women suffer from recurrent painful spasms of the muscles of the lower extremities, especially the calves. Leg cramps are more frequent at night and usually occur during the third trimester. Various prophylactic and therapeutic options have been suggested, most notably, calcium lactate and high potassium foods, such as banana, kiwi, or cantaloupe, but there are no data from controlled trials to show benefit over placebo for any of these. Massage, heat, and stretching the affected muscle(s) relieves the cramps when they occur. Backache Most pregnant women experience lower backaches as pregnancy progresses. These are usually alleviated by minimizing the amount of time spent standing, by increasing rest, by wearing a specially designed support belt over the lower abdomen, and by taking an analgesic such as acetaminophen. Exercises to increase muscular strength of the back and abdomen are sometimes helpful. Shoes with good support and avoidance of high heels that exaggerate the lordotic posture are essential. Increasingly severe or abrupt-onset back pain requires orthopedic consultation. Rhythmic cramping pains originating in the back may be a sign of preterm labor and necessitate appropriate evaluation. Round Ligament Pain This pain most frequently occurs during the second trimester when women report sharp, bilateral, or unilateral groin pain. It has been called “round ligament pain”, though it is not known if round ligament stretch is the true etiology. The pain may be increased with sudden movement or change in position. Resolution of unremitting ligament pain is sometimes achieved by having the patient assume a position on the hands and knees and lower the head to the floor, while keeping the buttocks in the air. Headache Generalized headaches are not uncommon during the first trimester of pregnancy. Muscle tension headaches may occur intermittently. The frequency and intensity of migraine headaches may increase or decrease during pregnancy. Headaches during the second and third trimesters are not an expected symptom of pregnancy. Pathologic headaches that occur with preeclampsia are discussed in Chapter 16. Emotional Changes Pregnancy is a time of significant psychological stress. Changes in hormonal levels, changes in relationships to partners, family and friends, and changes in body image all lead to increased psychological stress. Increased levels of placental corticotropin-releasing hormone toward the end of pregnancy also affect the maternal hypothalamic–pituitary axis and other brain loci involved in stress responses. There is a corresponding shift in most women toward primary process. Dreams become more vivid and dramatic. Emotional lability is common. It is very helpful to counsel the pregnant woman and her partner about these normal changes. Women with social stressors or with a history of depression may develop signs of atypical depression, necessitating counseling and medications. Sexual Relations Coital activity during normal pregnancy need not be restricted. The couple can be counseled regarding changing positions to achieve better comfort. Deep penetration may be more uncomfortable as pregnancy progresses. It is common for women to have changes in sexual desire over the course of gestation. Many women achieve orgasm easier during pregnancy; however libido often decreases in the first and third trimesters. Nipple stimulation, vaginal penetration, and orgasm can cause uterine contractions secondary to the release of prostaglandins and oxytocin. However, there are no proven adverse effects on the fetus or the onset of labor. The question of the effect of coitus in women at risk for preterm labor or early spontaneous pregnancy loss remains unanswered. Couples at risk may prefer to avoid sexual relations to minimize any feelings of guilt or responsibility if a problem occurs subsequently. There are two concrete interdictions to coitus during pregnancy. The first is that intercourse should not occur after membrane rupture or in the presence of known placenta previa. The second is that forceful introduction of air into the vagina should be avoided because of the risk of fatal air embolism. Employment Most patients are able to continue to work throughout their pregnancy. In general, work activities that increase the risk of falls or trauma, especially to the abdomen, should be avoided. Hazardous toxic or chemical exposures should be identified early and avoided. Strenuous physical activity, including repetitive lifting and prolonged standing for more than 5 hours, has been associated with a greater rate of adverse outcomes, and work routines should be modified accordingly. Urinary Frequency Patients often experience urinary frequency during the first 3 months of pregnancy, as the enlarging uterus compresses the bladder, and again during the last weeks, as the fetal head descends into the pelvis. If frequency occurs in conjunction with dysuria, hematuria, or urgency/hesitancy, the patient should be evaluated for a urinary tract infection. Skin Changes Hair growth has variable patterns in pregnancy, although many women experience increased growth during pregnancy and hair loss postpartum. Skin commonly darkens over the face and median ventral line of the abdomen in many women. Any nevi that change color should be excised. Leukorrhea An increase in the amount of vaginal discharge is physiologic and expected during pregnancy. Discharge accompanied by itching or burning, or a malodorous discharge, should be evaluated and treated accordingly. Douching has no place in the treatment or management of leukorrhea in pregnancy. Increased watery discharge may precede preterm labor and should be evaluated. Syncope Venous pooling in the lower extremities increases as the pregnancy progresses. This can lead to dizziness or lightheadedness, especially after standing upright abruptly or for long periods of time. Other causes of syncope include dehydration, hypoglycemia, and the shunting of blood flow to the stomach after eating a large meal. Syncope during exercise is a sign of overexertion. In general, syncopal episodes resolve rapidly and should be managed acutely, just as in a nonpregnant patient. Syncope in the supine position is avoidable by resting in the lateral recumbent position, right or left, thereby relieving uterine compression of the vena cava. X-rays/Ionizing Radiation The adverse effects on the fetus of ionizing radiation are dose-dependent. While there is no single diagnostic procedure that results in a dose of radiation high enough

to threaten the fetus or embryo, cumulative exposures or multiple procedures should be avoided, especially during the first trimester when the fetus is at highest risk for possible anomalies. Patients may undergo dental x-rays as needed, provided that the abdomen is fully covered by a lead apron. Studies using radioactive isotopes are best avoided. As in all diagnostic procedures, the risks and the potential benefits must be evaluated and individualized for each patient. Exposure to video display terminals is safe in pregnancy. Travel Most issues concerning travel involve the comfort of the mother. When prolonged sitting is involved, the patient should try to stretch her legs and walk for 10 minutes every 2 hours to decrease the risk of thrombosis that can occur secondary to the hypercoaguable pregnancy state and mechanical compression of venous blood flow from the extremities. Dependent edema may also be more pronounced after prolonged sitting. If the patient will be away from home for a significant period of time, she should take a copy of her medical record with her. Pregnant women can and should always wear seat belts when riding in a car. Travel in a pressurized airplane presents no additional risk to pregnant women. In traveling abroad, especially to underdeveloped countries, the usual precautions should be taken regarding ingestion of unpurified drinking water and uncooked fruits and vegetables. Immunizations Four immunizations using vaccines containing live viruses are relatively contraindicated during pregnancy. These are measles, mumps, rubella, and yellow fever. However, in certain circumstances, risk/benefit assessment may lead to receiving the immunizations. The risks for the fetus from the administration of rabies vaccine are unknown and each case must be considered individually, since the indications for prophylaxis are not altered by pregnancy. Tetanus toxoid, if needed, is acceptable in pregnancy. Flu vaccine is recommended for pregnant women. Women who are receiving hepatitis B vaccine may continue receiving it during pregnancy. Immune globulin for acute exposures to hepatitis A is considered safe also.

PREPARING FOR CHILDBIRTH Prenatal Education Classes Few empiric studies have examined the impact of prepared childbirth education on perinatal outcomes, but such education is generally believed to be helpful and valuable. The landmark volume Birth of a Child by Grantley Dick-Read, published in 1958, changed the modern face of childbirth education. Formal childbirth classes evolved rapidly in a multiplicity of settings. The goals of these classes are to educate and to answer questions in an environment conducive to the woman's new state of being, so that both the patient and her partner have the opportunity to decrease their anxiety level and increase their knowledge. Classes are designed to be an empowering experience, helping the parent(s) become a part of the process rather than the object of the actions of others. The content of childbirth classes varies but usually includes topics such as normal labor and delivery, anesthesia, breathing and concentration techniques, obstetric complications and interventions, and obstetric operations. Many instructors encourage patients to formulate a “birth plan” and to put this in writing and share it with their clinician. This can facilitate communication between the parents themselves and between patient and caregiver. Certain decisions are best considered and made prior to delivery. These include issues such as breast-feeding, postpartum contraception, return to work, and circumcision of a male infant. It is helpful if parents are given information in a nonjudgmental, nonthreatening environment so that they may make appropriate, well-considered, and informed decisions. Signs of Labor The final element in preparing for childbirth is knowledge of when labor is occurring and when it is appropriate to notify the health care provider. Patients should be given a 24-hour phone number to call for assistance. A course of action should be made clear to the patient and her partner. As has been reinforced throughout pregnancy, any warning signs of potential adverse outcomes mandate an immediate telephone call.

CONCLUSION The future of effective and efficient prenatal care in the United States depends largely on access, our ability to demonstrate clear benefit for patients and the incorporation of evidence-based data and practices that have well-defined outcomes and use cost-effective methods. Of equal importance to the content of prenatal care is the manner in which it is delivered. A crucial determinant of effective prenatal care is the clinician–patient relationship. Virtues of trust, honesty, and ethical treatment are integral to achieving the goal of prenatal care, which emphasizes allowing patients to be active in their health care. The skill of a caregiver can prevent the alienating experience of a bad outcome, but caregivers who look for a solution to their own powerlessness in the domination of patients or the domination of natural processes can make good outcomes alienating experiences for their patients. Caregivers must be aware of cultural diversity and assess the care they give, not in terms of their own needs or the needs of a profession or a medical organization, but in terms of those who have entrusted them with care.

SUMMARY POINTS Pregnancy is a normal physiologic event in a woman's life and most pregnancies are normal. Pregnancy is a time of significant psychological transitions that are experienced in varied ways by each mother. The preconception visit is a focused visit that allows systematic identification of potential risks and the implementation of early interventions. The appropriate content of the first prenatal visit and subsequent prenatal care is contained in formalized published forms. A firm scientific foundation for the content and timing of prenatal care visits and the relationship of such care to maternal and newborn outcomes is under continued review. Pregnancy is a time of change, expectation, anticipation, concern, and uncertainty for many women and their families; the provision of direct, concise, and accurate information in a compassionate and reassuring manner by the clinician is therapeutic. Each prenatal visit should evaluate and address interval history, maternal changes, fetal growth, specific interventions, and general patient concerns. SUGGESTED READINGS 1.

Baskett T, Nägele F. Naegele's rule: a reappraisal. Br J Obstet Gynaecol 2000;107:1433.

2.

Bastian LA, Piscitelli JT. Is this patient pregnant? Can you really rule out early pregnancy by clinical examination? JAMA 1997;278:586.

3.

Brown SS, ed. Prenatal care: reaching mothers, reaching babies. Washington, DC: National Academy Press, 1988.

4.

Carroli G, Villar J, Piaggio G, et al. WHO systematic review of randomised controlled trials of routine antenatal care. Lancet 2001;357:1565.

5.

Clapp JF III, Little KD. The interaction between regular exercise and selected aspects of women's health. Am J Obstet Gynecol 1995;173:2.

6. Clement S, Candy B, Sikorski J, et al. Does reducing the frequency of routine antenatal visits have long term effects? Follow up of participants in a randomised controlled trial. Br J Obstet Gynaecol 1999;106:367. 7.

Fiscella K. Does prenatal care improve birth outcomes? A critical review. Obstet Gynecol 1995;85:468.

8.

Gilbert W, Nesbitt TS, Danielsen B. Childbearing beyond age 40: pregnancy outcome in 23,032 cases. Obstet Gynecol 1999;93:9.

9.

Lacroix R, Eason E, Melzack R. Nausea and vomiting during pregnancy: a prospective study of its frequency, intensity, and patterns of change. Am J Obstet Gynecol 2000;182:931.

10.

Kogan MD, Martin JA, Alexander GR, et al. The changing pattern of prenatal care utilization in the United States, 1981–1995, using different prenatal care indices. JAMA 1998;279:1623.

11.

Lederman R. Psychosocial adaptation in pregnancy. New York: Springer Publishing Co, 1996.

12.

McDuffie RS Jr, Beck A, Bischoff K, et al. Effect of frequency of prenatal visits on perinatal outcome among low-risk women: a randomized controlled trial. JAMA 1996;275:847.

13.

McFarlane J, Parker B, Soeken K, et al. Assessing for abuse during pregnancy. JAMA 1992;267:3176.

12.

National Academy of Sciences. Recommended dietary allowances. Washington, DC: National Academy Press, 1989.

13.

Pederson AL, Worthington-Robets B, Hicko DE. Weight gain patterns during twin gestation. J Am Diet Assoc 1989;89:642.

14.

Rothman KJ, Moore LL, Singer MR, et al. Teratogenicity of high vitamin A intake. N Engl J Med 1995;333:1369.

15. Rouse DJ, Andrews WW, Goldenberg RL, Owen J. Screening and treatment of asymptomatic bacteriuria of pregnancy to prevent pyelonephritis: a cost-effectiveness and cost-benefit analysis. Obstet Gynecol 1995;86:119. 16.

Rubin R. Maternal identity and the maternal experience. New York: Springer Publishing Co, 1984.

17.

Sokol RJ, Martier SS, Ager JW. The T-ACE questions: practical prenatal detection of risk-drinking. Am J Obstet Gynecol 1989;160:863.

18.

Signorello LB, Nordmark A, Granath F, et al. Caffeine metabolism and the risk of spontaneous abortion of normal karyotype fetuses. Obstet Gynecol 2001;98:1059.

19.

Stewart DE, Cecutti A. Physical abuse in pregnancy. Can Med Assoc J 1993;149:1257.

20.

Villar J, Ba'aqeel H, Piaggio G, et al. WHO antenatal care randomised trial for the evaluation of a new model of routine antenatal care. Lancet 2001;357:1551.

21.

American College of Obstetricians and Gynecologists. Domestic violence. ACOG Educational Bulletin No. 257, Dec 1999.

22.

American College of Obstetricians and Gynecologists. Exercise during pregnancy and the postpartum period. ACOG Committee Opinion No. 267, Jan 2002.

Appendix I

Figure. ACOG ANTEPARTUM RECORD (FORM A).

Figure. ACOG ANTEPARTUM RECORD (FORM B).

Figure. ACOG ANTEPARTUM RECORD (FORM C).

Figure. ACOG ANTEPARTUM RECORD (FORM D).

Figure. ACOG ANTEPARTUM RECORD (FORM E).

TABLE cellSpacing=0 cellPadding=0 align=left border=0 hspace="10" vspace="5"> Figure. ACOG ANTEPARTUM RECORD (FORM F).

Figure. OBSTETRIC MEDICAL HISTORY.

Appendix II The following table shows the body mass index (BMI) for weight and height. From the bottom line: Every number below the first dark line is “underweight.” Every number between the first and second dark line is “nomal weight.” Every number between the second and third dark line is “overweitht.” The rest of the chart above the third dark line represents “obese.”

Figure. BMI for weight and height.

Appendix III Maternal and child health care currently accounts for a large proportion of health care dollars spent annually in the United States. This is due to the large number of patients and the numerous procedures performed during pregnancy and the first months of life. The effectiveness of prenatal care is still often measured in terms of finite outcomes such as maternal and infant mortality. While these end points do not take into account such parameters as patient satisfaction, emergency interventions, length of hospitalization, and degree of access, they are used as a crude measure of quality of care of mothers and infants and are often cited in comparisons between states and countries, even when the actual data and methodologies for obtaining them are diverse and incomparable. The definitions currently used by the National Center for Health Statistics when collecting data on maternal and infant mortality and morbidity are given here. Live birth: the complete expulsion or extraction from the mother of a product of human conception, regardless of the duration of the pregnancy, that then breathes or shows any other evidence of life, whether or not the umbilical cord has been cut or the placenta is attached. Fetal death (stillbirth): death before the complete expulsion or extraction from the mother of a product of human conception, regardless of the duration of the pregnancy. This definition excludes terminations of pregnancy. Early neonatal death: death of a live-born infant during the first 7 days of life. Late neonatal death: death of an infant after 7 days but before 29 days of life. Fetal death rate: the number of stillborn infants per 1000 infants born. Neonatal mortality rate:the number of infant deaths before 29 days of life per 1000 live births. Perinatal death rate: a combination of fetal and neonatal deaths per 1000 total births. Birth rate: the number of births per 1000 members of the population. Fertility rate: the number of live births per 1000 female members of the population between the ages of 15 and 44 years. Maternal mortality: the death of a woman from any cause related to or aggravated by pregnancy or its management, regardless of duration or site of pregnancy (but not from accidental or incidental causes), occurring during the pregnancy or up to 42 days after the pregnancy. It includes direct and indirect obstetric deaths. Direct obstetric death: the death of a woman from obstetric complications of pregnancy, labor, or the puerperium, resulting from interventions, omissions, or treatment, or from a chain of events resulting from any of these. Indirect obstetric death: the death of a woman resulting from a previously existing disease or a disease that developed during pregnancy, labor, or the puerperium that was not related to direct obstetric causes, although the physiologic effects of pregnancy were partially responsible for the death. Maternal mortality rate: maternal deaths per 100,000 live births.

Chapter 2 Normal Labor, Delivery, Newborn Care, and Puerperium Danforth’s Obstetrics and Gynecology

Chapter 2 Dwight J. Rouse and Elaine St. John

Normal Labor, Delivery, Newborn Care, and Puerperium

LABOR Labor Progress MANAGEMENT OF LABOR AND DELIVERY Antepartum Instructions Admission Management of the First Stage of Labor Management of the Second Stage of Labor Management of the Third Stage of Labor Management of the “Fourth Stage” of Labor Postpartum Care PUERPERIUM Complications of the Puerperium Breast-feeding Nursing Problems NEWBORN CARE Immediate Assessment and Resuscitation Meconium The Apgar Score General Evaluation and Treatment Early Onset Sepsis Hyperbilirubinemia and Discharge Planning SUMMARY POINTS SUGGESTED READINGS

LABOR Labor is defined as regular uterine contractions that lead to effacement and dilation of the cervix. If the estimated gestational age is accurate, labor usually begins within 2 weeks of the estimated date of confinement (EDC), which is 280 days (i.e., 40 weeks after the first day of the last menstrual period). However, because only 3% to 5% of patients actually deliver on their EDC, it has been suggested that EDCs be framed in terms of a range (e.g., from 38 to 42 weeks). Conceptualizing EDCs in this manner, it has been argued, might lessen the pressure for date-predicated interventions such as induction of labor. Prior to the onset of true labor, there is a general softening and stretching of pelvic ligaments and the soft tissues of the vagina. There is also shortening and dilation (ripening) of the cervix. Braxton Hicks contractions (weak, irregular, regional contractions) usually occur for weeks before the onset of actual labor. The normal stimulus for the biochemical cascade that finally results in labor is unknown. Corticotropin-releasing factor (CRF) plays a role. It is released into the maternal circulation early in the second trimester, and its concentration rises exponentially as pregnancy advances. CRF regulates the secretion of adrenal cortisol, which can increase the strength of uterine contractions. CRF also stimulates production of oxytocin by the fetus and prostaglandins by the placenta. Other factors probably play a role. For instance, progesterone inhibits and estrogen stimulates uterine contractility. Like CRF, estrogen also stimulates the production of oxytocin receptors in the uterus. Thus, as progesterone levels decline near term, estrogen may stimulate myometrial contractility. Mechanical stretch can also increase uterine contractility, as occurs with twin gestations and pregnancies complicated by polyhydramnios. Once labor is initiated, the process is thought to involve multiple positive feedback loops. For instance, contractions stretch the cervix. Stretching of the cervix elicits a reflex contraction by the uterus, pushing the fetal head against the cervix to stretch it more, and so on. Labor Progress Clinically recognizable labor is typically divided into three stages, each with statistically derived normative rates and durations. Many normative labor values were derived from the investigations of Emanuel Friedman, who studied thousands of normal and abnormal labors and plotted cervical dilation and fetal descent against time. The resulting graphic labor curve can be used to recognize individual labor patterns which deviate from normal and to guide the nature and timing of interventions. In a World Health Organization study of 35,484 women, use of a partogram ( Fig. 2.1) and an agreed upon labor management protocol was associated with a reduction in the percentage of prolonged labors, the proportion of labors requiring augmentation, and postpartum sepsis.

FIG. 2.1. Flow sheet for following labor progress. (From Chua S, Arulkumaran S. Poor prognosis in labor, including augmentation, malpositions and malpresentations. In: James DK, Steer PJ, Weiner CP, Gonik B, eds. High risk pregnancy, second ed. London: Harcourt Brace, 1999:1105; with permission.)

The first stage of labor begins with the onset of regular uterine contractions and ends with complete cervical dilation. This stage is further divided into three phases: latent, active, and deceleration, although the existence of the last phase has been questioned. During the latent phase, contractions become progressively stronger, longer, more frequent, and better coordinated. The mother's discomfort may be minimal or it may be severe. The latent phase is considered prolonged if it lasts longer than 20 hours in a nullipara, or 14 hours in a parous woman. The active phase commences when the slope of cervical dilation reaches its maximum. Typically, and especially in the nullipara, this occurs at 3 to 4 cm of dilation. During the active phase, contractions are usually strong and regular, occurring every 2 to 3 minutes. The active phase ends with complete (10 cm) cervical dilation. This phase of labor is generally quite painful. The length of the active phase is more predictable than the latent, lasting, on average, about 5 hours in nulliparas, and 2 hours in multiparas. Without labor epidural analgesia, respective minimum (fifth percentile) rates of dilation are 1.2 cm per hour and 1.5 cm per hour. With epidural, the rates are slower. Progressive descent of the fetal head into the maternal pelvis occurs to a variable degree during the active phase. The second stage is defined as the period from complete cervical dilation to complete delivery of the baby. During the second stage, contractions are strong and regular, with a frequency of every 1 to 3 minutes. The baby's head descends more deeply into the pelvis, and in women without regional anesthesia, each contraction stimulates a strong urge to push. In combination, uterine contractions and maternal expulsive efforts effect delivery of the baby. The second stage typically lasts about 50 minutes in nulliparas and 20 minutes in multiparas, but is often longer in women with regional anesthesia. The third stage of labor is defined as the period from delivery of the baby to delivery of the placenta. Regardless of parity, the third stage of labor is usually brief (under 10 minutes). The third stage of labor is prolonged if it persists beyond 30 minutes. The three classic determinants of the progress of labor are: Power—uterine contractions and, in the second stage, maternal expulsive efforts Pelvis—the bony pelvis and the overlying maternal soft tissues Passenger—the fetus and its lie, presentation, and position. The power of labor can be assessed clinically by uterine palpation or with the use of an intrauterine pressure catheter. Assessment of the pelvis involves manual

evaluation of the pelvic inlet, midpelvis, and outlet. Pelvic inlet—The transverse diameter of the pelvic inlet averages 13 cm. It cannot be measured clinically, but a narrow transverse inlet is a very rare cause of abnormal labor progress. The anteroposterior (AP) diameter of the inlet is more important. It is estimated clinically by determining the distance between the lower margin of the symphysis pubis and the sacral promontory ( Fig. 2.2). This value is known as the diagonal conjugate. The obstetric conjugate—or true AP diameter—is 1.5 to 2.0 cm shorter. The pelvic inlet is an adequate size for a normal fetus if the diagonal conjugate is 12 cm or greater.

FIG. 2.2. The pelvic inlet anteroposterior diameter is estimated from the diagonal conjugate. Midpelvis—The specific diameters of the midpelvis cannot be measured clinically. Contraction of the midpelvis is suspected if the ischial spines are quite prominent (or the sacrosciatic notch is less than two fingerbreadths wide), the pubic arch is narrow, the pelvic side walls converge, or the sacral concavity is quite shallow ( Fig. 2.3).

FIG. 2.3. The transverse diameter of the midpelvis is estimated by evaluating the distance between the ischial spines. Pelvic outlet—The transverse diameter of the pelvic outlet should be greater than 8 cm. This diameter can be estimated by placing a fist on the perineum to measure the distance between the ischial tuberosities. The AP diameter is estimated by noting the angle made by the pubic rami. A contracted outlet is rarely the sole cause of dystocia; however, it is often associated with midpelvis contraction. Consideration of these measurements allows assignment to one of the various pelvic types, and thus an appreciation of how and where labor may be stalled if the pelvis is not favorable for childbirth ( Fig. 2.4). Careful evaluation of the midpelvis is most important, as those women found to have a contracted midpelvis are poor candidates for forceps-assisted vaginal delivery. However, because the fetal skull has the ability to mold, borderline pelvimetry is not a contraindication to a trial of labor.

FIG. 2.4. Pelvic types. A: Gynecoid pelvis—most common, round to oval inlet, ischial spines not prominent, curved sacrum, wide pubic arch; best suited for childbearing. B: Android pelvis—heart-shaped inlet, narrow midpelvis with anterior sacrum, prominent ischial spines, convergent side walls, narrow pubic arch. C: Platypelloid pelvis—least common, decreased anteroposterior dimensions at all levels with wide transverse dimensions. D: Anthropoid pelvis—narrow inlet, midpelvis, and pubic arch.

From the perspective of the passenger (fetus), labor involves movement progressively downward through the pelvis by the following cardinal movements ( Fig. 2.5).

FIG. 2.5. Cardinal movements of labor. A: Engagement. B: Flexion. C: Descent and internal rotation. D, E: Extension. F: External rotation.

Engagement occurs days to weeks prior to labor for primigravidas and at the onset of labor for multigravidas. Flexion of the neck allows the occiput to lead, thus the smallest possible diameter of the fetal head travels downward through the pelvis. Descent is progressive as the cervix thins, and the lower uterine segment lengthens. Internal rotation occurs during descent. The vertex rotates from transverse to either a posterior or anterior position to pass the ischial spines. Extension occurs as the fetal head distends the perineum and the occiput passes beneath the symphysis. External rotation of the head after delivery to a transverse position allows the shoulders to rotate internally to an AP position.

MANAGEMENT OF LABOR AND DELIVERY Antepartum Instructions All women should be advised of the circumstances that should prompt them to seek evaluation for labor. These include (a) possible rupture of the membranes, (b) regular uterine contractions, (c) bleeding per vagina, and (d) back, pelvic, or abdominal pain greater than they are anticipating. Nulliparas are more likely to confuse false labor or Braxton Hicks contractions with true labor. Although distinguishing the two can be problematic, the contractions of false labor tend to be irregular both in intensity and in interval, and the associated discomfort, if any, is typically limited to the lower abdomen and groin. The contractions of false labor usually abate with time, analgesia, or sedation. With true labor, the contractions progressively increase in intensity. They occur every 2 to 4 minutes and cause discomfort in the abdomen and back with the associated sensation of increasing pelvic pressure. For many women labor will be preceded, by several hours or even days, by the passage of bloody show (a small amount of bloody mucous discharge from the cervix). In 10% of pregnancies, chorioamnion rupture precedes the onset of labor, and amniotic fluid leaks through the cervix and out of the vagina. Optimal management for such women has recently been clarified. A randomized trial of over 5000 women with

prelabor rupture of membranes at term demonstrated that induction of labor with oxytocin (as opposed to expectant management, or induction with prostaglandins) resulted in the lowest rate of neonatal infection and maternal postpartum fever, and did not increase the cesarean delivery rate. In many cases, the only way to confirm the diagnosis of true labor is observation over several hours and examination of the cervix for change. Admission If the woman is having contractions, their time of onset and frequency should be recorded. Questions should focus on spontaneous rupture of the membranes, presence or absence of bleeding, and fetal activity. A review of the patient's prenatal record should take specific note of her EDC and its reliability, as well as her past medical and surgical history, and details of previous pregnancies: number, gestation, fetal size, duration of labor, and any complications (e.g., shoulder dystocia). Prenatal laboratory data should be reviewed, including blood type, hematocrit, Rh 0 (D) immune globulin requirements, VDRL test, rubella immunity, and hepatitis and human immunodeficiency virus (HIV) status. The admission physical examination should include vital signs (temperature, pulse, blood pressure), auscultation of the heart and lungs, and a brief neurologic examination. Leopold maneuvers should be performed to assess fetal position ( Fig. 2.6), and the uterus should be palpated, or a tocodynamometer employed, to determine the frequency, intensity, and duration of uterine contractions. Fundal height should be evaluated and a clinical assessment of fetal weight should be performed. Fetal heart tones should be assessed, either by auscultation or via electronic monitoring, with specific attention to the response of the fetal heart rate (FHR) to the uterine contractions.

FIG. 2.6. Leopold maneuvers. First maneuver: The uterine contour is outlined; the fundus is palpated, allowing identification of the fetal parts. Second maneuver: By palpation of the sides of the maternal abdomen, the location of the fetal back is determined. Third maneuver: The presenting part is grasped, identified, and evaluated for engagement. Fourth maneuver: With palpation toward the pelvis, the identity of the presenting part is confirmed, and flexion or extension of the fetal head is evaluated.

The vulva should be examined for herpetic lesions. If membrane rupture is suspected, it should be confirmed (or ruled out). Several signs support rupture, including pooling of amniotic fluid in the vagina (observed by sterile speculum examination) or direct visualization of fluid leakage through the cervix. The pH of the pooled fluid can be checked with nitrazine paper, which turns blue in the presence of amniotic fluid (but also in the presence of blood), and an air-dried sample (on a slide) of the fluid can be examined under a microscope for the characteristic “fern” pattern that confirms the presence of amniotic fluid ( Fig. 2.7). Since cervical mucus and maternal serum can demonstrate a fern pattern, care must be taken in collection of this sample. Palpation of the cervix includes attention to the following:

FIG. 2.7. Typical ferning pattern of dried amniotic fluid (400×). (Original photo, courtesy of Dr. Dwight Rouse.)

assessment of consistency (soft or firm) degree of effacement ( Fig. 2.8)

FIG. 2.8. Degree of cervical effacement. A: No effacement. B: 75% effacement. C: 100% effacement. dilation of the cervical os location of the cervical os with respect to the vaginal axis (posterior, midplane, or anterior). The presenting part should be identified by palpation. Station is determined by noting the position of the fetal presenting (bony) part relative to the ischial spines ( Fig. 2.9). The fetal position is determined by noting the orientation of occiput relative to the maternal pelvis ( Fig. 2.10). The attitude of the fetal head is the position of the fetal head relative to the fetal chest and any lateral flexion of the head ( Fig. 2.11, Fig. 2.12 and Fig. 2.13). Pelvimetry should be performed as previously described.

FIG. 2.9. Stations of the fetal head. At the 0 station, the fetal head is at the bony ischial spines and fills the maternal sacrum. Positions above the ischial spines are referred to as -1 through -5, referring to the number of centimeters that the head is positioned above the spines. As the head descends past the ischial spines, the stations are referred to as +1 through +5 (head visible at the introitus).

FIG. 2.10. Fetal position. The orientation of the presenting vertex within the maternal pelvis.

FIG. 2.11. A, B: The bones, sutures, fontanelles, and clinically important diameters of the fetal head.

FIG. 2.12. Fetal attitude and dimensions of a term-size fetus. A: Full flexion presents the smallest circumference of the fetal head to the narrower planes of the pelvis. B: Military attitude usually changes to full flexion with descent into the pelvis. C: Brow presentation usually converts to full flexion or a face presentation, as the occipitomental diameter is too large for all except the largest pelves to accommodate. D: Face presentation shows dimensions that allow descent through the pelvis, unless the chin is posterior. Persistent mentum posterior must be delivered by cesarean section.

FIG. 2.13. Fetal attitude and lateral flexion of the fetal head. A: Synclitism—The plane of the biparietal diameter is parallel to the plane of the inlet. B: Asynclitism—Lateral flexion of the fetal head leads to anterior parietal or posterior parietal presentation.

In women at term with rupture of the membranes prior to the onset of labor, it is appropriate to defer digital examination of the cervix until they are in active labor, to decrease the risk of chorioamnionitis. It is important, however, to confirm the baby's presentation, and an ultrasound examination should be performed if the presentation is not clear. Management of the First Stage of Labor A variety of management approaches to the first stage of labor have been employed and evaluated, and no single approach is clearly superior. The primary management goals are to monitor fetal well-being, support the woman through what can be a lengthy, uncomfortable period, and offer intervention as it may become appropriate. Support of the woman through labor includes allowing her to assume whatever position is most comfortable (or least uncomfortable) for her. All forms of monitoring, be it intermittent auscultation, external fetal monitoring, or internal monitoring, can be accomplished in a lying, sitting, or upright position. In most normal labors, the only time a healthy woman's movements must be limited is after she has received analgesia and would not be steady on her feet. Uterine blood flow is maximal in the lateral recumbent position, and this should be the initial position adopted if there is concern over fetal well-being. Because of impaired venous return and reduced maternal cardiac output, the supine position is usually best avoided. With these considerations in mind, women should be free to position themselves as they like. Vital signs should be monitored at least every 4 hours, or more frequently as clinically indicated. By consensus of the Centers for Disease Control and Prevention and the American College of Obstetricians and Gynecologists (ACOG), all women whose 35-37 week rectovaginal group B streptococcus (GBS) culture was positive should receive intrapartum antibiotic prophylaxis for the prevention of early onset neonatal GBS disease. Acceptable prophylactic antibiotic regimens include intravenous penicillin G 5 million units initially, and then 2.5 million units every four hours until delivery, or intravenous ampicillin 2 gm initially, followed by 1 gm every 4 hours until delivery. For women allergic to penicillin, clindamycin, erythromycin, vancomycin, or, in women without a urticarial or anaphylactoid response to penicillin, a first

generation cephalosporin may be substituted. Irrespective of rectovaginal culture results, women with GBS bacteriuria during the pregnancy and those who have previously delivered an infant with early onset GBS disease should also receive antibiotic prophylaxis. If an appropriate rectovaginal GBS culture was not performed or is not available, prophylaxis should be based on the following risk factors: threatened pre-term delivery (< 37 weeks), GBS bacteriuria at any time in the current pregnancy, fever (= 38.0°C), rupture of membranes for at least 18 hours, or delivery of a prior infant with early onset GBS disease. Separate from the issue of GBS prophylaxis, if a woman develops an intrapartum fever, defined by some authorities as a temperature of at least 37.8°C (100.0°F) and by others as at least 38.0°C (100.4°F), evaluation and treatment for chorioamnionitis may be indicated. Placement of an intravenous line is not necessary for all women in labor. However, women who are dehydrated may benefit from intravenous hydration. It seems prudent to establish intravenous access for administration of fluids and medication, should they be necessary, in women at increased risk of postpartum hemorrhage (i.e., those with prior postpartum hemorrhage, prolonged labor, or overdistended uterus). In most women, laboratory evaluation upon presentation in labor can be minimized. Although in many units it is customary to perform routine admission blood type and antibody screen, hemoglobin and hematocrit, and syphilis serology, the necessity and cost-effectiveness of repeating these tests in healthy women who have received adequate prenatal care is debatable. Certainly, if a woman has hypertension on admission, she should be evaluated for possible preeclampsia. In women who have not had any prenatal care, it is advisable to order a hemoglobin and hematocrit, blood type, and Rh status, an antibody screen, a rubella titer, and syphilis, hepatitis, and HIV screens. The benefits of continuous caregiver support (by nurses, midwives, or laypeople) throughout labor have been established by multiple randomized trials. Salutary effects of such support include reduced pain medication requirements, lowered rates of operative vaginal and cesarean delivery, and a diminution in the frequency of 5-minute Apgar scores below 7. Thus, to the extent possible, all women should have access to continuous caregiver support throughout labor. No evidence-based criteria are available to direct how often the cervix should be examined during the first stage of labor. In general, frequent examinations in the latent phase of labor serve little purpose. They may unrealistically increase a woman's expectation for progress in labor and in addition, are uncomfortable and increase risk for infectious morbidity. However, during observation, if the membranes rupture, the cervix should be examined and the FHR evaluated if the presenting part was not well applied to the cervix on previous examination. These measures should enhance the detection of umbilical cord prolapse. In the active phase of labor, monitoring the progress of labor with cervical examinations every 2 hours allows identification of those women who are not making normal progress and who should therefore be evaluated for oxytocin augmentation. During labor, the FHR should be monitored. For most women, both intermittent auscultation and intermittent or continuous electronic monitoring are acceptable (see Chapter 9). While artificial rupture of membranes (amniotomy) may shorten the duration of labor by 1 to 2 hours and reduce the use of oxytocin, it may also increase the risk of cesarean delivery by 25%. Thus, in most women, amniotomy should be reserved for such indications as placement of internal fetal monitoring devices for better assessment of the FHR pattern, or for abnormal labor progress. To avoid umbilical cord prolapse, amniotomy is best performed when the presenting part is well applied to the cervix, and preferably during a contraction (or with fundal pressure) to minimize the chance for dislodging the fetal head. Although typically performed with a thin, plastic, hook-ended tool designed specifically for the purpose, amniotomy may be achieved simply by placing a fetal scalp electrode. In cases of excess amniotic fluid, or when the vertex is not well applied, a spinal or pudendal block needle may be used to create a small opening in the membranes to allow the gradual egress of amniotic fluid, thereby avoiding umbilical cord prolapse or the rapid decompression of the uterus and the associated risk of placental abruption. Meconium staining of the amniotic fluid occurs in up to one-fifth of deliveries. Meta-analysis of randomized trials performed in settings of standard perinatal surveillance demonstrates that with moderate or severe meconium staining, amnioinfusion of 500 to 1000 mL (or more, depending on the protocol) of sterile saline into the uterus (through an intrauterine catheter) will reduce variable heart rate decelerations and may lower the risk of cesarean delivery. In settings of limited perinatal surveillance, amnioinfusion reduces the risk of meconium aspiration syndrome, neonatal hypoxic ischemic encephalopathy, neonatal ventilation or intensive care unit admission, and perhaps perinatal mortality. Management of the Second Stage of Labor The onset of the second stage of labor with complete dilation of the cervix is usually noted by the woman as the presenting part descends into the vagina and she experiences the urge to push with contractions. However, regional anesthesia may partially or completely blunt this urge. The cervix should be examined at this time to confirm complete dilation, and the position and station of the presenting part should be ascertained. Whether nulliparous women with regional anesthesia should begin expulsive efforts in the absence of an urge to push was the subject of a multicenter, randomized clinical trial. In the trial, women allocated to immediate pushing were less likely to have intrapartum fever (4.5% vs. 8.5%). Moreover, their second stages were one hour shorter. However, difficult deliveries (defined as midpelvic operative vaginal deliveries, low-pelvic rotational deliveries, or second stage cesarean delivery) were less common (18% vs. 23%) in the delayed pushing group, who were instructed to not push until the urge was irresistible or the head was on the perineum. Neonatal outcomes were equivalent between the groups. This trial thus supports individualization in the timing of maternal expulsive efforts. Support of a woman in the second stage involves allowing her to find the most comfortable and effective position for pushing, as well as encouraging her efforts. Preparations for delivery should be made when the presenting part begins to distend the perineum and often sooner for multigravidas. Local or pudendal anesthesia, if necessary, should be administered at this time. An episiotomy is an incision in the perineum used to facilitate vaginal delivery. It is useful for patients in whom the perineum does not readily stretch and when delivery must be expedited. In the past, routine episiotomy was advocated to prevent subsequent pelvic relaxation; however, there is no evidence that it does so. Meta-analysis of randomized episiotomy trials, in which restrictive episiotomy use (˜ 30% of women underwent episiotomy) was compared to liberal use (˜70% episiotomy rate), reveals the following. Restrictive use of episiotomy lowers the risk of posterior perineal trauma, lessens the need for suturing, and is associated with fewer healing complications. However, restrictive use of episiotomy increases the risk of anterior perineal trauma. There is no difference between the two policies in severe vaginal or perineal trauma, or subsequent dyspareunia, urinary incontinence, or perineal pain. There are two basic types of episiotomy: midline and mediolateral. The benefits of a midline episiotomy include anatomic end results, easy repair, and lower incidence of postpartum pain or dyspareunia. A mediolateral episiotomy is less likely to extend into the anal sphincter or rectal mucosa, but is generally believed to result in more pain, dyspareunia, and excessive blood loss. Episiotomy should not be performed until delivery is imminent. A midline incision is made with scissors from the midpoint of the posterior fourchette directly backward toward the rectum ( Fig. 2.14). The midline episiotomy is subdivided as follows:

FIG. 2.14. Midline episiotomy. A: As the fetal head distends, with the perineum under adequate anesthesia, the episiotomy is cut through the perineal body and the tissues of the vagina and the rectovaginal septum. B: The episiotomy is repaired by reapproximating the vaginal mucosa in a running fashion with a delayed absorbable suture. C: The submucosal tissue of the vagina and the subcutaneous tissue and fascia of the perineal body are then closed. D: The skin is then reapproximated with a running subcuticular suture.

first degree—through the mucosa only

second degree—through the mucosa and subcutaneous tissues, including the muscles of the perineal body third degree—into or through the anal sphincter fourth degree—through the rectal mucosa. A mediolateral incision is made with scissors from the midpoint of the posterior fourchette at a 45-degree angle laterally on either side. Regardless of use of an episiotomy, tears and extensions into the rectum are best prevented by keeping the baby's head well flexed until the occiput passes beyond the subpubic arch. As the vertex appears beneath the symphysis, the perineum is supported by direct pressure from a draped hand over the coccygeal region. As the head delivers, it will rotate to a transverse position; the mother should be encouraged to continue to push to achieve delivery of the anterior shoulder. A quick evaluation for the presence of the umbilical cord around the baby's neck (nuchal cord) should be carried out. A nuchal cord can usually be slipped around the baby's head. Occasionally, it must be double-clamped and cut if it is too tight to reduce. Maternal pushing efforts can be assisted by placing gentle downward traction on the baby's head. With delivery of the anterior shoulder, a moment can be taken to bulb suction the baby's nose and mouth. Renewed pushing by the mother in combination with gentle upward traction by the obstetrician will achieve delivery of the posterior shoulder. The baby's body generally delivers easily following the shoulders. Care must be taken to support the baby's head and ensure that the baby does not slip from one's grip. After confirmation of good respiration and normal heart rate (easily checked at the cord insertion at the umbilicus), the baby may be given to the mother to cradle. The cord is then double-clamped and cut, and cord blood collected. The baby should be dried and wrapped soon following delivery to maintain its body temperature. Management of the Third Stage of Labor The third stage of labor begins following the delivery of the baby and ends with the delivery of the placenta. Signs of spontaneous placental separation include an apparent lengthening of the umbilical cord, a gush of vaginal bleeding, and a change in shape of the uterus from discoid to globular along with a rise in fundal height. Active management of the third stage of labor, which involves prompt umbilical cord clamping and cutting, administration of an oxytocic agent, and gentle umbilical cord traction, reduces maternal blood loss and the frequency of postpartum hemorrhage, and lessens the risk that the third stage will be prolonged. Cord traction should be used only against fundally applied counter-traction ( Fig. 2.15) to lessen the potential for uterine inversion and catastrophic hemorrhage. If at any time heavy bleeding occurs during the third stage of labor or if the placenta is not delivered within 30 minutes of the birth, the placenta should be manually removed. The anesthesiologist should be alerted at this time, as general anesthesia may be required for women who have no regional anesthesia, and curettage may be necessary if the placenta does not readily separate from the uterine wall. Manual removal is accomplished by developing a cleavage plane with the intrauterine hand between the maternal surface of the placenta and the uterine wall, while simultaneously fixing the uterus with the abdominal hand, and progressively peeling the placenta free. To ensure complete placental removal, a 4-inch by 4-inch gauze may be wrapped around the hand and used to abrade the uterine wall. The placenta should then be carefully inspected for cord insertion, confirmation of a three-vessel cord (infants with only two cord vessels have a higher rate of malformations and warrant closer evaluation by the newborn caregiver), and completeness of the placenta and membranes. If any portion of the placenta or the membranes is missing, the uterine cavity should be manually explored. Some advocate routine exploration of the uterine cavity to reduce the risk of infection and bleeding from retained placental fragments. In most women, especially those without regional anesthesia, the benefit of manual exploration is outweighed by the discomfort it causes, as well as the increased risk for uterine infection.

FIG. 2.15. Stage three of labor: delivery of the placenta. A: Spontaneous separation of the placenta is confirmed. B: With gentle traction on the umbilical cord and suprapubic palpation of the fundus, the placenta and membranes are delivered. C: If spontaneous separation of the placenta does not occur or bleeding ensues, the placenta is manually separated from the wall of the uterus and removed.

The uterus should be frequently palpated following delivery of the placenta to ensure that it remains well contracted. Oxytocin, 10 to 20 U administered intramuscularly or as a dilute intravenous solution, has been demonstrated to decrease the incidence of postpartum hemorrhage secondary to uterine atony. The birth canal, including the cervix, vagina, and perineum should be inspected for lacerations requiring repair. During this time, under most circumstances and after drying and swaddling, the baby should be kept with the family and, for women planning to breast-feed, placed to the breast to nurse within the first 10 to 20 minutes after birth. This first suckling not only stimulates endogenous oxytocin release but also begins the process of milk production and successful breast-feeding. Management of the “Fourth Stage” of Labor Because many complications of birth occur or become evident during the first hour after delivery, this time has been referred to as the “fourth stage” of labor, though it is not officially defined as a labor stage. The new mother should be seen at least every 15 minutes by a trained labor and delivery nurse to assess vital signs and look for evidence of uterine atony or postpartum hemorrhage. The perineum should be inspected for any signs of hematoma formation, which may be signaled by inordinate vulvar or rectal pain. The newborn should undergo its initial assessment at this time and be observed closely for any signs of compromise. Postpartum Care After the first 24 hours, postpartum recovery is rapid. A regular diet can be offered to most women as soon as they request food, sometimes shortly after delivery. Because of their risk to require general anesthesia, it is prudent to delay feeding women who have experienced, or are at risk for, significant postpartum hemorrhage. Full ambulation is encouraged as soon as possible. Exercises to improve tone and strengthen abdominal muscles may be started after 1 day. The perineum should be cleansed with warm water two or three times daily after voiding or bowel movements. Both showers and baths may be taken, but vaginal douching is prohibited during the early puerperium to avoid the risk of introducing infection into the uterus. Pain from an uncomfortable perineal laceration or episiotomy can be relieved with warm sitz baths and an analgesic (e.g., acetaminophen, 650 mg, with or without codeine or ibuprofen, 600 to 800 mg) ( Table 2.1).

TABLE 2.1. Examples of orders after routine vaginal delivery: immediately postpartum

When regional anesthesia is used, care must be taken to avoid problems with urinary retention and bladder overdistention. Women should be encouraged to void every 2 to 3 hours, even if they are not aware of any sensation of bladder fullness, as rapid diuresis may occur following delivery, especially when oxytocin is discontinued. Prior to discharge, a hemoglobin or hematocrit should be checked if there is any concern that a woman has experienced heavier than normal postpartum bleeding, or if she entered labor significantly anemic. In addition, seronegative women should be immunized against rubella on the day of discharge. If the mother is Rh-negative, is not sensitized, and has an Rh-positive infant, she should be given Rh 0 (D) immune globulin to prevent sensitization from occurring. As discharge from the hospital now

occurs anywhere from 12 to 48 hours following an uncomplicated birth, most women have not begun milk production prior to their discharge. To support breast-feeding, it is critical to advise women of the availability of assistance, should they experience any difficulty with nursing. Binding and avoiding any stimulation of the breasts are advised if the mother does not plan to breast-feed. She should also take aspirin, acetaminophen, or ibuprofen if engorgement becomes uncomfortable. The length of inpatient hospitalization following childbirth has changed remarkably over the past two decades, reflecting the decreasing risk of postpartum complications and changing medical and societal attitudes toward birth, as well as pressures from third-party payors. Numerous reports, including three randomized, controlled trials have shown early discharge, 12 to 24 hours postdelivery, with appropriate follow-up in the home results in a low and acceptable rate of readmission of mothers and babies. Several studies show that longer hospital stays improve maternal postpartum adjustment and breast-feeding success. Stays of 12- to 24-hours are only acceptable when patients have adequate help in the home and the community has a home care nursing program with 24-hour coverage, 7 days a week. Home care services, while not always necessary, can be very helpful and appear to increase the safety of early discharge. All women and their babies discharged within 24 hours of birth should be offered a minimum of one home visit or an appointment for outpatient follow-up in the first few days following discharge.

PUERPERIUM The puerperium is the 6 to 8 weeks following delivery of the placenta, in which the uterus returns to its normal state. Following delivery of the placenta, the uterus rapidly contracts to half of its predelivery size. The involution that then occurs over the next several weeks is most rapid in women who breast-feed their newborns. Postpartum vaginal discharge, or lochia, changes as the uterus involutes. Initially, the discharge is grossly bloody (lochia rubra), persisting for 3 to 4 days. It then decreases in volume and changes to pale brown and becomes thinner (lochia serosa), persisting for 10 to 12 days. Finally, the discharge becomes yellowish white, occasionally tinged with blood (lochia alba), and may persist for several weeks. The total volume of lochia is about 250 mL, and women are usually encouraged to use external pads to absorb it, rather than intravaginal tampons, in order to minimize the risk of infection. When followed closely, a woman's hematocrit may actually rise due to diuresis following delivery and autotransfusion occurring as the uterus involutes. After 1 week, the uterus is firm and nontender and extends to about midway between the symphysis and the umbilicus. By 2 weeks postpartum, the uterus should no longer be palpable abdominally. The contractions of the involuting uterus may be painful, especially during the first few days following delivery. These “after pains” are usually relieved with acetaminophen or ibuprofen. Prior to discharge, women should receive instructions regarding what they can expect during the puerperium and recommendations for activity. In general, women should be encouraged to rest as they feel necessary and gradually increase their activity following delivery. They do not need to be restricted in terms of ambulation, but should be cautioned that driving in the early postpartum may not be advisable if perineal discomfort is distracting, or would cause a delay in reaction in case of an emergency. Women can resume a regular diet, and, if nursing, should be encouraged to keep well hydrated. Perineal pain should gradually diminish. If it does not, or it worsens, women should be instructed to seek evaluation for possible infection or hematoma. Normal bladder function should resume, but women should be cautioned that they may experience some difficulty with stress and urge incontinence, which can be expected to gradually improve in the months following delivery. Parents should receive information regarding lochia, expected volume, changes, and duration; activities; care of the breasts, perineum, and bladder; and dietary and, specifically, fluid requirements. The specific signs of complications should be stressed, including fever, chills, leg pain or swelling, episiotomy pain or drainage, and abnormal duration or amount of bleeding. Couples should be advised that they can safely resume coitus when desired and comfortable; however, abstaining for at least 2 weeks following the birth is often recommended. They should be forewarned of potential discomfort of the perineum and problems with vaginal dryness due to atrophic vaginitis, as well as the potential for transient changes in libido related to pregnancy and delivery. For women who are breast-feeding, ACOG has summarized hormonal contraceptive options ( Table 2.2). Women who are not nursing and desire a combination oral contraceptive can safely have this method initiated 1 to 2 weeks following delivery. Long-acting progestin contraception such as medroxyprogesterone acetate (Depo-Provera) injections or levonorgestrel implants (Norplant System) can be started or placed anytime following delivery. A diaphragm should be fitted only after complete involution of the uterus at 6 to 8 weeks postpartum. In the meantime, condoms and spermicidal foam or jelly should be used. Intrauterine devices can be placed immediately following delivery of the placenta; however, insertion at this time is associated with an increased risk of expulsion. Thus, waiting until 6 to 8 weeks postpartum may be preferable.

TABLE 2.2. ACOG recommendations for hormonal contraception if used by breast-feeding women

In nonlactating women, ovulation usually occurs about 6 weeks postpartum. The first menses following delivery is often an anovulatory cycle; however, conception has been reported as early as 2 weeks postpartum, so contraception must be advised even in the early postpartum period. While women who exclusively breast-feed their babies usually do not ovulate for at least 3 months following delivery, an occasional nursing mother will ovulate early, and thus contraception use should be advised. Women who receive rubella immunization, which is a live vaccine, are advised to delay pregnancy for at least a month. Women should be cautioned about postpartum depression. Instructions from the pediatrician regarding care of the neonate should be reinforced, as well as encouraging follow-up should any problems occur. Although its value is open to question, it is customary to schedule an office visit 4 to 6 weeks postpartum. At this visit, inquiries should be made regarding breast-feeding, continued bleeding, resumption of sexual intercourse, use of appropriate contraception, and any difficulty with regard to voiding or bowel movements. A focused physical examination can be performed, and a Pap smear obtained, if indicated. Women should be counseled regarding emotional lability in the postpartum period and reassured that intervention is available should she experience clinically significant depression. Complications of the Puerperium Puerperal Infection Fever (a temperature of at least 38°C or 100.4°F) is the most commonly observed sign of puerperal infection. The differential diagnosis includes infections of the perineum, vagina, uterus, parametrium, bladder, kidneys, or breast. Noninfectious causes of fever such as thrombophlebitis and breast engorgement must also be considered. After vaginal delivery, endometritis is infrequent but may be presaged by chorioamnionitis. Various obstetric factors increase the risk of endometritis; two of the most consistent are prolonged rupture of the membranes and prolonged labor, both of which are often associated with multiple intrapartum cervical examinations, which also add to the risk. Endometritis is usually caused by normal vaginal flora, including Escherichia coli, ß-hemolytic streptococci, facultative organisms such as Streptococcus faecalis, and anaerobes such as Peptostreptoccus, Bacteroides, and Prevotella species. Most infections involve mixed flora, but progression to an anaerobic infection is a particular risk following cesarean delivery. Women suffering with serious puerperal infection of the reproductive tract frequently complain of chills, headache, malaise, and anorexia. Physical examination usually reveals tachycardia and often pallor. If the uterus is involved, it is usually tender and may be large and boggy. Lochia may be diminished if the cervix is blocked, or it may be profuse and malodorous. Parametrial involvement, peritonitis, and pelvic thrombophlebitis may complicate the illness. Endotoxic shock or a virulent puerperal sepsis may develop in a relatively short time and can be fatal. Prompt and aggressive management with broad-spectrum antibiotics—and potentially surgery—is mandatory (see Chapter 19). Postpartum Hemorrhage Postpartum hemorrhage has been defined as a blood loss of greater than 500 mL during or after the third stage of labor. This definition is somewhat paradoxical, as careful measures of blood loss following delivery indicate that the average is approximately 500 mL. After hypertensive disorders and embolism, postpartum hemorrhage is the major cause of maternal mortality for women who reach the second trimester. The major cause of postpartum hemorrhage is uterine atony, or failure of the uterus to contract sufficiently after delivery. This failure inhibits the major hemostatic mechanism of the postpartum uterus—myometrial contraction with constriction of placental bed arteries and veins. Risk factors for uterine atony include prolonged or oxytocin-augmented labor, uterine overdistension (as with polyhydramnios, fetal macrosomia, or multiple gestation), chorioamnionitis, grand multiparity, and a history of postpartum hemorrhage. Halogenated anesthetic agents also inhibit uterine contractility, and very rapid labors have been associated with failure of the uterus to contract adequately after delivery. Additional causes of postpartum hemorrhage are vaginal or cervical lacerations, and retained products of conception. Most serious hemorrhages occur within the first 24 hours after delivery but may occur even weeks after delivery. If hemorrhage occurs, and uterine atony is the presumptive cause, bimanual uterine compression (using a fist in the anterior vaginal fornix and the other hand on the abdomen, open and compressing the posterior uterine wall) should be performed and dilute intravenous oxytocin (40 to 80 units per L) administered. Large bore intravenous access should be secured. If bleeding persists, the vagina and cervix should be inspected for bleeding lacerations, and these should be sutured if found. The uterus should be manually explored for retained placental fragments or membranes. Intramuscular injection of methylergonovine maleate (0.2 mg) or prostaglandin F 2 (0.25 mg) can aid in stimulating contraction of the atonic uterus. Both may exacerbate or provoke

hypertension. Preparation should be made to replace blood and blood products as necessary to prevent the development of a consumptive coagulopathy. If contractions cannot be stimulated in a refractory atonic uterus, laparotomy with uterine artery ligation, hypogastric artery ligation (if a skilled, experienced surgeon is available and the woman is hemodynamically stable) or hysterectomy may be required. If bleeding from a vaginal wall laceration is persistent, packing, angiography, and selective embolization may be helpful. Depression Many mothers (at least half) feel an emotional letdown during the first few days after delivery. These “maternity blues” are characterized by bouts of sadness, crying, and lability of mood. This state is usually, but not always, short-lived and seldom persists beyond 2 weeks. No specific therapy other than emotional support is indicated. More seriously, up to 20% of women may experience a major postpartum depression, characterized by such symptoms as insomnia, pessimism, lethargy, feelings of inadequacy, inability to cope, and fatigue. In women with these symptoms, it is important to rule out postpartum hypothyroidism. Both counseling and medication (usually with a selective serotonin reuptake inhibitor) may be indicated. Profound, persistent depression or that associated with lack of interest in the infant, suicidal or homicidal thoughts, hallucinations, or psychotic behavior should prompt psychiatric consultation. The recurrence risk for postpartum depression is high, and prophylactic therapy beginning 2 to 3 weeks before delivery may be indicated. An earlier postpartum visit should be considered for women at risk of postpartum depression including those with a prior history of depression not associated with pregnancy. Puerperal psychosis (hallucinations, suicidal impulses) is unusual but may complicate as many as 1 in 500 pregnancies. It is an emergency which should be immediately attended to by a psychiatrist to prevent the mother from harming herself or her infant. Breast-feeding Breast-feeding has undergone somewhat of a renaissance in the United States over the past 30 years: in 1971 only one-fourth of mothers breast-fed at hospital discharge, versus almost two thirds in 1998. In 1997, the AAP summarized the manifest benefits of breast-feeding for the infant ( Table 2.3).

TABLE 2.3. Research on established and potential protective effects of human milk and breast-feeding on infants

Breast-feeding mothers benefit as well. Suckling-triggered oxytocin release promotes uterine contractions and leads to diminished blood loss. Oxytocin and prolactin levels are elevated in breast-feeding mothers, enhancing feelings of attachment and relaxation. Women who breast-feed face lower risks of ovarian and premenopausal breast cancer, and are less subject to pregnancy-induced obesity. By delaying ovulation, breast-feeding contributes to longer interpregnancy intervals. Almost all women may breast-feed. However, some women should not. According to the ACOG, these include women who: take street drugs or do not control alcohol use have an infant with galactosemia are infected with HIV have active, untreated tuberculosis take certain medications are undergoing treatment for breast cancer. Additionally, women with active varicella infections, herpes simplex virus infections of the breast, and active hepatitis A and B (until the infant has been passively and actively vaccinated) should refrain from breast-feeding. The list of medications contraindicated in breast-feeding is not extensive. It includes bromocriptine, cocaine, cyclophosphamide, cyclosporine, doxorubicin, ergotamine, lithium, methotrexate, phencyclidine, phenindione, and radioactive iodine and other radiolabeled elements. Women who want to breast-feed but who receive radioactive elements on a one-time or short-term basis (as for a diagnostic study) can pump and discard their milk until the radioactivity has cleared their bodies. Premature infants can be breast-fed, but may have special nutritional requirements, especially those born very prematurely. Consultation is necessary to ensure adequate nutrition for such babies. Mothers whose babies are temporarily unable to breast feed may pump their breasts (which is most efficiently done with an electric pump) and the milk may be administered to the baby or stored for later use (frozen unless use is planned within 2 days). Lack of education and knowledge, both on the part of caregivers and mothers, is the biggest barrier to breast-feeding. To redress this deficiency, ACOG has compiled two lists of resources ( Table 2.4, Table 2.5). It is incumbent upon obstetric and pediatric care providers to provide education, support, and assistance to allow women to begin and maintain successful breast-feeding. These activities should begin at the first prenatal visit (or even sooner, e.g., routine gynecologic health visits) and continue throughout the pregnancy and postpartum period. Hospital protocols and care delivery should be conducive to breast-feeding, and specific practices to support breast-feeding have been enumerated ( Table 2.6). Since most women are discharged before their milk comes in, most questions and problems with breast-feeding are handled on an outpatient basis. Therefore, to achieve successful, sustained breast-feeding among their patients, physicians and their office staff must have the ability to field questions and promptly correct problems with nursing.

TABLE 2.4. Patient breast-feeding education materials

TABLE 2.5. References for health care workers and patients seeking in-depth information about breast-feeding

TABLE 2.6. Ten hospital practices to encourage and support breast-feeding a

Women who make an informed decision not to breast-feed should be reassured that milk production will gradually diminish over the first few days after delivery. During this time, a properly fitted bra, ice packs, and oral analgesics should minimize the discomfort of breast engorgement. Medical therapies to hasten this process are no longer advised. Other than encouraging breast-feeding women to follow a well-balanced, varied diet, and to keep adequately hydrated, no specific dietary modifications are necessary. It is reasonable to recommend a multivitamin with iron supplementation to all nursing women, especially strict vegetarians and those who do not eat dairy products. Nursing women have a greater delay in resumption of intercourse, compared to those who bottle-feed. In some women, this is clearly attributable to discomfort due to relative vaginal atrophy and decreased lubrication from low estrogen levels. If lubrication during intercourse does not provide relief, women with a well-established milk supply may benefit from vaginal estrogen cream. Not working, or working part-time, is associated with a longer period of breast-feeding. However, working is rarely the reason given by women for not breast-feeding. The barriers to breast-feeding while working include lack of a flexible or part-time work schedule, inadequate maternity leave and job security, lack of on-site day care, and lack of an appropriate place to pump or breast-feed. Often, the most significant problem facing a mother trying to work and breast-feed is the people around her who do not understand her choice to nurse. Increasingly, however, societal attitudes, and social policies and laws have become more favorable to, and supportive of, breast-feeding. Nursing Problems Evaluating problems with nursing requires the observation of the mother and baby while nursing, to which the office setting may not be conducive—and thus the advantage of lay assistance and lactation specialists. Adequate rest is important, and encouragement to rest should accompany almost all calls about problems with nursing. Concerns about Insufficient Milk Supply The most common reason women give for discontinuing breast-feeding is a concern over insufficient milk supply. Milk letdown can be inhibited by cold, pain, and emotional stress and is sometimes difficult to achieve using a breast pump. A quiet, private setting and pictures of the baby to look at may help. Thus simple supportive measures and environmental modification may be all that is required. If the problem proves refractory, oxytocin in nasal-spray form may be helpful. The spray contains 40 USP units (IU) per milliliter, which is administered into one or both nostrils 2 to 3 minutes prior to nursing or pumping. Painful Nipples Painful nipples are usually caused by problems with positioning and latching on. For some women, comfort will be gained by leaving their nipples exposed and using either a lamp or hair dryer to dry their nipples after feedings. Additionally, breast shields and purified lanolin cream may help provide relief. Mastitis One to two percent of breast-feeding women will develop mastitis, which can occur any time, but is most frequent in the first through fifth week of breast-feeding. Typical symptoms are unilateral breast pain, chills, and malaise. Signs include fever, and an erythematous, indurated segment of breast. Staphylococcus aureus is the most common agent, although a broad- spectrum of bacteria may be causative. In women not allergic or sensitive to penicillin, a 7- to 10-day course of oral dicloxacillin, 500 mg q.i.d., is usually curative. Oral hydration is advised and antipyretics may lessen discomfort and fever. For penicillin-intolerant women, erythromycin may be substituted. Breast-feeding should continue, as drainage of the breast has therapeutic value. Neglected or recurrent infections and infections with resistant organisms can lead to a breast abscess. Treatment of an abscess usually involves intravenous antibiotics and aspiration or surgical drainage.

NEWBORN CARE Immediate Assessment and Resuscitation The transition from fetus to newborn infant is the most dramatic physiologic change that occurs in the human life span. The fetus that received all of its oxygen and nutritional needs via the placenta must now use two entirely different, essentially dormant organ systems to meet these needs. The circulation is rerouted and the pulmonary bypass paths (ductus arteriosus, foramen ovale, and umbilical circulation) are no longer used. The lungs and left side of the heart that once handled about 15% of the circulation must now deal with 100% of the circulation in series with the right heart and remainder of the body. The fluid-filled, unexpanded lungs must be inflated and cleared of fluid to allow gas exchange, as they are now the sole source of oxygen for the infant. Resuscitation in the delivery room is geared toward helping the infant accomplish this transition. The AAP Neonatal Resuscitation Program (NRP) is a training course that all those specializing in pediatrics complete and is strongly encouraged for all individuals who deliver infants or attend deliveries; in fact, many institutions now require current certification for delivery room personnel. The resuscitation algorithm from the current NRP is shown in Figure 2.16.

FIG. 2.16. Algorithm for resuscitation of the newly born infant. (From American Academy of Pediatrics © 2000; with permission.)

While every delivery should be attended by personnel trained in newborn resuscitation, the presence of personnel experienced in resuscitation is most desirable for deliveries where the need for resuscitation is more likely. Examples of these situations are summarized in Table 2.7.

TABLE 2.7. Antepartum factors and intrapartum factors

Meconium Meconium passage prior to birth occurs in up to 20% of term deliveries and is a common reason for neonatal resuscitation. In the 1970s it was recognized that early suctioning by the obstetrician or pediatrician decreased the incidence of meconium aspiration syndrome (MAS). In the 1980s, routine suctioning of the oropharynx at the perineum or endotracheal suctioning were standard practice; however, it became clear that not all cases of MAS could be prevented and that aspiration in utero could occur. In the 1990s, many neonatologists were concerned that vigorous babies with thin meconium had a very low risk of MAS, and might actually do better if not intubated for tracheal suctioning. Unfortunately, the majority of studies pertinent to this issue have not been prospective, randomized, or controlled. The most recent Cochrane Library review concluded that until further evidence is available, endotracheal suctioning for meconium should be reserved for those infants who are depressed or have respiratory difficulties. The Apgar Score In 1952, Dr. Virginia Apgar devised a 0 to 10 scale scoring system intended to focus attention on the newborn and allow the systematic assessment of its condition and immediate needs ( Table 2.8). Since that time, Apgar scores have become a mainstay in immediate newborn assessment. The Apgar score is a useful tool to guide those charged with assessment and care of the newborn. It provides an overall picture of infant status, and has been used in clinical research to correlate later outcomes with initial appearance. It cannot be used to predict neurologic outcome without other tests or examinations, nor is it a validated tool for determining asphyxia. Moreover, it was never intended for use in premature infants. Scores are assigned at 1 minute of age, and every 5 minutes thereafter until the score is over 7, or until 20 minutes of age. Clearly, resuscitation should begin if necessary prior to the 1-minute score.

TABLE 2.8. Apgar mnemonic

General Evaluation and Treatment Every newborn infant should have a thorough physical examination. The newborn exam is focused on identification of congenital abnormalities, birth trauma, and signs of transitional difficulties ( Table 2.9). Although infants of 35 or more weeks and over 2 kg will generally be able to maintain their temperature and feed successfully, all infants born prior to 37 weeks may require more frequent monitoring of temperature and blood glucose. Infants who are large or small for their delivery dates are at higher risk for hypoglycemia. If dates are uncertain, a Dubowitz or Ballard examination may be performed, ideally between 24 to 48 hours of age. In addition to monitoring blood sugar prior to established feeding, routine laboratory work often includes blood type, Coombs test, and hematocrit. A positive Coombs test can be caused by Rh or ABO incompatibility. Hematocrit levels between 40% to 65% are acceptable. Most states screen for phenylketonuria (PKU) and congenital hypothyroidism. Other tests used by some states include screens for congenital adrenal hyperplasia, cystic fibrosis, and other inborn errors of metabolism. These tests may be initiated prior to discharge according to state mandates. Infants born in U.S. hospitals are given either dilute silver nitrate drops or erythromycin ophthalmic ointment to prevent Neisseria gonorrhoeae conjunctivitis, which was once the most common cause of congenital blindness. Vitamin K is administered intramuscularly to prevent vitamin K deficiency bleeding (VKDB). Factors that can contribute to this deficiency include low vitamin K stores at birth, poor placental transfer of vitamin K, low levels of vitamin K in breast milk, and sterility of the gut. Because standard commercial infant formulas contain supplemental vitamin K, VKDB is almost exclusively a problem of breast-fed infants. The most common sites of bleeding are the umbilicus, mucous membranes, gastrointestinal tract, the penis after circumcision, and skin (from venipuncture). Intracranial bleeding can occur and is the main cause of mortality and long-term morbidity. In the United States, all infants are actively immunized against hepatitis B, the initial dose of which is generally given on the day of birth. If the mother is positive for hepatitis B surface antigen or if her status cannot be determined within a few days, the infant should also receive passive immunization within 7 days in the form of hepatitis B immune globulin.

TABLE 2.9. Examination of the newborn

Early Onset Sepsis One of the most serious problems encountered in the immediate newborn period is early onset sepsis, defined as occurring in the first 5 days after birth. The current estimated rate in the United States is 1 to 2 per 1000 live births. Group B streptococci (GBS) remain the most common causal organisms, but with the routine use of intrapartum antibiotic prophylaxis for GBS, Escherichia coli now constitutes a larger proportion of causative organisms. The symptoms of neonatal sepsis include respiratory distress, lethargy, poor feeding, hypotonia, seizures, and shock. The management of infants with symptoms of sepsis includes blood and, if indicated, cerebrospinal fluid cultures, followed by the administration of broad-spectrum intravenous antibiotics, generally a penicillin in combination with an aminoglycoside. The management of asymptomatic infants of mothers with risk factors for neonatal sepsis is less clear and has become a topic of considerable debate since the routine use of intrapartum GBS chemoprophylaxis. Based on expert opinion and limited data the following general approach has been recommended. Because of an increased risk for early-onset GBS septicemia and greater difficulty in assessing symptoms, asymptomatic infants born prior to 35 weeks gestation should be evaluated with a complete blood count and blood culture, and observed in the hospital without empirical antimicrobial therapy for at least 48 hours. In infants born at and beyond 35 weeks, clinical assessment is more dependable. Routine laboratory evaluation is not recommended in these infants. They should, however, be observed in-hospital for at least 48 hours. Hyperbilirubinemia and Discharge Planning Timing of newborn discharge has been a topic of controversy for nearly 20 years. Prior to and during much of this time, the traditional stay was 3 days for a vaginally delivered baby and 5 days for those delivered by cesarean (due to the mother's extended stay). This minimum of 3 days allowed detection of the most common serious problems of the newborn, such as ductal-dependant congenital heart disease, early onset sepsis, hyperbilirubinemia, and failure to establish adequate breast-feeding. Discharges at 24 hours (or less) became popular in the late 1980s, a practice driven largely by third-party payors. While numerous small studies were performed to assess the safety of early discharge, most suffered from insufficient numbers to rule out an increase in adverse outcomes that, individually, are relatively uncommon. Bilirubin encephalopathy, a disorder rarely seen for the past 30 to 40 years, has recently been reported to be on the rise. Bilirubin peaks at 5 days in formula-fed infants but may peak as late as 7 to 10 days in breast-fed infants. Breast milk production is minimal in the first 48 hours after delivery and does not reach 80% of full volume until 4 days postpartum. Because of this time frame, it is not easy to predict at 3 days of age which breast-fed infants with a negative Coombs test will be at risk of dangerously high bilirubin levels (>25 mg/dL), and it is nearly impossible to do so at less than 48 hours. In recognition of these issues, the pendulum appears to be swinging back to a 3-day stay for most breast-fed infants. Infants readmitted after discharge prior to 48 hours are most likely breast-fed and jaundice is the most common reason for admission. In addition to insufficient breast milk intake, hyperbilirubinemia can result from relative polycythemia, bruising, cephalohematoma, or be idiopathic. The current AAP guidelines for management of term infants with nonhemolytic jaundice are given in Table 2.10.

TABLE 2.10. American Academy of Pediatrics hyperbilirubinemia treatment guidelines: total serum bilirubin (mg/dL)

SUMMARY POINTS Labor is defined as regular uterine contractions that lead to effacement and dilation of the cervix. Labor can be conceptualized as depending on the Power, the Passage, and the Passenger. A variety of management approaches to labor have been employed and evaluated, and no single approach is clearly superior. Both infection and hemorrhage can complicate normal labor and delivery, and caregivers must be vigilant for these complications and competent in their management. Encouraging and providing support for breast-feeding has significant benefits for both infants and mothers. Every delivery should be attended by personnel trained in newborn resuscitation. Newborn hospital stays shorter than 3 days may compromise the detection of hyperbilirubinemia in breast-fed infants. SUGGESTED READINGS American Academy of Pediatrics Provisional Committee for Quality Improvement Hyperbilirubinemia Subcommittee. Practice parameter: management of hyperbilirubinemia in the healthy term newborn. Pediatrics 1994;94:558–565. American College of Obstetricians and Gynecologists. Breastfeeding: maternal and infant aspects. ACOG Educational Bulletin No. 258, July 2000. American College of Obstetricians and Gynecologists. Prevention of early-onset Group B streptococcal disease in newborns. ACOG Committee Opinion #279, December 2002. Carroli G, Belizan J. Episiotomy for vaginal birth. Cochrane Database of Systematic Reviews 2001;issue 3. Chua S, Arulkumaran S. Poor prognosis in labor, including augmentation, malpositions and malpresentations. In: James DK, Steer PJ, Weiner CP, Gonik B, eds. High risk pregnancy, second ed. London: Harcourt Brace, 1999:1105. Fraser WD, Turcot L, Krauss I, Brisson-Carrol G. Amniotomy for shortening spontaneous labour. Cochrane Database of Systematic Reviews 2001;issue 2. Fraser WD, Marcoux S, Krauss I, et al. Multicenter, randomized, controlled trial of delayed pushing for nulliparous women in the second stage of labor with continuous epidural analgesia. Am J Obstet Gynecol 2000;182:1165–1172. Halliday HL. Endotracheal intubation at birth for preventing morbidity and mortality in vigorous, meconium-stained infants born at term. Cochrane Database of Systematic Reviews 2002;issue 1. Hannah ME, Ohlsson A, Farine D, et al. Induction of labor compared with expectant management for prelabor rupture of the membranes at term. N Engl J Med 1996;334:1005–1010. Hodnett ED. Caregiver support for women during childbirth. Cochrane Database of Systematic Reviews 2001;issue 2. Hofmeyr GJ. Amnioinfusion for meconium-stained liquor in labour. Cochrane Database of Systematic Reviews 2001;issue 2. Kattwinkel J, ed. Neonatal resuscitation, fourth ed. Elk Grove Village, IL: American Academy of Pediatrics and American Heart Association, 2000. Prendiville WJ, Elbourne D, McDonald S. Active versus expectant management in the third stage of labour. Cochrane Database of Systematic Reviews 2001;issue 2. Radmacher P, Massey C, Adamkin D. Hidden morbidity with “successful” early discharge. J Perinatol 2002;22:15–20. World Health Organization Maternal Health and Safe Motherhood Programme. World Health Organization partograph in management of labour. Lancet 1994;343:1399–1404.

Chapter 3 Obstetric Analgesia and Anesthesia Danforth’s Obstetrics and Gynecology

Chapter 3 Joy L. Hawkins

Obstetric Analgesia and Anesthesia

PAIN OF PARTURITION SYSTEMIC ANALGESIA AND SEDATION Use of Systemic Medications Systemic Narcotics Patient-controlled Intravenous Analgesia Narcotic Antagonists Sedative Drugs Inhalational Agents REGIONAL ANALGESIA Local Anesthetic Agents Side Effects of Local Anesthetic Drugs Use of Regional Anesthetic Blocks Complications of Regional Block Analgesia OTHER METHODS OF PAIN RELIEF Prepared Childbirth Hypnosis Acupuncture Biofeedback Transcutaneous Electrical Nerve Stimulation Intracutaneous Nerve Stimulation ANESTHESIA FOR CESAREAN DELIVERY Epidural Anesthesia Subarachnoid or Spinal Anesthesia General Anesthesia Analgesia after Cesarean Section SUMMARY POINTS REFERENCES

The purpose of this chapter is to acquaint the obstetrician with the various techniques of obstetric analgesia (pain relief) and anesthesia (for surgical procedures) and to describe their indications, advantages, disadvantages, and complications. The technical aspects, including the methods of administration, will not be described in detail. Readers seeking specific information on how to perform the various obstetric anesthetic techniques are referred to one of the basic obstetric anesthesia textbooks ( 1 ). Obstetric analgesia or anesthesia refers to the multiple techniques useful for the alleviation of the pain associated with labor, delivery, or surgery. The choice of an appropriate analgesic technique must be made by the patient, the obstetrician, and the anesthesiologist and should take into consideration the patient's anatomy and physiology, the status of her fetus, the obstetric plan for delivery, and the pharmacology of the drugs to be employed ( 2 ). Most women now request some form of analgesia during childbirth ( Table 3.1).

TABLE 3.1. Types of labor analgesia provided by size of hospital in three time periods

PAIN OF PARTURITION Increasing dilation of the cervix, contraction and distention of the uterus, and distention or tearing of the vagina, vulva, and perineum causes the pain that occurs during labor and delivery. Pain may be generated through stretching or application of pressure to adjacent pelvic organs. The pain that occurs in the first stage of labor increases in severity as the cervix becomes more dilated. The onset of pain lags approximately 15 to 30 seconds behind the onset of the uterine contraction and is first perceived when the intraamniotic pressure reaches 15 mm Hg above that of resting tonus. In the second stage of labor, sharp pain occurs as the tissues of the vagina and perineum are stretched. Stretching stimulates the second, third, and fourth sacral nerve roots, which carry nociceptive information to the spinal cord through the sensory fibers of the pudendal nerve. Adnexal pressure and traction on the bladder, urethra, rectum, and peritoneum also contribute to the pain of parturition. Compression of the lumbosacral plexus by the fetal head, particularly in the occiput posterior position, may cause pain even before the onset of labor. The pain of uterine contractions is conducted through small sensory nerve fibers of the paracervical and inferior hypogastric plexuses to join the sympathetic nerve chain at L2-3. The ascending fibers enter the spinal cord through the nerve roots of T-10 to T-12, with a variable contribution from L-1 ( Fig. 3.1). Because the cutaneous branches of the lower thoracic and upper lumbar nerves migrate caudally for a considerable distance before they innervate the skin, the pain of uterine contractions is often referred to the area over the upper sacrum and the lower lumbar spine.

FIG. 3.1. Sympathetic nerve supply of the uterus from the pelvic and abdominal distribution. The uterine nerves arise from the upper part of the uterus (i.e., upper uterine segment), the contraction of which contributes to pain; from the lower part of the uterus (i.e., lower uterine segment), the distention of which contributes to pain; and from the cervix, the dilation of which contributes to pain. The ovarian nerve supplies the ovary, fallopian tube, broad ligament, round ligament, and the side of the uterus, and it communicates with the uterine plexus. The sympathetic efferent and afferent fibers are shown together. (Adapted from Abouleish E. Pain control in

obstetrics. Philadelphia: JB Lippincott Co, 1977.)

Pain from the uterus and cervix is transmitted through the small-diameter myelinated d-A fibers and unmyelinated C fibers. Because there are relatively fewer nociceptive afferent nerves from visceral structures than from somatic structures, visceral pain is perceived as being diffuse and difficult to localize. These visceral afferents also synapse on and excite the same dorsal horn neurons as afferents from somatic structures. This arrangement is responsible for the phenomenon of referred pain. The gate theory of Melzack and Wall holds that stimulation of the large cutaneous ß-A nerve fibers closes a “gate” in the substantia gelatinosa of the spinal cord, preventing pain impulses from being carried rostrally by the d-A and C nerve fibers. This theory forms the basis for the use of acupuncture, transcutaneous electrical nerve stimulation, and intracutaneous nerve stimulation with sterile water injections for the relief of pain associated with parturition. Pregnancy appears to reduce anesthetic requirements. It has been postulated that high progesterone levels lead to increased quantities of endogenous endorphins, which may increase the maternal threshold to pain. One study correlated pain intensity during labor and plasma levels of ß-endorphin. The lowest endorphin levels were found after abolition of labor pain by epidural analgesia. The highest concentrations were observed in the first few minutes after delivery, immediately after cessation of the severe pain of expulsive labor. The nature of the pain of labor varies in intensity with the stages of labor. The intensity of pain is related to physical factors such as the strength and duration of uterine contractions, the rapidity of cervical dilation, the degree of distention of the vaginal and perineal tissues, the requirement for operative delivery, and the size, presentation, and position of the infant. Augmentation of labor with oxytocin increases the strength and pain of uterine contractions. The primiparous woman may perceive greater pain than the multipara who enters labor with more advanced cervical dilation and who may also be more psychologically prepared ( Fig. 3.2). Exhaustion, psychological factors, and protracted nausea and vomiting may also increase the parturient's perception of labor pain.

FIG. 3.2. Comparison of pain scores obtained from women during labor and from patients in a general hospital pain clinic or emergency department using the McGill pain questionnaire. (From Melzack R. The myth of painless childbirth. Pain 1984;19:321–37.)

Pain management is an important part of modern obstetric care. The obstetrician should appreciate the importance of providing pain relief during labor through the use of nonpharmacologic techniques, systemic analgesics, or regional block analgesia. During the second stage of labor, additional analgesia may also be needed through perineal extension of a segmental epidural or through the use of pudendal or spinal blocks.

SYSTEMIC ANALGESIA AND SEDATION In the management of labor pain, systemic narcotics are usually considered to be the first step beyond the less invasive or “natural” methods such as massage, water baths, and birth attendants (doulas). They may also be necessary for patients who are not candidates for regional analgesia. Research indicates the analgesic effects of parenteral agents used in labor is limited and the primary mechanism of action is heavy sedation ( 3 ). Although narcotics may be effective for some patients in relieving the pain of labor, their side effects prohibit the use of large doses. The physician must balance maternal and neonatal respiratory depression with effective relief of the pain of labor. Because the pain of labor occurs intermittently with contractions, maternal hyperventilation during a contraction leads to hypoventilation for the 2 to 3 minutes between contractions, especially when narcotics have depressed the carbon dioxide response curve. There are advantages and disadvantages of all available narcotics, and a drug should be chosen with knowledge of its side effects and pharmacokinetics ( Table 3.2).

TABLE 3.2. Parenteral medications for labor analgesia

Use of Systemic Medications Because the use of systemic narcotics and tranquilizers does not require special training or the availability of anesthesia personnel, they are used extensively in the United States to provide labor analgesia (see Table 3.1). These compounds do not induce fetal heart rate (FHR) abnormalities other than changes in variability and rarely sinusoidal heart rate patterns, nor do they cause fetal acidosis. The drug-related adverse maternal effects of narcotics can include nausea, vomiting, pruritus, sedation, decreased gastric motility, loss of protective airway reflexes, and hypoxia due to respiratory depression. The adverse neonatal effects of these agents include central nervous system (CNS) depression, respiratory depression, impaired early breast-feeding, altered neuroadaptive behavior, and decreased ability to regulate body temperature. To minimize these side effects, the lowest effective dosage should be employed, and the timing with respect to delivery must be carefully considered. Resuscitation equipment should be kept at hand, and naloxone, used to antagonize opioids, should be readily available. Benzodiazepine effects may be reversed with flumazenil (Romazicon). Systemic Narcotics Meperidine Meperidine (Demerol) has achieved wide popularity for systemic analgesia during labor. It is preferred over morphine because it produces less emesis and does not depress the newborn carbon dioxide response curve as much as morphine. It can be administered intravenously or intramuscularly during labor. Current usage most often consists of small, incremental intravenous doses of 25 to 50 mg. Small doses can also be used to treat shivering. Placental transfer of meperidine occurs rapidly. Maximal depression of the infant occurs when delivery takes place 2 to 4 hours after maternal intravenous or intramuscular administration. Delivery of the infant within 1 hour of administration produces little evidence of newborn depression. A study using meperidine for labor analgesia by a patient-controlled infusion device found 5% of infants required naloxone at delivery ( 4 ). Meperidine has as its principal metabolite the compound normeperidine, which is equipotent with meperidine in its ability to produce respiratory depression and can also cause seizures. Repeated intravenous administration of small doses of meperidine leads to increasing maternal and fetal levels of normeperidine. Meperidine has an elimination half-life in neonatal blood of 22.7 hours, and that of normeperidine is measured in days, reflected in abnormal neurobehavioral scores for up to 3 days. Morphine Morphine is pharmacologically more potent than meperidine by a factor of approximately 10. Morphine depresses the newborn carbon dioxide response curve more than meperidine, perhaps due to greater permeability of the infant brain to morphine. Because of this reputation, morphine has virtually disappeared from the armamentarium of analgesic drugs used by the obstetrician for the management of labor. Fentanyl Fentanyl (Sublimaze) is a potent synthetic narcotic with analgesic activity approximately 100 times that of morphine. Its onset of action is rapid, and its duration of activity is short (i.e., 20–30 minutes) because of its rapid distribution from plasma. The terminal drug elimination half-life after a single small dose is 1 to 2 hours. Fentanyl is highly bound to protein, which may limit its placental transfer. It has no active metabolites. Fetal–maternal blood concentration ratios average 0.31 over the first 10 minutes after intravenous administration. Fentanyl produces moderate analgesia and mild sedation. There may be a brief period of decreased FHR variability, but no other disturbing FHR patterns have been reported. Comparative studies with meperidine indicate that the need for newborn naloxone (Narcan) administration is greater after use of meperidine than after administration of fentanyl.

Nalbuphine Nalbuphine (Nubain) is a potent narcotic agonist–antagonist agent which, at equianalgesic doses, produces respiratory depression equivalent to that of morphine. The advantage and disadvantage of nalbuphine is that as the dosage is increased, a ceiling effect is seen for respiratory depression, and unfortunately also for analgesia. Maximal respiratory depression occurs with a dose of 30 mg in a 70-kg adult. Sedation and dysphoric reactions may also occur. Reversal of other opioid effects may precipitate withdrawal in a susceptible patient. A possible transient depressive effect of nalbuphine on the fetal CNS has been identified. Butorphanol Butorphanol (Stadol) is another synthetic narcotic with agonist–antagonist properties. It is five times more potent than morphine and 40 times more potent than meperidine. It has achieved moderate popularity in the United States in the management of the pain of the first stage of labor. It is usually administered intravenously in doses of 1 to 2 mg. Butorphanol exhibits the same ceiling effect for analgesia and respiratory depression as nalbuphine. Maternal side effects may include sedation, dysphoric reactions, and reversal of other opioid effects. Newer Opioids Sufentanil and remifentanil, the newer synthetic opioids, have not been studied extensively for systemic analgesia during labor. Patient-controlled Intravenous Analgesia Intravenous patient-controlled analgesia (PCA) is widely available and provides pain relief through self-administration of intravenous opioids. Fentanyl, remifentanil, and meperidine are the analgesics most commonly employed with this technique. The infusion pump is programmed so that the patient receives an incremental dose when she pushes the button, followed by a lockout interval when additional requests by the patient will not be administered. An hourly maximum may also be programmed. A basal infusion is rarely used in labor because of the risk of respiratory depression between contractions. The actual settings are dictated by the pharmacokinetics of the narcotic chosen. The main advantage of PCA is improved patient satisfaction due to a feeling of control and not having to wait for a nurse to bring pain medication. Use of PCA may also decrease nursing staffing requirements. In a study by Rosenblatt and colleagues ( 5 ), metoclopramide (Reglan) was used as an analgesic adjunct to PCA for patients undergoing prostaglandin induction of labor for second-trimester termination of pregnancy. Patients were given intravenous metoclopramide, 10 mg, or saline placebo followed by PCA-administered morphine. Those receiving metoclopramide used 54% less morphine and had lower pain scores. Narcotic Antagonists Naloxone Because all narcotics cross the placenta and can produce respiratory depression in the neonate, availability of an effective antagonist is essential. Naloxone reverses opioid-induced respiratory depression without producing side effects of its own. Because it also reverses analgesia, its prophylactic use is not advised. Naloxone may be administered to the parturient as an intravenous bolus of 0.1 to 0.4 mg to treat maternal respiratory depression. Care must be taken to titrate naloxone to the desired effect, since large doses have been implicated in the causation of myocardial infarction, pulmonary edema, and severe hypertension. Naloxone, 0.01 mg per kg, may also be administered intravenously, intramuscularly, or through the endotracheal tube to the newborn to reverse the respiratory depressant effects of placentally transferred narcotics. The effect is usually apparent within a few minutes and persists for as long as 2 hours. The neonate must be carefully observed for evidence of renarcotization, because the half-life of naloxone is less than that of most narcotics. Sedative Drugs Benzodiazepines The principal benzodiazepine drugs are diazepam (Valium) and midazolam (Versed). Diazepam has been used extensively in other parts of the world for seizure prophylaxis in patients with severe preeclampsia. However, because of its side effects on the newborn, it has found little favor in the United States. Newborns exposed to diazepam characteristically exhibit hypotonicity, hypoactivity, and impaired temperature regulation and metabolic response to cold stress. Midazolam is a newer benzodiazepine anxiolytic, a sedative drug with significant amnestic properties. It is five times more potent than diazepam and is soluble in water, a property that reduces pain associated with intravenous administration. Midazolam crosses the sheep placenta, achieving a fetal–maternal concentration ratio of 0.15. Its metabolites are inactive, and the drug is excreted more rapidly than diazepam. Midazolam has been used as an induction agent for cesarean delivery, but because of its ability to cross the placenta, it has produced neonatal respiratory depression and decreased body tone and temperature. Midazolam has not been recommended for use as a tranquilizer-sedative in labor, because its amnestic properties are unacceptable to most parturients. Barbiturates Barbiturates may be used in the latent phase of labor. Although they cause maternal sedation and decreased anxiety, barbiturates lack analgesic properties and may increase the perception of pain when given without concomitant administration of a narcotic. Most barbiturates have long elimination half-lives and readily cross the placenta. Prolonged neonatal effects have led to the virtual elimination of these drugs from use during labor. Other Sedatives Phenothiazine derivatives, such as promethazine (Phenergan), have been used in obstetrics and provide sedation and decrease nausea. Hydroxyzine, although not a phenothiazine, has similar properties when used in combination with narcotics. Maternal sedation is achieved without significant maternal or newborn side effects. It is important to remember that none of these drugs provide analgesia, and some parturients may object to the heavy sedation they cause. In addition, both medications are very painful intramuscular injections. They should have little place in the management of labor analgesia. Ketamine (Ketalar), when administered intermittently at low doses (10–15 mg), can produce analgesia in parturients without causing maternal loss of consciousness or neonatal respiratory depression. An improvement on this technique using low-dose ketamine infusion has been described by Maroof and associates ( 6 ). In this study, patients were given an intravenous bolus of ketamine, 0.5 mg per kg, upon achieving 4 cm of cervical dilation. This was followed by an infusion of ketamine at 0.25 mg per kg per hour. Patients served as their own controls, evaluating labor with and without the benefit of ketamine. Pain scores were significantly better with ketamine. The authors concluded that ketamine infusion produced acceptable analgesia throughout labor and was not associated with delirium or loss of consciousness in any patient. In addition, neonatal respiratory depression was not noted. The profound amnesia and potential for dysphoria or other psychomimetic effects when using ketamine limit its general use for labor analgesia. It is most useful for short painful procedures such as urgent forceps delivery or manual removal of the placenta. Inhalational Agents Nitrous oxide can be inhaled periodically with contractions in a 50% mixture with oxygen. During a painful contraction, the mother breathes from a mask connected to the regulator valve of a breathing circuit. A scavenging system is required by the Occupational Safety and Health Administration (OSHA) to eliminate exhaled waste anesthetic gases. Unfortunately, when it is breathed at the onset of a contraction, maximal analgesia is achieved only after the contraction has ended. When nitrous oxide is used in conjunction with narcotics, maternal oxygen saturation may decrease. Use of a pulse oximeter to ensure adequate maternal oxygenation is recommended. In practical terms, since almost all deliveries in the United States now take place outside the operating room, an anesthesia machine will probably not be available to safely administer inhalational agents.

REGIONAL ANALGESIA Local Anesthetic Agents Most local anesthetic agents share a common structure consisting of a hydrophilic amino group connected by an intermediate chain to a lipophilic aromatic residue. Their presumed mechanism of action is to block exchange of sodium and potassium ions across the cell membrane, probably through mechanical interruption of ion flow through cell wall channels. Local anesthetic drugs are manufactured as chloride salts. The nonionized base is able to diffuse across tissues, while the ionized form is actually the active component. The amounts of nonionized (mobile) and ionized (active) drug depend on the pK a of the local anesthetic and tissue pH. After injection of lidocaine (Xylocaine), the sensory nerve action potential decreases more sharply in pregnant women than in nonpregnant women. This implies that pregnant women have an increased susceptibility to the effects of local anesthetic agents ( 7 ). Local anesthetics belong principally to two groups, those of ester and amide configurations. Ester drugs are generally characterized by their rapid onset of action, short duration, and low toxicity. Chloroprocaine (Nesacaine) is a representative of this group. It is rapidly metabolized by serum pseudocholinesterase, forming paraaminobenzoic acid. Lidocaine, bupivacaine, ropivacaine, and levobupivacaine are representatives of the amide group. These drugs are more highly bound to protein and have a slower onset and a longer duration of action. They are metabolized in the liver. Toxicity is usually greater for amides than for drugs of the ester group ( Table 3.3).

TABLE 3.3. Characteristics of local anesthetics commonly used in obstetric anesthesia

Local anesthetic drugs are absorbed systemically. Distribution to the fetus depends on maternal tissue uptake, maternal blood concentration, uterine blood flow, and maternal and fetal metabolism and excretion. Fetal-tissue drug distribution is also affected by asphyxia. Fetal asphyxia leads to increased P a CO 2 , which results in cerebral and coronary vessel dilation and increased brain and myocardial blood flow. The increased perfusion of these organs with anesthetic drug leads to greater toxicity. The decrease in pH seen with fetal asphyxia results in increased ionized (active) drug in the fetal circulation and tissues, a phenomenon referred to as ion trapping. The clinical significance of this is unclear. Side Effects of Local Anesthetic Drugs Systemic Toxicity Systemic complications involving the use of local anesthetics include toxic blood levels of the drug and allergic reactions, as well as reactions due to epinephrine that is often added to local anesthetic solutions to retard systemic absorption and prolong duration of action. Maximal safe doses for healthy young adults are 7 mg per kg (300 mg) of lidocaine, 2 to 3 mg per kg (175 mg) of bupivacaine, and 20 mg per kg (1000 mg) of chloroprocaine (see Table 3.3). The most common reason for high blood levels of local anesthetic drugs is accidental intravascular injection. This most commonly occurs when an epidural catheter has been placed or migrated into a vein. To minimize accidental intravenous injection, gentle aspiration should be undertaken before each injection. Injection should be done slowly and incrementally with only 2 to 5 mL of local anesthetic drug to reduce the chance of a sudden increase in plasma levels. A marker such as epinephrine may be added to the local anesthetic solution so that intravascular injection will manifest as tachycardia. The infiltration of a local anesthetic agent into an area rich in vessels, such as the region of the uterine artery (e.g., paracervical block), pudendal vessels, or the epidural space may be associated with absorption of the drug through blood vessel walls. The serum levels tend to rise slowly, and toxic manifestations usually occur only after multiple injections. Repeated injections of slowly metabolized local anesthetic drugs, such as the amides, may lead to accumulation in the serum such that toxic levels are achieved. This phenomenon does not occur readily with esters such as chloroprocaine, which are rapidly metabolized (maternal serum half-life of 21 seconds and fetal serum half-life of 43 seconds for chloroprocaine). To minimize the likelihood of producing high serum levels, care should be taken to record the amount and concentration of local anesthetic solution and to limit use to approximately 25% less than the maximal safe dose. Signs and symptoms of local anesthetic drug toxicity include, in order of their appearance, a relaxed feeling, drowsiness, lightheadedness, tinnitus, circumoral paresthesias, metallic taste, slurred speech, blurred vision, unconsciousness, convulsions, and cardiac dysrhythmias and arrest. In 1983, the U.S. Food and Drug Administration issued an advisory, warning that 0.75% bupivacaine should no longer be used in obstetrics because of reports of bupivacaine-induced cardiac arrest occurring at blood levels of only 3 to 5 µg per mL. The advisory stated that the resuscitation in these cases had been “difficult or impossible despite apparently adequate preparation and appropriate management.” Inadvertent intravascular injection causes high serum levels, which produce cardiac arrest through blockade of the cardiac sodium channels, inhibiting repolarization of the nerve cell membranes of the conduction system of the heart. Bupivacaine has been found to bind avidly to nonspecific cardiac protein–binding sites, slowing the conduction of impulses arising in pacemaker cells and causing a dose-dependent reduction in the strength of myocardial contractility, leading to cardiac arrest. Management is best accomplished through prevention, as described previously. Therapy is symptomatic. Initial treatment includes the use of mask oxygen, a reliable intravenous line, and measures to ensure and protect the airway. These include use of cricoid pressure to occlude the esophagus, the availability of adequate suction, and the capability to perform endotracheal intubation, if needed. Adequacy of respirations must be ensured, if necessary by means of positive pressure ventilation with 100% inspired oxygen. The patient should be hyperventilated to help correct metabolic acidosis caused by seizure activity and decreased cardiac output. CNS hyperreactivity and convulsions are treated with thiopental (Pentothal) in small, incremental doses of 25 to 50 mg given intravenously, or with 1 to 5 mg of midazolam given intravenously. In the event of cardiovascular depression, elevate the lower extremities and verify left uterine displacement. Vasoactive drugs such as ephedrine, phenylephrine, epinephrine, and calcium may be employed to support the circulation. If cardiopulmonary resuscitation is indicated, the fetus should be delivered within 5 minutes to relieve maternal central venous compression and advanced cardiac life support (ACLS) protocols should be followed. Use of Regional Anesthetic Blocks Local Infiltration of the Perineum Local infiltration of the perineum is commonly performed when an episiotomy is needed and time or fetal head position does not allow a pudendal block to be administered. An average of 10 to 20 mL of local anesthetic solution is employed. The preferred drugs are lidocaine 1% or chloroprocaine 2%. Pudendal Block The pudendal block provides analgesia of the vaginal introitus and perineum. There are several advantages of this analgesic technique. Because the elapsed time between administration and delivery is short, there is relatively little systemic absorption and therefore little opportunity for the drug to directly affect the fetus. The block is easy to accomplish and provides analgesia of the perineum only. The disadvantages include the need for large drug doses (i.e., 10 mL on each side) and the potential for local anesthetic toxicity, hematoma, and the possibility of infection leading to retropsoas or subgluteal abscess. With the transvaginal approach, the ischial spine must first be identified. Through a guide, a needle is inserted into the vagina and directed laterally and posteriorly to the ischial spine. A submucosal wheal is made, and the needle is advanced into the sacrospinous ligament, where resistance is felt. As the needle passes the ligament, a loss of resistance is felt. The needle has now entered the pudendal canal, which contains the pudendal nerve and associated vessels ( Fig. 3.3). After aspirating the needle for blood, 3 to 5 mL of local anesthetic solution (usually lidocaine 1%) is injected, and the needle is advanced another 0.5 to 1 cm. If aspiration is again negative, 5 to 7 mL of solution is injected. A total of 10 mL is injected on each side. Approximately 10 minutes are required for anesthesia to occur. Chloroprocaine 1% to 2% may also be used for this block. Analgesia with chloroprocaine lasts less than 1 hour, but lidocaine analgesia is more prolonged.

FIG. 3.3. The pudendal nerve and its branches. The inferior hemorrhoidal nerve can arise higher up from the pudendal nerve or separately from the sacral plexus. (Adapted from Abouleish E. Pain control in obstetrics. Philadelphia: JB Lippincott Co, 1977.) Paracervical Block Paracervical block (PCB) anesthesia may be used when the active phase of labor begins, and it can be employed until approximately 8 cm of dilation has been achieved. Although formerly popular, this block has fallen into relative disuse since the description of bradycardia after PCB and its proven association with fetal acidosis. The PCB is useful when anesthesia personnel are unavailable and parenteral narcotics are inadequate. The PCB relieves the pain associated with uterine contractions, but it is not effective for pain associated with distention of the pelvic floor. The two drugs of choice are chloroprocaine and lidocaine in 1% concentrations. Typically, 6 mL of drug is administered superficially, just under the vaginal mucosa, at the 4- and 8-o'clock positions ( Fig. 3.4). In this way, bradycardia, which occurs in 10% to 30% of cases, is less likely to appear. The landmark study by Baxi and colleagues ( 8 ), using a transcutaneous oxygen electrode attached to the fetal scalp, demonstrated that bradycardia is related to decreasing fetal oxygenation, which becomes marginal approximately 10 minutes after injection. This research has been corroborated by the study of isolated human uterine artery segments and by work in animals, indicating that direct uterine artery vasoconstriction and uterine hypertonus in response to the injection of a local anesthetic drug diminish uterine blood flow and fetal oxygenation.

FIG. 3.4. Technique of paracervical block. Notice the position of the hand and fingers in relation to the cervix and fetal head and the shallow depth of the needle insertion. No undue pressure is applied at the vaginal fornix by the fingers or needle guide. (Adapted from Abouleish E. Pain control in obstetrics. Philadelphia: JB Lippincott Co, 1977.) Lumbar Epidural Analgesia Standard Technique Lumbar epidural analgesia was first performed in 1884 by Corning, who recognized that analgesia could still occur when attempted spinal analgesia failed. In 1921, Pages applied the technique to surgery. Obstetric applications were made by Graffagnino and Seyler in 1935. The most commonly used

anesthetic agents for lumbar epidural analgesia for labor are bupivacaine 0.0625% to 0.25% and ropivacaine 0.1% to 0.2%. The technique of lumbar epidural analgesia involves the insertion of a 17- or 18-gauge hollow-bore needle through the ligamentum flavum into the epidural space at the L4-5, L3-4, or L2-3 interspace. Although the hanging-drop technique for identification of the epidural space is generally effective, most physicians prefer the loss-of-resistance technique as the one that affords the least risk of penetration of the dura. Use of air or saline may be used to identify the epidural space. A 20-gauge catheter is passed through the epidural needle for a distance of 3 to 5 cm within the epidural space. This catheter is securely taped in place and serves as an avenue for intermittent or continuous infusion of local anesthetic agents or opioids. Combined Spinal–Epidural Analgesia The combined spinal–epidural (CSE) technique adds a subarachnoid injection of an opioid with or without a small dose of local anesthetic in a needle-through-needle technique to provide a faster onset with a smaller dose of medication than is possible using epidural medications alone ( 9 ). The epidural catheter is still available for additional analgesia. A prospective, double-blinded, randomized study compared CSE analgesia with standard epidural analgesia in spontaneously laboring nulliparous parturients. There was no difference in the rate of progress of labor, the amount of epidural local anesthetic required, and the incidence of instrumental deliveries between the CSE and the standard epidural analgesia groups. However, analgesia was more complete and there was higher patient satisfaction in the CSE group than in the epidural analgesia group. Patient-controlled Epidural Analgesia Patient-controlled epidural analgesia (PCEA) is a technique by which the patient self-administers on-demand doses of an analgesic mixture via an epidural catheter, whenever she perceives discomfort. To avoid overdosage, a lockout period follows each self-administration. This technique is associated with a decreased use of local anesthetic solution and less demand on staff time compared to continuous epidural infusion (CEI). Most studies have found patients will self-administer less local anesthetic solution than continuous infusions provide and that anesthesia workforce needs are reduced by about 40% ( 10 ). Quality of analgesia, complications, and amount of motor block are similar between the two techniques. Epidural Medications in Labor Epidural injection of opioids alone has been shown to be of limited value for the relief of labor pain. High doses of morphine (7.5 mg) have provided satisfactory analgesia, but only during the first stage of labor. Because of its slow onset of action (often 1 hour or more) and high incidence of side effects (nausea and pruritus), morphine is not a satisfactory agent for this use. Fentanyl and sufentanil used alone can provide analgesia for early labor, but require high doses—100 µg fentanyl or 30 µg sufentanil. In contrast, spinal injection of opioids alone provides excellent, although time-limited analgesia for labor in small doses—10 to 25 µg fentanyl or 5 to 10 µg sufentanil. Fortunately, the addition of opioids to dilute concentrations of epidural local anesthetics has been proven to be quite effective in the relief of labor pain. The combination is a rational one because local anesthetic solutions relieve somatic pain preferentially, whereas opioids are more effective in relieving visceral pain. By combining a lipid-soluble opioid such as fentanyl or sufentanil to bupivacaine or ropivacaine, the concentration of local anesthetic can be dramatically decreased and motor block can be minimized. The addition of fentanyl or sufentanil approximately doubles the analgesic efficacy of any concentration of bupivacaine or ropivacaine, while shortening the time to complete analgesia. Opioid and Chloroprocaine Mixtures. Mixtures of morphine with chloroprocaine have not been shown to be useful, as chloroprocaine appears to antagonize the analgesic effects of the opioid, while increasing its side effects, particularly nausea and pruritus. The mechanism of this unfavorable interaction is unknown. Chloroprocaine may also prolong the onset of morphine analgesia and decrease the effectiveness of bupivacaine if used before these agents. The most satisfactory analgesia appears to be produced when opioids are combined with amide local anesthetics. The beneficial effects of epidural opioids in labor appear to be the following: reduction in motor block, allowing improved mobility of the patient reduction in shivering decreased incidence of hypotension use of lower doses of local anesthetic agents greater maternal satisfaction with the analgesia that they provide. Effects of Epidural Analgesia on Uterine Blood Flow Studies investigating changes in intervillous blood flow and mean arterial pressure with lumbar epidural analgesia in adequately preloaded patients have demonstrated only a negligible reduction in these parameters with the onset of effective analgesia. Well-hydrated patients with preeclampsia have experienced improvement in intervillous blood flow along with a slight decrease in blood pressure. Advantages and Disadvantages of Lumbar Epidural Analgesia There are three principal advantages of lumbar epidural analgesia: the parturient remains awake and cooperative. the incidence of complications is very low when the technique is used correctly. once an epidural catheter is in place, it can be used to provide analgesia or anesthesia for a vaginal or cesarean delivery. The disadvantages of lumbar epidural analgesia include: the possibility of poor perineal analgesia the presence of “hot spots”, where analgesia is insufficient delayed onset of action technical difficulty intravascular injection accidental dural puncture hypotension. Technical failure occurs in approximately 4% of cases. Indications and Contraindications for Lumbar Epidural Analgesia Indications for lumbar epidural analgesia include pain in labor, management of the patient with preeclampsia who does not have a coagulation abnormality, management of labor in patients with certain cardiac lesions, and management of breech delivery. A joint statement by the American College of Obstetricians and Gynecologists (ACOG) and the American Society of Anesthesiologists (ASA) ( 11 ) notes that: “Labor results in severe pain for many women. There is no other circumstance in which it is considered acceptable for a person to experience untreated severe pain, amenable to safe intervention, while under a physician's care. In the absence of a medical contraindication, maternal request is a sufficient medical indication for pain relief during labor. Of the various pharmacologic methods of pain relief used in labor and delivery, regional analgesia techniques—spinal, epidural and combined spinal epidural (CSE) are the most flexible, effective, and least depressing to the CNS, allowing for an alert, participating mother and an alert neonate.” There are absolute and relative contraindications to the induction of lumbar epidural analgesia. Absolute contraindications include the following: patient refusal, hemodynamic instability, infection at the anticipated site of puncture, and absence of resuscitation equipment. Relative contraindications may include fever, preexisting CNS disease, hypovolemia, hypotension, lack of experience by the anesthetist, and blood coagulation defects. Although an arbitrary platelet count of 100,000 per mm 3 has been advocated as the lower limit for safe lumbar epidural analgesia, successful blocks without epidural bleeding complications have been obtained with platelet counts as low as 50,000 per mm 3 . In one report, Beilin and colleagues ( 12 ) described a study of 80 women who presented for labor and delivery and had platelet counts less than 100,000 per mm 3 during the peripartum period. Of these 80, 30 were given an epidural anesthetic. The range of platelet counts was 69,000 to 98,000 per mm 3 . No patient had any documented neurologic complication. These authors concluded that regional anesthesia should not necessarily be withheld when the platelet count is less than 100,000 per mm 3 . However, given that the risk of epidural hematoma is extremely rare, a study this small would be unlikely to reveal a problem. The underlying cause of the thrombocytopenia is also important and must be considered along with any absolute number. For example, the patient with idiopathic thrombocytopenic purpura or gestational thrombocytopenia is much less likely to bleed at a low platelet count than the patient with HELLP (hemolysis, elevated liver enzymes, low platelet count) syndrome. The best indicator of potential bleeding is a patient history of bruises, contusions, petechiae, bleeding from the gums, and so on. Subarachnoid Analgesia Subarachnoid or spinal analgesia for labor has become increasingly popular. The major advantages of spinal analgesia include (a) use of a very low dose of local anesthetic or narcotic analgesic drug and (b) the excellent analgesia provided. Onset of action is rapid, and uterine activity is not affected. The disadvantages include (a) the possibility of postdural puncture headache (PDPH) which is increasingly rare with the use of pencil-point needles and (b) the time-limited nature of a single-shot technique. For this reason spinal analgesia is commonly combined with an epidural catheter as a CSE technique. Indications and contraindications are similar to epidural analgesia. Morphine and fentanyl can provided analgesia during labor when administered intrathecally in small doses. Labor analgesia usually lasts for 4 to 8 hours and is not accompanied by motor block. The major disadvantages include pruritus and nausea. Pruritus can be antagonized with small doses of intravenous naloxone or oral naltrexone. PDPH can be minimized through the use of 22- to 27-gauge Whitacre, Sprotte, or other pencil-point needles. A 28-gauge microcatheter can be passed through 22-gauge spinal needles, permitting the use of continuous spinal analgesia in labor and surgery. These catheters were withdrawn from the market because of associated cauda equina syndrome, probably related to local anesthetic neurotoxicity. Complications of Regional Block Analgesia Hypotension Hypotension is common, occurring in 10% to 20% of patients undergoing epidural analgesia for labor and 50% to 80% for cesarean delivery. It often occurs despite left uterine displacement and administration of an adequate vascular preload. Treatment should consist of the following steps: 1. 2. 3. 4.

Ensure or verify left uterine displacement. Increase intravenous fluid infusion to the maximal available rate. Administer oxygen by facemask. If hypotension does not immediately resolve, administer ephedrine intravenously in 5- to 10-mg increments until hypotension resolves.

Phenylephrine, an a-agonist, has been avoided in the past because of concerns about compromising uterine blood flow, but more recent work has shown no adverse clinical effects and higher fetal pH values than ephedrine alone ( 13 ).

Postdural Puncture Headache When a large-bore epidural needle (e.g., 18-gauge Tuohy) penetrates the dura and arachnoid membranes, the incidence of PDPH is greater than 50%. Other factors governing the incidence of PDPH include the number of times the dura has been punctured, the direction of the bevel, and the type of needle used. PDPH occurs because a decrease in cerebrospinal fluid (CSF) volume causes compensatory cerebral vasodilation and traction on the pain-sensitive blood vessels and meninges. Assumption of the erect position increases traction on these structures and aggravates the pain. Therefore, the main diagnostic criterion of PDPH is that it is postural. Ocular and auditory symptoms, such as vertigo, ataxia, and sixth cranial nerve palsy may be associated with PDPH. Many different treatment regimens have been employed for this condition, including administration of saline through the epidural catheter, the use of abdominal binders, the administration of intravenous or oral caffeine (as a cerebral vasoconstrictor), bed rest, analgesics, and epidural blood patches. The most effective of these is the epidural blood patch. Evidence suggests that prophylactic epidural blood patch can substantially reduce the incidence and severity of PDPH. In a study by Lowinwirt and associates ( 14 ), patients with accidental dural puncture with 16- or 17-gauge needles were randomized and allocated to two treatment groups. In one group, patients received 15 to 20 mL of autologous blood through the indwelling epidural catheter at least 5 hours following the last dose of local anesthetic and just before catheter removal. Patients in the control group were managed conservatively with intravenous hydration, bed rest, theophylline, or caffeine. Eighty-three percent of patients receiving the prophylactic epidural blood patch avoided PDPH. Only 4% of patients treated conservatively avoided PDPH. Use of a prophylactic epidural blood patch was not associated with any complications. Chronic Back Pain Low-back pain is a common complaint in the postpartum period. A controversy exists as to the role that epidural analgesia for labor and delivery might play in the subsequent development of low-back pain. Whereas several retrospective studies have demonstrated an association, other prospective studies have shown no increased risk of acute low-back pain after epidural use. A study by Loughnan and colleagues ( 15 ) examined the risk of low-back pain 6 months after delivery in patients who received epidural analgesia, as compared with those who did not. This prospective follow-up study showed no difference in the prevalence of low-back pain at 6 months after delivery. Therefore, it seems unlikely that epidural analgesia makes any significant contribution to the 19% to 33% prevalence of low-back pain among postpartum women. Neurologic Complications Neurologic complications of epidural and spinal analgesia are rare, occurring in about 1 per 10,000 blocks. Many postpartum neurologic sequelae are related to intraoperative positioning problems. An example is foot drop associated with pressure on the lateral popliteal nerve and caused by an improperly placed stirrup. In the lithotomy position, pressure applied on the femoral cutaneous nerve by the inguinal ligament may cause pain and numbness in the lateral thigh. Pain and numbness in the distribution of the sciatic nerve may result from forceps delivery or passage of the baby's head through the pelvis. Spinal nerve root neuropathy may be caused by traumatic insertion of a spinal needle or an epidural needle or catheter. In this case, pain and paresthesias along the distribution of the nerve are perceived immediately, but they tend to disappear when the needle or catheter is removed. Rarely, symptoms may appear as long as 2 days after the procedure. Recovery usually occurs in 1 to 2 weeks, but injury can be permanent. Accidental injection of an irritant solution (e.g., thiopental) or a prep solution into the CSF may produce adhesive arachnoiditis, which can cause permanent loss of spinal cord function. Epidural abscess usually is caused by hematogenous spread and not by injection of contaminated anesthetic solutions. Epidural hematoma is a serious complication that, although rare, may occur in conjunction with coagulopathy. A hematoma should be suspected if recovery from the block is slow or absent or if neurologic function worsens after a period of initial recovery. The primary symptoms are pain and weakness, which may progress rapidly to paralysis. Early surgical drainage provides the only chance for recovery of neurologic function. Effects of Epidural Analgesia on Progress of Labor Whether use of regional analgesic techniques is associated with an increased risk of cesarean delivery remains controversial, but that concern is mainly based on results from small retrospective studies. There is definitely an association between use of epidural analgesia for labor and cesarean delivery, but the problem is that nulliparous women with longer and more painful labors are the patients more likely to choose epidural analgesia, and they are also those who are at higher risk to have a cesarean delivery regardless of their choice of labor analgesia. The presence of severe pain during early labor signals an increased risk for prolonged labor and operative delivery ( 16 ). Also, regional analgesia is often recommended to women in whom operative or instrumental delivery is thought to be likely. Designing a study to remove that bias or association has proven difficult, both for the ethical reason you cannot refuse a treatment available to a patient if requested (i.e., if a patient is randomized to receive “no epidural” but she later requests it because of inadequate pain relief she must be allowed to cross over) and because there is no other form of pain relief to offer that provides equivalent analgesia (i.e., a “control group”). For this reason, many studies done prior to the last decade were methodologically flawed. More recent studies have attempted to control for the fact that women already at increased risk for an operative delivery are more likely to choose epidural analgesia, and have not found regional analgesia to be associated with cesarean delivery ( 17 ). There are different methodological ways to accomplish this. Several retrospective, population-based studies have found that the introduction of an epidural analgesia service or the increased use of epidural analgesia did not increase the cesarean delivery rate. In a “natural” experiment, a military hospital went from a 1% to 84% epidural analgesia rate in one year while other conditions remained unchanged ( 18 ). A review of singleton, nulliparous, term patients in spontaneous labor before and after the change found no differences in rates of cesarean delivery overall and for dystocia, no change in instrumental delivery rates, no change in duration of first and active stages of labor, but an increase in the second stage of labor by about 25 minutes. At the National Maternity Hospital in Dublin, the epidural rate increased from 10% to 57% over 3 years with no effect on cesarean delivery rates ( 19 ). These clinicians suggested use of active management of labor and use of oxytocin in the second stage of labor to overcome increases in instrumental vaginal delivery rates. Other studies have used generous doses of narcotics in the control group to prevent crossover to the epidural analgesia group. A review of 802 nulliparous patients randomized to epidural analgesia or intramuscular meperidine using standardized labor management found similar cesarean delivery rates in an intent-to-treat analysis ( 20 ). The spontaneous vaginal delivery rate was also similar between groups. A randomized trial of 715 women who received epidural analgesia or patient-controlled meperidine found no difference in cesarean delivery rates and lower pain scores in the epidural group ( 4 ). Pain relief is a worthy goal unto itself! Obstetric management must also influence the cesarean delivery rate. Even randomized trials cannot blind the obstetrician to the type of analgesia being used, and the obstetrician makes the decision regarding the need for cesarean delivery. If epidural analgesia has an influence on the risk of cesarean delivery, then those obstetricians with a higher use should have a higher cesarean delivery rate. In contrast, a review of 110 obstetricians in a single hospital practice found no relationship between frequency of epidural analgesia use and rate of cesarean section for dystocia across practitioners (R 2 = 0.019) ( 21 ). They concluded that after accounting for a number of known patient risk factors, obstetric practice style appears to be a major determinant of rates of cesarean delivery. As noted earlier, the intensity of labor pain may be predictive of an increased risk of cesarean delivery for dystocia. A secondary analysis of women randomized to receive patient-controlled intravenous meperidine analgesia for labor found that those who required higher doses of narcotic had higher pain scores initially, longer labors (9 vs. 5 hours), and more cesarean deliveries for dystocia (14% vs. 1.4%) ( 22 ). Thus, it would appear that there are many variables with the potential to affect the risk of cesarean delivery besides choice of analgesia during labor. These would include patient-related factors such as parity, induction, level of pain, labor pattern, and oxytocin use, as well as obstetrician-related factors such as active or passive management. An editorial suggested that “Perhaps the time has now come to look at the problem in a different way and to ask why labor is prolonged and spontaneous delivery less likely in some units than in others. The realization that labor with regional analgesia is not the same as that without and that management must be modified will hopefully result in improvements in outcome” ( 23 ). Chestnut ( 24 ) has made some astute comments on the current state of affairs regarding anesthesia in labor: 1. Use of dilute solutions of local anesthetics in labor may be less likely to have an effect on progress of labor. 2. Administration of epidural analgesia should be delayed until labor is well established; however, it is unnecessary to wait until an arbitrary cervical dilation, such as 5 cm. 3. Studies on patients attempting vaginal birth after cesarean delivery have not shown an increase in cesarean delivery rates when epidural analgesia is used, and good analgesia may encourage some women to attempt vaginal delivery. 4. Use of epidural analgesia during labor may decrease the use of general anesthesia with its attendant risks if emergency cesarean delivery is required. 5. Maternal administration of high doses of narcotics may result in substantial neonatal effects such as respiratory depression ( 25 ). 6. Maternal–fetal factors and obstetric management, not epidural analgesia, are the most important determinants of the cesarean delivery rate.

OTHER METHODS OF PAIN RELIEF Prepared Childbirth Prepared childbirth techniques are based on the belief that pain can be eliminated or reduced by conditioned reflexes of controlled relaxation and that education about the birth process can diminish the pain resulting from fear of the unknown. Parturients and significant others are offered a series of five to ten weekly lectures and are educated about pregnancy, labor, and the delivery process. The parturient is taught how to relax and engages in exercises to strengthen her back and abdominal muscles. She also learns specific breathing patterns to be used while she experiences the discomfort of uterine contractions. All parturients should have access to emotional support, whether by her husband, a family member, a birth attendant (doula), or professional hospital staff. Insurance companies are beginning to pay for doulas as the literature has supported the beneficial effect on decreasing interventions ( 26 ). Effective courses also teach pregnant women that additional methods of pain relief are available and that these do not cause harm to the fetus. The pregnant woman should be advised that to ask for these other methods does not imply that she is a failure. Hypnosis Hypnosis is a state of altered consciousness that requires deep concentration. The patient is not asleep, but she initiates a trance as labor begins and continues it until delivery is completed. The patient must undergo a time-consuming series of training sessions with a hypnotist, and this technique is not always successful ( 27 ). Acupuncture Acupuncture has been used to help control labor pain in China and the Far East for many years. Since the early 1970s, mixed reports of its efficacy have been published in the West. Some studies indicate that acupuncture can significantly lower pain scores and may decrease the duration of the first stage of labor. Although enthusiasm for the technique varies in Western countries, the Chinese continue to report 99% rates of excellent and good success. When acupuncture is used for

cesarean section in conjunction with intravenous meperidine and local anesthetic infiltration, blood pressure, pulse rate, and respirations are stable, and the degree of patient acceptance is reported to be high. Biofeedback Biofeedback is provided by a portable electromyographic device through an audible sound and visual monitor. Electrodes placed over the maternal abdomen monitor tension of the abdominal musculature. This technique may be helpful during the first stage of labor and may reduce its duration. Transcutaneous Electrical Nerve Stimulation Transcutaneous electrical nerve stimulation (TENS) analgesia is based on the observation that application of a mild electric current to the skin can result in reduction of pain. Activation based on the gate theory and release of enkephalins are possible modes of action. Studies evaluating the effectiveness of TENS suggested that, although the method does no harm, it probably does little good and should not be advocated for widespread use for labor analgesia ( 28 ). Intracutaneous Nerve Stimulation Trolle and colleagues ( 29 ) evaluated the analgesic effect of intradermal sterile water blocks in women complaining of severe low-back pain during labor. Saline solution was used as a control. Sterile water or saline (0.1 mL) was injected at four different spots in the low-back area, approximately corresponding to the borders of the sacrum. Eighty-nine percent of women in the sterile water group reported an analgesic effect, compared with 45% in the saline group. However, meperidine use in the two groups was similar, as were the rates for oxytocin use and dystocia. The cesarean section rate in the sterile water group was significantly lower due to more cephalopelvic disproportion and malposition of the occiput in the saline group. This technique is free of adverse effects and enjoys a high degree of patient acceptance although the mechanism is unclear.

ANESTHESIA FOR CESAREAN DELIVERY There are three anesthetic choices for cesarean delivery. Selection of one over the others depends on the patient's desires, medical status, and the urgency of the operation. Regional anesthesia is strongly preferred in the United States ( Table 3.4).

TABLE 3.4. Types of cesarean anesthesia provided by size of hospital in three time periods

Epidural Anesthesia Epidural anesthesia accounts for approximately 40% of anesthetics used for cesarean section. It offers the advantages of unlimited duration, minimizing the risks of airway management, and providing a route for postoperative pain management. To carry out a cesarean delivery, a sensory dermatome level of at least T-4 is required. Anesthesia to this level eliminates proprioception from the respiratory muscles of the chest wall, and the parturient may experience a subjective sensation of dyspnea. Reassurance will usually allay this fear. The patient should be placed on the operating table with the uterus displaced laterally through elevation of the right hip or by tilting of the operating table, to prevent aortocaval compression. A vascular preload of 1000 to 1500 mL of a non–glucose-containing crystalloid solution should be administered before dosing of the epidural needle or catheter with the anesthetizing solution. A surgical concentration of a local anesthetic (often 2% lidocaine) with or without added opioid is administered through the catheter in 3- to 5-mL increments. The patient should be given oxygen by nasal cannula or by mask. One study failed to reveal any differences in the clinical condition of neonates, as assessed by Apgar scores and blood gas analyses, when oxygen administration by these two modalities was compared. Epidural opioids have been useful for relieving pain of visceral origin, which affects as many as one-third of the women who have cesarean sections under epidural anesthesia. Visceral pain occurs primarily during bladder retraction, exteriorization of the uterus, and suturing of the peritoneum. The addition of fentanyl or sufentanil to the local anesthetic reduces the time of onset of analgesia, decreases the incidence of nausea, and increases the quality of analgesia without depressing the neurobehavioral status of the newborn. Studies of fentanyl concentrations in neonates demonstrate that fentanyl crosses the placenta. Even high maternal doses (e.g., 100 µg) of fentanyl yield safe levels in the newborn. Morphine 2 to 4 mg is often administered through the epidural catheter after delivery to provide 12 to 24 hours of postoperative analgesia, although the incidence of associated pruritus and nausea can be high. Use of epidural morphine may reactivate herpetic labialis (herpes simplex virus-1) infections ( 30 ). There are several advantages to using epidural analgesia for cesarean section delivery: if an epidural catheter is already in place, it can be used expeditiously for the cesarean section delivery. maternal hypotension may be less pronounced and slower in onset with epidural than with spinal anesthesia. headache is usually avoided, unless the patient sustains an accidental dural puncture. the length of anesthesia is controllable in case surgery is prolonged. the technique is adaptable for postoperative pain relief. The disadvantages include: the slower onset of analgesia the requirement for a larger amount of anesthetic solution with its attendant increased risk of systemic toxicity the lower success rate than that experienced with subarachnoid block. An unexpectedly high level of anesthetic block may be achieved with epidural or spinal anesthesia and, as a result, this should be monitored. The likelihood of inadvertent spinal anesthesia while attempting epidural block can be minimized through gentle aspiration of the catheter combined with using a test dose of sufficiently small volume that it is unlikely to produce a high block. A total spinal block generally occurs within 90 seconds after injection, but it may be delayed for as long as 20 minutes. Dyspnea, hypotension, unconsciousness, and apnea are signs and symptoms of total spinal block. Treatment includes ventilation through an endotracheal tube and, if needed, circulatory support. Subarachnoid or Spinal Anesthesia A subarachnoid block provides excellent anesthesia for cesarean section delivery. Over 50% of cesarean deliveries are performed using this technique (see Table 3.4). Prehydration is administered with 1500 to 2000 mL of a non–glucose-containing crystalloid solution. A 22- to 27-gauge pencil-point spinal needle is inserted into the subarachnoid space, which is identified by the characteristic feel of the needle penetrating the dura and observing CSF in the needle. Bupivacaine 0.75%, 10 to 12 mg, with dextrose, is most commonly used. Analgesia of shorter duration is obtained with the use of lidocaine 5% in 7.5% dextrose. Opioids may also be administered intrathecally to improve quality of analgesia, decrease nausea, and improve cardiovascular stability by allowing a lower dose of local anesthetic. Spinal (as well as epidural) narcotics are associated with a high incidence of pruritus, as well as the rare but real potential for delayed respiratory depression. Fentanyl may be administered to improve intraoperative analgesia while the addition of morphine can provide postoperative analgesia that may last for 18 to 24 hours.

Contraindications to spinal anesthesia are the same as for epidural anesthesia and include patient refusal, septicemia, infection of the puncture site, acute or chronic hypovolemia, and abnormal clotting parameters. Spinal anesthesia is usually avoided in pregnant women with acute CNS disease. The most common complication of spinal anesthesia is hypotension. This should be treated promptly with fluid administration and intravenous ephedrine (5- to 10-mg bolus). Oxygen should be given, and the parturient's oxygen saturation should be monitored with a pulse oximeter. In the event of a high spinal block which compromises ventilation or airway control, cricoid pressure should be applied and endotracheal intubation performed to prevent aspiration of gastric contents. With the increasing use of spinal and epidural narcotics, pruritus is becoming a commonplace adverse effect. Its incidence approaches 60% when spinal or epidural morphine is employed. The cause seems to be related to stimulation of opioid receptors rather than to release of histamine. Naloxone can be used to control pruritus, but the dose must be titrated carefully to avoid antagonism of analgesia. Nalbuphine may also be used and is less likely to antagonize analgesia. Approximately 23% of women experience shivering during normal labor and delivery, and the rate increases to approximately 68% with epidural analgesia. Shivering can be diminished or abolished through epidural injection of opioids (e.g., fentanyl 100 µg) or small intravenous doses of meperidine (e.g., 12.5 mg). General Anesthesia General anesthesia is used for cesarean delivery when the patient refuses regional analgesia or has a contraindication to regional analgesia, or when a need exists for rapid delivery because of fetal distress, cord prolapse, shoulder dystocia, or maternal hemorrhage. The American College of Obstetricians and Gynecologists in the fifth edition of Guidelines for Perinatal Care, cites the risk factors for failed intubation and urges obstetricians to be alert to the presence of the factors that place parturients at increased risk for complications from emergency general anesthesia. Among these are marked obesity, severe facial and neck edema, extremely short stature, short neck, difficulty opening the mouth, a small mandible, protuberant teeth, arthritis of the neck, anatomic abnormalities of the face or mouth, a large thyroid gland, asthma, serious medical or obstetric complications, and a history of problems with anesthetics. If any of these factors is identified, a member of the anesthesia team should be consulted to prepare for the unexpected need to induce general anesthesia. If the anesthesiologist has concerns about his or her ability to intubate the patient, early placement of a regional anesthetic should be planned or arrangements for an awake intubation should be made. Pneumonitis resulting from aspiration of gastric contents has long been feared as a complication of general anesthesia for obstetrics, but is extremely rare. One review compared the incidence of aspiration in obstetric and gynecological patients ( 31 ). The incidence of clinically significant aspiration was 0.11% in women undergoing cesarean delivery compared to 0.01% in gynecology inpatients. No patient died, but morbidity was significant. The prudent anesthetist administers a nonparticulate oral antacid, such as sodium citrate, given prophylactically to increase the gastric pH. If time allows, an H 2 -blocker (e.g., ranitidine 50 mg intravenously) should be administered. Intravenous metoclopramide, 10 mg, hastens gastric emptying, increases gastroesophageal sphincter tone, and may decrease nausea. Before induction of anesthesia, the patient should be preoxygenated with 100% oxygen by mask for at least 3 minutes. Induction is commonly carried out using thiopental (3–4 mg per kg i.v.). Propofol (Diprivan) has also been used. Propofol is associated with a blunted hypertensive response to endotracheal intubation and has yielded similar and satisfactory Apgar scores, neurologic and adaptive capacity scores, and umbilical cord blood gas analyses. If propofol infusion is used for maintenance of anesthesia for a prolonged time before delivery, neonatal blood levels are high, and neurologic and adaptive capacity scores may be impaired. If the patient is hemodynamically unstable, ketamine 1 mg per kg or etomidate 0.3 mg per kg may be used. Intubation is facilitated by use of succinylcholine. Cricoid pressure is maintained during induction of anesthesia until the endotracheal tube is in place, the cuff has been inflated, respirations have been auscultated, and end-tidal carbon dioxide has been seen. After successful intubation, a mixture of equal parts of nitrous oxide and oxygen may be administered, and a low dose of an inhalational agent, such as desflurane 3%, sevoflurane 1% or isoflurane 0.75%, is administered to optimize maternal analgesia and amnesia. These low concentrations have minimal effects on uterine contractility and are not associated with postpartum hemorrhage. After delivery of the infant, the nitrous oxide concentration may be increased to 70%, and narcotics may be given intravenously to supplement the anesthesia. Midazolam may be used to decrease the risk of maternal recall. The advantages of general anesthesia include: reliability of the technique rapidity of induction of anesthesia avoidance of sympathetic blockade and hypotension. The disadvantages include: the risks of maternal aspiration of gastric contents failed intubation maternal awareness hypertension during manipulation of the larynx. If the cords are poorly visualized during laryngoscopy, no more than three attempts at endotracheal intubation should be made before beginning a failed intubation drill ( 32 ). The initial maneuver in the failed intubation drill depends on the obstetric indication for cesarean section. If the operation is not emergent, the patient should be awakened and an epidural or spinal block performed. If a regional anesthetic cannot be accomplished, an awake fiberoptic intubation should be considered. In an obstetric emergency where surgery must proceed, the patient must be ventilated with bag and mask or laryngeal mask airway, and anesthesia may be maintained with inhalational or intravenous agents throughout the remainder of the cesarean section. The continuation of cricoid pressure is important to reduce the maternal risk of aspiration. If it should prove impossible to ventilate the patient with bag and mask, an emergency maneuver such as cricothyroidotomy must be performed. Use of an esophageal gastric tube airway (Combitube) or laryngeal mask may enable adequate ventilation. Inability to intubate has been estimated to occur seven times more commonly in the obstetric patient than in the general operating room, and continues to contribute significantly to anesthetic causes of maternal mortality ( 33 ). Anticipation of a difficult intubation allows the anesthesia team to be prepared to avoid general anesthesia or plan an awake intubation. Analgesia after Cesarean Section Considerable advances have been made in the management of pain after cesarean section. The availability of spinal and epidural narcotics has enabled the anesthesia team to provide the postsurgical patient with effective, long-term analgesia. Morphine is the most commonly used neuraxial opioid because of its long duration and lack of motor block compared to local anesthetic infusions. The most feared complication is respiratory depression. The rate of analgesia-related respiratory depression is approximately 0.09%. Because this complication is rare, patients receiving postoperative neuraxial opioid analgesia may be safely nursed on the general ward, if the nurses are appropriately educated in monitoring the degree of somnolence and the respiratory rates of their patients. Patients who did not receive regional anesthesia may receive intravenous PCA. Combining narcotics and nonsteroidal antiinflammatory medications such as ketorolac or ibuprofen improves the quality of analgesia and allows reduced doses of narcotics. Naloxone should be readily available to antagonize respiratory depression.

SUMMARY POINTS Pain management is an important part of modern obstetric care. Most women will request some form of analgesia during childbirth. Parenteral narcotics for labor analgesia may be administered by intermittent injection or patient-controlled intravenous infusion. There are advantages and disadvantages of all opioids. High systemic blood levels of local anesthetic caused by intravascular injection or excessive absorption may lead to convulsions and cardiac arrest. Resuscitation equipment must be immediately available whenever regional blocks are used. Modern techniques of regional analgesia for labor (dilute concentrations of epidural local anesthetics, combined spinal–epidural analgesia and patient-controlled epidural infusions) emphasize pain relief with minimal motor block. Studies indicate these techniques do not impact progress of labor. Although modern anesthetic care for cesarean delivery is extremely safe (anesthesia-related maternal mortality = 1.1 per million live births), general anesthesia complications are more common than regional anesthetic complications because of difficulties with airway management.

REFERENCES 1. Rosen MA, Hughes SC, Levinson G. Regional anesthesia for labor and delivery. In: Hughes SC, Levinson G, Rosen MA, eds. Anesthesia for obstetrics, fourth ed. Philadelphia: Lippincott Williams & Wilkins, 2002:123–148. 2.

Hawkins JL, et al. ASA Practice Guidelines for Obstetrical Anesthesia. Anesthesiology 1999;90:600–611.

3.

Olofsson CH, Ekblom A, Edman-Ordeberg G, et al. Lack of analgesic effect systemically administered morphine or pethidine in labour. Br J Obstet Gynaecol 1996;103:968–972.

4.

Sharma SK, Sidawi JE, Ramin SM, et al. A randomized trial of epidural versus patient-controlled meperidine analgesia during labor. Anesthesiology 1997;87:487–494.

5.

Rosenblatt WH, Cioffi AM, Sinatra R, et al. Metoclopramide: an analgesic adjunct to patient-controlled analgesia. Anesth Analg 1991;73:553–558.

6.

Maroof M, Hakin S, Khan RM, et al. Low-dose ketamine infusion if effective in relieving labor pain. Anesth Analg 1998;86:S380(abst).

7.

Butterworth JF IV, Walker FO, Lysak SZ. Pregnancy increases median nerve susceptibility to lidocaine. Anesthesiology 1990;72:962–965.

7a.

Weinberg GL. Current concepts in resuscitation of patients with local anesthetic cardiac toxicity. Reg Anes Pain Med 2002;27:568–575.

8.

Baxi LV, Petrie RH, James LS. Human fetal oxygenation following paracervical block. Am J Obstet Gynecol 1979;135:1109–1113.

9.

Shennan AH. Effect of low-dose mobile versus traditional epidural techniques on mode of delivery: a randomized controlled trial. Lancet 2001;358:19–23.

10.

Viscomi C, Eisenach JC. Patient-controlled epidural analgesia during labor. Obstet Gynecol 1991;77:348–351.

11.

American College of Obstetricians and Gynecologists. Pain relief during labor. ACOG Committee Opinion No. 231, 2000.

12.

Beilin Y, Zahn J, Comerford M. Safe epidural analgesia in thirty parturients with platelet counts between 69,000 and 98,000 mm-3. Anesth Analg 1997;85:385–391.

13. Lee A, Kee WDN, Gin T. A quantitative, systematic review of randomized controlled trials of ephedrine versus phenylephrine for the management of hypotension during spinal anesthesia for cesarean delivery. Anesth Analg 2002;94:920–926. 14. Lowenwirt I, Cohen S, Zephyr J, et al. Can prophylactic epidural blood patch reduce the incidence and severity of postpartum dural puncture headache in obstetrics? Anesth Analg 1998;86:S378(abst). 15. Loughnan BA, Carli F, Romney M, et al. Epidural analgesia and backache: a randomized controlled comparison with intramuscular meperidine for analgesia during labour. Br J Anaesth 2002;88:466–442. 16.

Wuitchik M, Bakal D, Lipshitz J. The clinical significance of pain and cognitive activity in latent labor. Obstet Gynecol 1989;73:35–42.

17.

Sharma SK, Leveno KJ. Update: epidural analgesia during labor does not increase cesarean births. Curr Anesth Rep 2000;2:18–24.

18.

Zhang J, Yancey MK, Klebanoff MA, et al. Does epidural analgesia prolong labor and increase risk of cesarean delivery? A natural experiment. Am J Obstet Gynecol 2001;185:128–134.

19.

Impey L, MacQuillan K, Robson M. Epidural analgesia need not increase operative delivery rates. Am J Obstet Gynecol 2000;182:358–363.

20.

Loughnan BA, Carli F, Romney M, et al. Randomized controlled comparison of epidural bupivacaine versus pethidine for analgesia in labour. Br J Anaesth 2000;84:715–719.

21.

Segal S, Blatman R, Doble M, et al. The influence of the obstetrician in the relationship between epidural analgesia and cesarean section for dystocia. Anesthesiology 1999;91:90–96.

22.

Alexander JM, Sharma SK, McIntire DD, et al. Intensity of labor pain and cesarean delivery. Anesth Analg 2001;92:1524–1528.

23.

Russell R. The effects of regional analgesia on the progress of labour and delivery. Br J Anaesth 2000;85:709–712.

24.

Chestnut DH. Epidural analgesia and the incidence of cesarean section. Anesthesiology 1997;87:472–476.

25.

Halpern SH, Leighton BL, Ohlsson A, et al. Effect of epidural vs parenteral opioid analgesia on the progress of labor. JAMA 1998;280:2105–2110.

26.

Kennell J, Klaus M, McGrath S, et al. Continuous emotional support during labor in a US hospital: a randomized controlled trial. JAMA 1991;265:2197–2201.

27.

Freeman RM, Macaulay AJ, Eve L, et al. Randomized trial of self hypnosis for analgesia in labour. Br Med J 1986;292:657–658.

28.

Carroll D, Tramer M, McQuay H, et al. Transcutaneous electrical nerve stimulation in labour pain: a systematic review. Br J Obstet Gynaecol 1997;104:169–175.

29.

Trolle B, Moller M, Kronborg H, et al. The effect of sterile water blocks on low back labor pain. Am J Obstet Gynecol 1991;164:1277–1281.

30.

Crone LA, Conly JM, Clark KM, et al. Recurrent herpes simplex virus labialis and the use of epidural morphine in obstetric patients. Anesth Analg 1988;67:318–323.

31.

Soreide E, Bjornestad E, Steen PA. An audit of perioperative aspiration pneumonitis in gynaecological and obstetric patients. Acta Anaesth Scand 1996;40:14–19.

32.

Caplan RA, Benumof JL, Berry FA, et al. Practice guidelines for management of the difficult airway. Anesthesiology 1993;78:597–602.

33.

Hawkins JL, Koonin LM, Palmer SK, et al. Anesthesia-related deaths during obstetric delivery in the United States, 1979–1990. Anesthesiology 1997;86:277–284.

Chapter 4 Early Pregnancy Loss Danforth’s Obstetrics and Gynecology

Chapter 4 D. Ware Branch and James R. Scott

Early Pregnancy Loss

EPIDEMIOLOGY ETIOLOGY Embryonic Factors Parental Factors PATHOLOGY CLINICAL FEATURES AND TREATMENT Threatened Miscarriage Inevitable and Incomplete Miscarriage Complete Miscarriage Missed Miscarriage Septic Miscarriage RECURRENT MISCARRIAGE Known and Suspected Causes of Recurrent Miscarriage Recommendations for Recurrent Miscarriage SUMMARY POINTS SUGGESTED READINGS

The term spontaneous abortion, which has a negative connotation to many patients, is gradually being replaced by the word miscarriage. Both terms originally defined pregnancy losses prior to 20 weeks gestation, but they are more commonly used by physicians to describe first-trimester losses. These arbitrary time limits have become less useful with advances in developmental biology and diagnostic sonography. The preembryonic period is defined as conception through the first 5 weeks of pregnancy from the first day of the last menstrual period. The embryonic period encompasses 6 to 9 weeks gestation, and the fetal period is from 10 weeks until delivery.

EPIDEMIOLOGY Human reproduction is relatively inefficient, and miscarriage is the most common complication of pregnancy, with an overall incidence of approximately 15% among clinically recognized pregnancies. The rate of miscarriage is, however, heavily dependent upon the past obstetric history, being higher among women with prior miscarriages and lower among women whose past pregnancy(ies) ended in live births. Histologically defective ova found in hysterectomy specimens ( Fig. 4.1) and data on early pregnancies detected with sensitive ß-human chorionic gonadotropin (ß-hCG) assays indicate that the very early and often unrecognized pregnancy loss rate is two to three times higher than that of recognized pregnancy. The prevalence of miscarriage also increases with maternal age from 12% in women younger than 20 years of age to over 50% in women older than 45 years of age ( Fig. 4.2).

FIG. 4.1. Histologic comparison of (A) a morphologically normally implanted human ovum estimated to be about 11 to 12 days of age with (B) an abnormal conceptus, showing a defective trophoblast with pathologically large lacunae and an empty chorionic sac that is destined to abort. (From Hertig AT, Rock J, Adams EC. Am J Anat 1956;98:435, with permission.)

FIG. 4.2. Relation of maternal age to the risk of spontaneous abortion. (Data from Warburtin D, Kline J, Stein Z, et al. Cytogenetic abnormalities in spontaneous abortions of recognized conceptions. In: Porter IH, ed. Perinatal genetics: diagnosis and treatment. New York: Academic Press, 1986:133.)

ETIOLOGY In view of the complicated genetic, hormonal, immunologic, and cellular events requiring precise integration for fertilization, implantation, and embryonic development, it is remarkable that successful pregnancy occurs so often. When early pregnancy loss occurs, it can be due to a number of embryonic and parental factors. Embryonic Factors Most single, sporadic miscarriages are caused by nonrepetitive intrinsic defects in the developing conceptus, such as abnormal germ cells, chromosomal abnormalities in the conceptus, defective implantation, defects in the developing placenta or embryo, accidental injuries to the fetus, and probably other causes as yet unrecognized. Fifty percent of women presenting with spotting or cramping already have a nonviable conceptus by sonogram, and many of these embryos are morphologically abnormal. About one-third of abortus specimens from losses occurring before 9 weeks gestation are anembryonic. Some cases of empty gestational sacs or “blighted ova” actually represent pregnancy failures with subsequent embryonic resorption. The high proportion of abnormal aborted concepti is apparently the result of a selective process that eliminates about 95% of morphologic and cytogenetic errors. The frequency of chromosomally abnormal spontaneously aborted products of conception in the first trimester is approximately 60% and decreases to 7% by the end of week 24 ( Fig. 4.3). The rate of genetic abnormalities is even higher in anembryonic miscarriages. Autosomal trisomies are the most common (51.9%) and arise de novo as a result of meiotic nondisjunction during gametogenesis in parents with normal karyotypes. The relative frequency of each type of trisomy differs considerably.

Trisomy 16, which accounts for about one-third of all trisomic abortions, has not been reported in live-born infants and is therefore uniformly lethal. Trisomy 22 and 21 follow in frequency. The next most common chromosomal abnormalities, in decreasing order, are monosomy 45,X (the single most common karyotypic abnormality), triploidy, tetraploidy, translocations, and mosaicism.

FIG. 4.3. The frequencies of chromosomal anomalies among 3040 spontaneously aborted fetuses related to the duration of pregnancy. For comparison, the frequency of chromosomal anomalies among 54,749 newborn infants is shown. (Data from Shiota K, Uwabe C, Nishimaura H. High prevalence of defective human embryos at the early implantation period. Teratology 1987;35:309; Boue J, Boue A, Lazar P. Retrospective and prospective epidemiological studies of 1500 karyotyped spontaneous human abortions. Teratology 1975;12:11; Lauritsen JG. Aetiology of spontaneous abortion: a cytogenetic and epidemiological study of 288 abortuses and their parents. Acta Obstet Gynecol Scand Suppl 1976;52:1; and Creasy MR, Crolla JA, Alberman ED. A cytogenetic study of human spontaneous abortions using banding techniques. Hum Genet 1976;31:177.)

Parental Factors In a small percentage of cases, one member of the couple is the carrier of a balanced translocation, and the offspring of these parents may be repeatedly aborted. Media publicity tends to give the impression that a variety of agents such as infections, video display terminals, cigarette smoking, coffee, ethanol, chemical agents, and drugs (see Chapter 7) markedly increase the risk of miscarriage. In reality, there is little credible evidence that extrinsic factors account for anything other than a very small proportion of early pregnancy loss.

PATHOLOGY Most miscarriages occur within a few weeks after the death of the embryo or rudimentary analog. Initially, there is hemorrhage into the decidua basalis, with necrosis and inflammation in the region of implantation. The gestational sac is partially or entirely detached. Uterine contractions and dilation of the cervix usually result in expulsion of most or all of the products of conception. When the sac is opened, fluid is often found surrounding a small macerated embryo, or there may be no visible embryo in the sac. Histologically, hydropic degeneration of the placental villi caused by retention of tissue fluid is common.

CLINICAL FEATURES AND TREATMENT An unrecognized pregnancy episode should always be considered a possibility in any woman of reproductive age with abnormal bleeding or pain. Each new pregnant patient should also be instructed to notify her physician promptly about vaginal bleeding or uterine cramps. Since management depends on a number of factors, it is convenient to consider the clinical aspects of miscarriage under the following subgroups. Threatened Miscarriage Any bloody vaginal discharge or uterine bleeding that occurs during the first half of pregnancy has traditionally been assumed to be a threatened miscarriage. Because as many as 25% of pregnant women have some degree of spotting or bleeding during the early months of gestation, it is a common diagnosis. Bleeding associated with threatened miscarriage is typically scanty, varies from a brownish discharge to bright red bleeding, and may occur repeatedly over the course of many days. It usually precedes uterine cramping or low backache. On pelvic examination, the cervix is closed and uneffaced, and no tissue has passed. The differential diagnosis includes ectopic pregnancy, molar pregnancy, vaginal ulcerations, cervicitis with bleeding, cervical erosions, polyps, and carcinoma. A viable conceptus can be detected with modern ultrasound as early as 5.5 weeks of gestation. The ability to visualize the embryo and embryonic heart motion has made evaluation and management of threatened miscarriage more precise. However, accurate knowledge of gestational age is necessary for proper interpretation. Ultrasound findings are unreliable at 3 to 4 weeks gestation, and what appears to be an empty uterus can be misinterpreted as an abnormal intrauterine or ectopic pregnancy when it is actually a normal early gestation. If there is any doubt of normalcy, it is best to perform serial ß-hCG measurements and a follow-up sonogram. From 5 to 6 weeks, the yolk sac and gestational sac are visible by transvaginal ultrasound, and the embryo with cardiac activity is seen soon after that. Abnormal gestational sac and yolk sac size, an embryo small for dates, and slow embryonic heart rates suggest impending pregnancy loss. The presence of an appropriately sized embryo with a normal cardiac rate is encouraging, even in the setting of uterine bleeding, and more than two thirds of these will survive. In the absence of signs of miscarriage, more than 95% of pregnancies continue if a live embryo is demonstrated ultrasonically at 8 weeks gestation. These embryos have a very low mortality rate during the next few weeks, and the subsequent pregnancy loss rate is only 1% if a live fetus is seen at 14 to 16 weeks gestation. Although there is no convincing evidence that any treatment favorably influences the course of threatened miscarriage, a sympathetic attitude by the physician along with continuing support and follow-up are important to patients. This includes a tactful explanation about the pathologic process and favorable prognosis when the pregnancy is viable. An optimistic but cautious approach is prudent, since a few of these women will have a later embryonic or fetal death. It is reasonable to advise patients to remain available to medical care until it can be determined whether the symptoms will persist or cease. Continued observation is indicated as long as bleeding and cramping are mild, the cervix remains closed, quantitative ß-hCG levels are increasing normally, and a normal embryo or fetus is evident on follow-up sonogram. If the bleeding and cramping progressively increase, the prognosis becomes worse. An unfavorable outcome is also associated with negative or falling ß-hCG values, sonographic evidence of an embryo or fetus decreasing in size ( Fig. 4.4), a slow heart rate, and a uterus that is not increasing in size on pelvic examination. If careful clinical evaluation indicates that the conceptus is no longer viable, the treatment options are expectant management or evacuation of the uterus. In women with minimal intrauterine tissue by ultrasound, waiting for spontaneous passage of the products of conception is possible. The complication rate may be decreased by elective uterine curettage in patients with significant amounts of tissue (see “ Missed Miscarriage” later in the chapter).

FIG. 4.4. Ultrasonic comparison of (A) an anembryonic pregnancy with no fetal tissue that is destined to abort with (B) a normal gestational sac with a transonic area, echogenic rim, and fetal pole.

Inevitable and Incomplete Miscarriage Early pregnancy loss is a process rather than a single event. Previously classified as different entities, inevitable and incomplete miscarriages present a similar clinical picture and are treated in the same way. A miscarriage is inevitable when bleeding or gross rupture of the membranes is accompanied by pain and dilation of the cervix. The miscarriage is incomplete when the products of conception have partially passed from the uterine cavity, are protruding from the external os, or are in the vagina with persistent bleeding and cramping. Placental tissue is more likely to be retained when this occurs in the second trimester. Bleeding can be profuse and occasionally produces hypovolemia. A careful vaginal examination usually establishes the diagnosis. Rarely, one conceptus is aborted, and a normal retained twin proceeds to delivery at term. This unusual situation can be diagnosed by ultrasound at the time of first-trimester bleeding. There is otherwise no fetal survival in inevitable or incomplete miscarriages. Evacuation of the uterus is advisable to prevent maternal complications from further hemorrhage or infection. In most cases, suction curettage can be performed promptly and safely in an outpatient setting using analgesia, a paracervical block, and an intravenous infusion of normal saline containing 10 to 20 U of oxytocin. Often, the cervix is dilated, and the products of conception can be removed from the cervical canal and lower uterine segment with ring forceps to facilitate uterine contractions and hemostasis. Suction curettage is performed using a plastic curette and vacuum pressure. As the curettage proceeds, tissue can be seen as it flows through the curette and suction tubing. The curette is rotated 360 degrees clockwise as it is withdrawn, and the procedure is repeated in a counterclockwise direction. When a grating sensation is noted and no more tissue is obtained, the endometrial cavity has been emptied. Preparation is necessary to anticipate any problems such as allergic reactions to medication, uterine atony, uterine perforation, seizure, or cardiac arrest. In selected cases, a hemoglobin level should be obtained, and blood replacement may be necessary if hemorrhage occurs. If measures taken in the emergency room fail to promptly control bleeding, the patient should be transferred to the operating room for an examination under anesthesia and evacuation of the uterus. After curettage, the patient is observed for several hours. When stable, she is discharged and followed as an outpatient. Medical management of incomplete miscarriage has been explored in well-designed trials. Misoprostol may be used instead of surgical evacuation of the uterus in clinically stable patients, though the need for uterine curettage in the misoprostol-treated patients is high (50%). In spite of this, misoprostol treatment of incomplete miscarriage is associated with lower rates of short- and long-term complications compared to surgical evacuation. Rh-negative women with miscarriage should receive 50 g (in the first trimester) or the standard 300-g (in the second trimester) dose of Rh immune globulin to prevent Rh immunization. The tissue obtained should be examined to confirm the presence of products of conception and rule out the possibility of ectopic pregnancy. Complete Miscarriage Patients followed for a threatened miscarriage are instructed to save all tissue passed, so it can be inspected. When the entire products of conception have passed, pain and bleeding soon cease. If the diagnosis is certain, no further therapy is necessary. In questionable cases, ultrasound is useful to determine that the uterus is empty. In some circumstances, curettage may be necessary to be sure that the uterus is completely evacuated. Removal of remaining necrotic decidua decreases the incidence of bleeding and shortens the recovery time. Missed Miscarriage In this situation, expulsion of the conceptus does not occur despite a prolonged period after embryonic death. The reason that some dead embryos do not abort spontaneously is not clear. Typically, the patient's symptoms of pregnancy regress, the pregnancy test becomes negative, and no fetal heart motion is detected by ultrasound. Most patients do eventually abort spontaneously, and coagulation defects due to retention of a dead fetus are rare in the first half of pregnancy. However, expectant management is emotionally trying, and many women prefer to have the uterus evacuated. During the first trimester, this is done by suction curettage preceded by cervical preparation using misoprostol or insertion of laminaria if the cervix is closed. The procedure is often performed in a hospital setting with intravenous fluids and blood available in case significant bleeding occurs. Evacuation of the uterus by medical means is also an acceptable approach in the first trimester. In one randomized, controlled trial, an 80% complete abortion rate was achieved using 800-µg of misoprostol (four 200-µg tablets) per vagina every four hours, and the need for dilation and curettage (D&C) was reduced to 28%. Most patients responded to the first dose of misoprostol. Combination regimens that include methotrexate or RU-486 with misoprostol appear even more promising, but have not yet been tested in missed abortion. In the second trimester, the uterus can be emptied by dilation and evacuation (D&E) or induction of labor with intravaginal prostaglandin E 2 (PGE 2 ) or misoprostol. D&E is an extension of the traditional D&C and vacuum curettage. It is especially appropriate at 13 to 16 weeks gestation, although many proponents use this procedure through 20 weeks. The cervix is usually first prepared using misoprostol or passively dilated with laminaria to avoid trauma, and the fetus and placenta are mechanically removed with suction and instruments. If induction of labor is chosen, vaginal PGE 2 is used, one 20-mg suppository is placed high in the posterior vaginal vault every 4 hours until the fetus and placenta are expelled. Between 2.5 and 5 mg of diphenoxylate given orally and 10 mg of prochlorperazine given intramuscularly can control diarrhea and nausea, and narcotics or epidural anesthesia can be used to control pain. In this situation, a retained placenta is relatively common and may require manual removal and uterine curettage. Misoprostol, 200-µg tablets placed high in the vagina every 4 hours, is equally effective. This regimen may cause less nausea, vomiting, diarrhea, and fever than PGE 2. Septic Miscarriage Septic abortion, once a leading cause of maternal mortality, has become a less frequent occurrence, primarily because changes in abortion laws have made pregnancy terminations by physicians available to women with unwanted pregnancies. However, any type of spontaneous miscarriage can also be complicated by infection. The infection is most commonly endometritis but can progress to parametritis and peritonitis. These patients present with fever, abdominal tenderness, and uterine pain. In severe cases, local infection progresses to septicemia and septic shock. The polymicrobial infection mirrors the endogenous vaginal flora and includes Escherichia coli and other aerobic, enteric, Gram-negative rods, group B-hemolytic streptococci, anaerobic streptococci, Bacteroides species, staphylococci, and microaerophilic bacteria. The initial evaluation and management of septic abortion should include several steps: physical and pelvic examination complete blood cell count and determination of electrolyte, blood urea nitrogen, and creatinine levels type and screen or crossmatch of blood smears from cervix for Gram stain aerobic and anaerobic cultures of endocervix, blood, and available products of conception indwelling Foley catheter intravenous fluids (e.g., saline, Ringer lactate) through a large-bore angiocatheter administration of 0.5 mL of tetanus toxoid, given subcutaneously for immunized patients, or 250 U of tetanus immune globulin, administered deep within the muscle supine and upright radiographs of the abdomen to detect free air or foreign bodies. Optimal therapy consists of evacuation of the uterus and aggressive use of parenteral antibiotics before, during, and after removal of necrotic tissue by curettage ( Table 4.1). Prompt removal of the infected tissue is important and should be performed within a few hours after beginning intravenous antibiotics. Numerous antibiotic regimens have been recommended, but high-dose, broad-spectrum coverage as outlined in Table 4.1 is essential. Although most patients with septic abortions respond favorably to treatment, septic shock syndrome is a serious complication that requires aggressive management in an intensive care setting (see Chapter 25).

TABLE 4.1. Antibiotic regimens for septic abortion

RECURRENT MISCARRIAGE Recurrent miscarriage (RM), traditionally defined as three or more consecutive first-trimester spontaneous losses, affects up to 1% of couples. Primary recurrent miscarriage is diagnosed in women who have never had a successful pregnancy, and secondary recurrent miscarriage in those whose repetitive losses follow a live birth. There is no specific classification for women who have multiple miscarriages interspersed with normal pregnancies. It is generally agreed that a workup for possible causes of RM is indicated in most patients after two or three consecutive miscarriages. The management of couples with RM is controversial. This clinical entity has received much attention in the lay and medical literature during the past decade. A definite cause is established in no more than 50% of couples, and several alleged causes of RM are controversial. Despite publicity to the contrary, there is little evidence that poor nutrition, infections, unrecognized diabetes, toxic agents, or psychological trauma are significant etiologic factors. Some alleged experts and Internet sites inappropriately emphasize unproven hypotheses and results from poorly designed clinical studies. Seeking a solution, some patients and physicians may explore less well-accepted etiologies and empirical or alternative treatments. Moreover, new diagnostic tests for RM are continually being proposed to replace those that have been disproved and discarded. For example, antithyroid antibodies, elevated follicular-phase luteinizing hormone levels, circulating maternal embryotoxic factor, and abnormal lymphocyte subset ratios (elevated CD56 + levels) have been touted within the last decade. It is beyond the scope of this chapter to critically analyze each new “treatment” or assay, but the mechanism of pregnancy loss and potential relationship to each of these remains largely theoretical. Until effective treatments are identified and proven by properly designed studies, these screening tests have little use in the routine evaluation of patients with RM. Typically, investigation of anatomic, hormonal, genetic, and infectious factors has been recommended. Even these may be criticized because the derivation of their diagnostic use and treatments advocated are empirical and have come under scrutiny because they were never submitted to properly designed study. Importantly, evidence has mounted that the average women with RM has a fairly good prognosis for a successful next pregnancy without any specific treatment. Known and Suspected Causes of Recurrent Miscarriage Structural Uterine Defects Hysterosalpingography, magnetic resonance imaging, hysteroscopy, sonohysteroscopy, and laparoscopy can be used to diagnose septate uterus, other müllerian anomalies, uterine defects associated with diethylstilbestrol exposure, submucous myomas, and intrauterine synechiae. The prognosis for successful pregnancies in patients with müllerian anomalies is related to the type of malformation, with asymmetric fusion defects carrying the worst prognosis and septate, bicornuate, and didelphic uteri carrying increasingly better prognoses. In patients with RM, the prevalence of these anatomic defects is approximately 10% to 15%. The cause of pregnancy loss in women with uterine anomalies is uncertain. A diminished blood supply interfering with normal implantation and placentation and the reduced size of the uterine cavity are often cited as possible causes, but these reasons seem unlikely or insubstantial in the setting of the arcuate uterus. Abdominal metroplasty has been replaced in most cases by the hysteroscopic removal of uterine septa. This procedure can be accomplished in an outpatient setting and eliminates the need for cesarean delivery. Noncontrolled, retrospective studies suggest the subsequent live-birth rate is greater than 80%. Removal of synechiae and submucous myomas can also be performed hysteroscopically (see Chapter 47). Endocrine Problems The luteal phase defect (LPD) has long been thought to be a cause of spontaneous abortion, but the evidence linking LPD to recurrent abortion is subject to criticism. It is traditionally thought that women with LPD have short menstrual cycles, postovulatory intervals less than 14 days, and secondary infertility. LPD was initially thought to be due to failure of the corpus luteum to make enough progesterone to establish a mature endometrial lining suitable for placentation. This theory has evolved to implicate poor follicular-phase oocyte development, which results in disordered estrogen secretion, inadequate ovarian steroidogenesis, and subsequent maldevelopment of endometrial receptors. In turn, these effects could result from excess luteinizing hormone or hyperandrogenic states. Some investigators claim that LPD accounts for over one fourth of cases of RM, but studies of this disorder have not included concurrently tested controls. Also, there is little agreement on the criteria necessary to make the diagnosis. Endometrial biopsy or luteal-phase serum progesterone levels are the most widely accepted diagnostic tests. Both are timed for the late luteal phase of the cycle. The endometrial biopsy is histologically dated, and a lag greater than 2 to 3 days is considered suspect. However, this should be confirmed by repeat biopsy, because delayed endometrial histology can occur sporadically in women with no reproductive problems. To further confuse the issue, normal women have endometrial histology suggestive of LPD in up to 50% of single menstrual cycles and 25% of sequential cycles. Though the association between LPD and RM remains speculative, many clinicians have treated women with RM with progesterone in their next pregnancy. One commonly advocated treatment is a 25-mg progesterone suppository inserted into the vagina twice a day (morning and night) beginning after ovulation and continuing until menses begin or through the first 8 to 10 weeks of pregnancy. Comparable doses of oral micronized progesterone have also been used. No properly designed studies have evaluated the role of progesterone treatment in women with RM with LPD. Older studies and meta-analyses are difficult to interpret because of marked differences in inclusion criteria and how LPD was diagnosed, the use of various progesterone compounds, and the small number of women studied. In a more recent randomized trial, a subgroup of women with polycystic ovary syndrome (PCOS) and three or more miscarriages were randomized to treatment with either progesterone or placebo pessaries. There was no difference in the pregnancy outcomes. Clomiphene and other ovulatory agents have been tried to improve follicular development and corpus luteum function, but the results have been variable. Human chorionic gonadotropin has been used in an attempt to stimulate the corpus luteum support of pregnancy in women with RM. One placebo-controlled, multicentered trial found no significant difference in the successful pregnancy rates (83% vs. 79%). In summary, the relationship between the LPD and recurrent pregnancy loss remains a subject of controversy. It has not been shown conclusively that progesterone treatment or corpus luteum support influences pregnancy outcome in women with recurrent pregnancy loss. PCOS has been found in one third or more of women with RM. However, the diagnosis of PCOS in women with RM does not predict a worse pregnancy outcome than in women with RM without PCOS. There is no known effective therapy for women with PCOS and RM. Genetic Abnormalities Parental chromosomal anomalies are found in approximately 3% to 5% of couples with RM. Cytogenetic examination of both partners is helpful to predict recurrence and forms the basis for genetic counseling. Most abnormalities are balanced translocations, with two-thirds being reciprocal translocations and one-third robertsonian translocations. Couples with balanced translocations have spontaneous loss rates ranging from 50% for reciprocal translocations to 25% for robertsonian translocations. All couples with a parental chromosomal abnormality deserve counseling about genetic amniocentesis or chorionic villus sampling in any future pregnancy to exclude a serious fetal chromosomal abnormality. Parental chromosomal abnormalities do not usually preclude further attempts at pregnancy, because most couples eventually have normal offspring. For the rare homologous robertsonian translocation that prevents successful pregnancy, therapeutic possibilities include artificial donor insemination, in vitro fertilization with donor oocytes, and adoption. Chromosomal analysis of the products of conception is also clinically useful, particularly in the evaluation of the reason for failure of a treatment regimen. Molecular mutations that may be shown in the future to cause recurrent miscarriages include lethal, single-point mutations, possibly linked to MHC genes; mutations in genes that code for products critical for normal development; mutations in homeobox genes that control transcriptional regulation; mutations that lead to severe metabolic errors and embryonic death; and disorders of protooncogenes and oncogenes. One group has shown that certain polymorphisms of the HLA-G gene are associated with significantly higher rates of miscarriage among couples presenting with RM. Also, marked skewing of the normal 50:50 distribution of X chromosome inactivation in the mother, a condition termed highly skewed X-chromosome inactivation, may be associated with otherwise unexplained RM. Before testing is recommended, confirmation of these molecular genetic associations in different populations is required. For now, commercially available tests for these conditions are not widely available, and there are no proven treatment options. Autoimmune Disorders Antiphospholipid Syndrome Antiphospholipid syndrome (APS) has been recognized as a proven cause of pregnancy loss for over a decade. Approximately 5% to 15% of women with RM have lupus anticoagulant (LA), anticardiolipin (aCL), or both. These acquired antiphospholipid autoantibodies are induced by as yet unknown stimuli in the setting of aberrant immunoregulation. Low levels of immune globulin G or immune globulin M aCL are of questionable significance. Though women with APS may present with RM in the first trimester, fetal death in the second or early third trimesters may be more specific for the condition. Patients with high levels of aCL or a history of prior fetal death are at greatest risk of another fetal loss. The cause of fetal death appears to be a decidual vasculopathy that results in decidual infarction and insufficient blood flow to the placenta. Intervillous thrombosis has also been described. However, these lesions are nonspecific, and the degree of pathology is not always sufficient to explain the fetal death. The mechanisms by which aCL may cause decidual vasculopathy and fetal death are unknown. A number of pathophysiologic mechanisms have been proposed, including an imbalance of local prostacyclin and thromboxane production, enhanced platelet aggregation, decreased activation of protein C, increased tissue factor, and decreased trophoblast annexin V production or availability. Most recently, the complement system has been invoked as having a major role in antiphospholipid syndrome-related pregnancy loss. Maternally administered heparin is widely considered the treatment of choice for APS pregnancies, both to improve embryo-fetal outcome and protect the mother from thrombotic events ( Table 4.2). Treatment is usually initiated in the early first trimester after ultrasonographic demonstration of a live embryo. The dose of heparin required for safe and effective treatment, however, is debated. Some experts use relatively low doses of heparin (e.g., 5000 U of standard heparin b.i.d.), particularly when treating women with recurrent preembryonic or embryonic losses. However, higher doses of heparin are recommended for patients with APS with prior thrombosis, and some experts urge full anticoagulation. The optimal dose of heparin is controversial for women whose APS is diagnosed because of prior fetal loss or neonatal death after delivery before 34 weeks gestation due to severe preeclampsia or placental insufficiency, but who do not have a history of thromboembolism. These women are at risk for thromboembolic disease, and it is our opinion that these cases should receive sufficient thromboprophylaxis. Low molecular-weight heparins (LMWHs) are widely used in Europe for the treatment of APS pregnancy, and there is little reason to suspect that the appropriate use of LMWHs differs from that of standard heparin with regard to efficacy. In most case series and trials, daily low-dose aspirin is included in the treatment regimen. One important caveat deserves mention—a small, placebo-controlled trial found that otherwise

healthy women with RM and low titers of antiphospholipid antibodies do not require treatment.

TABLE 4.2. Subcutaneous heparin regimens used in the treatment of antiphospholipid syndrome during pregnancy Intravenous immune globulin has also been used during pregnancy, usually in conjunction with heparin and low-dose aspirin, especially in women with particularly poor past histories or recurrent pregnancy loss during heparin treatment. However, a randomized, controlled, pilot study of intravenous immune globulin treatment during pregnancy in unselected APS cases proved negative. Anticoagulant coverage of the postpartum period in women with APS and prior thrombosis is critical. We prefer switching the patient to warfarin thromboprophylaxis as soon as she is clinically stable from delivery. In most cases, an international normalized ratio of 3.0 is desirable, and postpartum coverage should extend for 6 to 8 weeks after delivery. Because of their risk for thrombosis, the same strategy is recommended in women without prior thrombosis but in whom APS is diagnosed because of prior fetal loss or neonatal death after delivery at or before 34 weeks gestation for severe preeclampsia or placental insufficiency. Both heparin and warfarin are safe for nursing mothers. The need for postpartum anticoagulation in women with primary APS diagnosed solely on the basis of recurrent preembryonic and embryonic losses is unclear. Other Autoimmune Disorders Autoantibodies to thyroid antigens are associated with a modest increased rate of pregnancy loss if identified in early pregnancy or immediately before pregnancy. Some investigators have found a significant proportion of women with RM to have antithyroid antibodies; others have not. Even if antithyroid antibodies are associated with RM, no treatment options have proven beneficial. Approximately 15% of women with RM have detectable antinuclear antibodies (ANA). Subsequent pregnancy outcomes among women with a positive ANA test result are similar to those among women with a negative ANA test result. A randomized treatment trial of women with recurrent pregnancy loss and a positive autoantibody result, including ANA, found no benefit to treatment with prednisone and low-dose aspirin and treatment with placebo. Thus, currently available data do not support testing women with recurrent pregnancy loss for ANA. Thrombophilic Disorders The relationship between inherited thrombophilic disorders and recurrent miscarriage has been the subject of intense study within the last several years. The most common inherited thrombophilic disorders are factor V Leiden and prothrombin G20210A mutation, found in approximately 8% and 3%, respectively, of Caucasian women in the United States. These mutations are associated with approximately 25% of isolated thrombotic events and approximately 50% of familial thrombosis. Other less common thrombophilias include deficiencies of the anticoagulants protein C, protein S, and antithrombin III. Hyperhomocysteinemia, most commonly due to the C677T polymorphism of the methylenetetrahydrofolate reductase (MTHFR) gene, is also associated with venous thrombosis. Data regarding the association of these thrombophilic abnormalities and RM do not allow clear and consistent conclusions. The rather obvious fact that most women with common thrombophilic mutations, such as factor V Leiden, the prothrombin G20210A mutation, or the MTHFR C677T mutation, do not have RM further confounds the picture. One prospective study of next pregnancies in women with RM found a significantly lower successful pregnancy rate among those with the factor V Leiden mutation compared to those without (37.5% vs. 69.3%). Various studies are more consistent in finding an association between thrombophilias and second- or third-trimester fetal loss. The odds ratio for stillbirth is significantly higher in women with combined thrombophilic defects. Some women with RM have evidence of ongoing, perhaps chronic, thrombin generation or the formation of thrombosis-related microparticles. These studies underscore the potential importance of prothrombotic states to pregnancy loss, but more research is required to bring the current findings into the clinical realm. Despite the recent interest in this field, no treatment trials have been performed. Thus, which therapy, if any, is effective in promoting successful pregnancy among women with recurrent pregnancy loss and thrombophilia is uncertain. Cervical Incompetence Incompetent cervix, also called premature cervical dilation, is an important cause of second-trimester pregnancy loss. It is characterized by gradual, painless dilation of the cervix with bulging and rupture of the membranes and subsequent expulsion of a fetus too immature to survive. Pregnancy loss from this cervical abnormality usually occurs in the second trimester and is thought to be an entirely different and distinct entity from a first-trimester miscarriage or premature labor in the third trimester. It results from different factors, presents a distinctive clinical picture, and requires different management. Moreover, miscarriage and premature labor are common, but mid-trimester premature cervical dilation is relatively rare. Unlike the rest of the uterus, the cervix is fundamentally a connective tissue structure. The cause of cervical incompetence is obscure, and various etiologic factors have been proposed. Previous surgery or trauma to the cervix, such as D&C, amputation, conization, cauterization, loop electrosurgical excision procedure (LEEP), or traumatic delivery, seem to be factors in some cases. In other instances, congenital cervical structural defects, uterine anomalies, or abnormal cervical development associated with in utero diethylstilbestrol exposure appear to play a role. Little agreement can be found regarding the diagnosis of cervical incompetence, except that it is one of exclusion that requires careful evaluation to rule out other potential causes of mid-trimester pregnancy loss. Other causes of very early delivery include abruptio placentae, chorioamnionitis, and uterine anomalies, but they usually present different clinical pictures. Whether or not the condition can be diagnosed during the nonpregnant state by methods designed to calibrate the diameter of the endocervical canal or during early pregnancy by sonographic findings is questionable. The absolute diagnosis of cervical incompetence can be made only by seeing the fetal membranes bulging through the partially dilated cervix of a patient in the second trimester of pregnancy who is not in labor. More typically, a presumptive diagnosis is made from the characteristic history of apparently silent dilation of the cervix followed by rupture of the membranes and a relatively painless, rapid labor with delivery of an immature infant. Also, the fetus is typically alive at the time of presentation to the hospital; delivery of a dead, macerated fetus makes the diagnosis of cervical competence questionable. Upon inspection in the nonpregnant state, the cervix may be shortened with a patulous os or may be deformed with lacerations that sometimes extend to the vaginal fornix. Although bed rest, various intravaginal devices, and pharmacologic agents have been used with some success, the generally accepted treatment for incompetent cervix is surgical. Various methods have been described, but the McDonald or Shirodkar procedures ( Fig. 4.5) are most commonly employed prophylactically. These are techniques performed vaginally, usually under regional anesthesia, designed to reinforce the cervix close to the level of the internal os. If there is insufficient cervical tissue to allow placement of a cerclage vaginally, an abdominal approach is sometimes used. The reinforcement suture is usually placed toward the end of the first trimester after ultrasound documentation of a live fetus, after the risk of miscarriage has passed, and before the cervix starts to dilate.

FIG. 4.5. Incompetent cervix can be treated by three procedures. A: In the McDonald cerclage procedure, a multiple-bite suture using large, monofilament nylon is placed around the cervix and tied securely to reduce the diameter of the cervical canal to a few millimeters. B: In the Shirodkar procedure, Merseline tape encircling the cervix is passed under the mucosa and anchored to the cervix anteriorly and posteriorly with interrupted sutures. C: With transabdominal cervicoisthmic cerclage, a Merseline band is placed in an avascular space medial to the uterine vessels at the level of the cervicouterine junction. Placement of the cerclage in the second trimester after cervical change has occurred is sometimes necessary but appears less effective. The procedure should not be used if the diagnosis is in doubt, if membranes are ruptured, or if vaginal bleeding and cramping are part of the clinical picture. There is no evidence that postoperative antibiotics, progesterone, or tocolytic agents are useful adjuvants. If membranes rupture or labor ensues at any time, removal of the cerclage should be strongly considered to prevent chorioamnionitis, sepsis, cervical laceration, and rupture of the uterus. Otherwise, the suture is removed when fetal maturity is achieved, usually after 37 weeks gestation, which is often followed by the onset of labor and a relatively rapid delivery. If the patient desires further pregnancies, some physicians leave the cerclage in place and deliver by cesarean section. Often, the history is not typical, and it is difficult to determine whether or not premature cervical dilation will occur in a subsequent pregnancy. These patients are usually followed with frequent vaginal examinations and serial sonograms to diagnose potential cervical changes. Transvaginal ultrasound can be used to accurately assess cervical length and may be useful in deciding for or against cerclage in women with an unclear history. The effectiveness of cerclage has often been questioned, even in women with a classic clinical picture. Nevertheless, when patients are carefully selected, this type of management is 80% to 90% successful in preventing delivery of an immature fetus. There is little difference in the fetal survival rates between the McDonald and Shirodkar techniques. The procedure has also been used prophylactically for patients with previous preterm deliveries with less convincing evidence for cervical incompetence. Results from prospective randomized studies have led to conflicting conclusions. Idiopathic Causes Because most cases of RM have no discernible cause, alloimmune factors have long been suspected. These have yet to be proven, largely because little is known about the mechanisms that prevent immunologic rejection of the conceptus in successful pregnancies (see Chapter 18). Early reports proposed that HLA compatibility between couples, the absence of maternal leukocytotoxic antibodies, or the absence of maternal blocking antibodies were related to RM. The

importance of these factors has not been substantiated, and these expensive tests are no longer clinically indicated. Contemporary research is focused on local decidual or trophoblast immunosuppressive factors such as cytokines, growth factors, hormones, enzymes, and endometrial proteins. Some of these immunoactive factors appear to be necessary for implantation and growth and development of the early placenta and embryo, and others may cause abortion, when expressed. There are, however, no practical clinical tests available for these factors and no proven treatment if they were found abnormal. Although no alloimmune mechanism has been unequivocally shown to cause RM in humans, several types of immunotherapy have been advocated. Originally, the attempt to improve maternal immunotolerance in recurrent aborters was based on evidence that pretransplant blood transfusions decreased rejection of organ allografts and that the rate of resorptions or abortions in animal models was reduced by prior immunization with spleen cells from a paternally related strain. The most popular regimen involves injections of the father's leukocytes. Though proponents persist, this treatment is questionable at best and harmful at worst. Most randomized trials have proven negative, and the largest and only multicenter randomized trial found that treated pregnancy outcomes were worse in the women who received leukocyte immunization. Based largely on this trial, the U.S. Food and Drug Administration has stated that the administration of this therapy for RM may only be done as part of a clinical investigation, and then only if there is an investigational new drug application in effect. Participating women should be counseled that immunization using viable leukocytes carries the risks of any blood transfusion, such as hepatitis, human immunodeficiency virus, and cytomegalovirus infections. Reactions have been uncommon but include soreness and redness at the injection site, cutaneous graft-versus-hostlike reaction, fever, maternal platelet and leukocyte alloimmunization, and blood group sensitization. Intravenous immune globulin has been proposed as an alternative therapy in patients with idiopathic RM. A number of randomized trials have been reported, and the results are conflicting. Nevertheless, this treatment seems to be no more successful than paternal cell immunization, and intravenous immune globulin is not recommended outside of a research protocol by either the American College of Obstetricians and Gynecologists or the American Society of Reproductive Medicine. It is imperative for physicians to recognize that the prognosis for idiopathic RM is by no means dismal. Numerous studies and several meta-analyses indicate that the average next pregnancy live-birth rate for placebo-treated women with idiopathic RM is 60% to 70%. Many couples see this modestly favorable prognosis in a somewhat positive light, and it compares favorably with the prognosis for conditions such as APS or parental karyotype abnormalities. Understanding this prognosis may allow the couple to choose against an expensive unproven treatment. One caveat—as expected, increasing maternal age and increasing number of miscarriages are negative variables. Recommendations for Recurrent Miscarriage The scheme for a reasonable and cost-effective evaluation of women with RM shown in Table 4.3 is based on current guidelines published by the American College of Obstetricians and Gynecologists and the Royal College of Obstetricians and Gynaecologists. A sympathetic attitude by the physician is crucial—establishment of trust and rapport and a sincere appreciation of the distress and grief experienced by these couples permit tactful and thorough discussions with patient and partner. It is reasonable to institute an evaluation after two consecutive miscarriages in anxious women or if the patient has few reproductive years remaining or has had an infertility problem. Couples interested in an investigational protocol are perhaps best referred to legitimate research centers.

TABLE 4.3. Proposed evaluation for recurrent miscarriage

SUMMARY POINTS Miscarriage is the most common complication of pregnancy, and the most frequent etiology is a chromosomal abnormality of the conceptus. Ultrasound is helpful in determining whether or not the embryo is viable, and appropriate modern management may be either observation or evacuation of the uterus. Recurrent early pregnancy loss is sometimes associated with underlying maternal abnormalities that can be detected with standard tests. Physicians should be aware of unproven tests and controversial treatments in order to best counsel their patients. SUGGESTED READINGS Aarts JM, Brons JTJ, Bruinse HW. Emergency cerclage: a review. Obstet Gynecol Surv 1995;50:459. Aldrich CL, Stephenson MD, Karrison T, et al. HLA-G genotypes and pregnancy outcome in couples with unexplained recurrent miscarriage. Mol Hum Reprod 2001;7:1167–1172. Althuisius SM, Dekker GA, Hummel P, et al. Final results of the Cervical Incompetence Prevention Randomized Cerclage Trial (CIPRACT): therapeutic cerclage with bed rest versus bed rest alone. Am J Obstet Gynecol 2001;185:1106–1112. American College of Obstetrician and Gynecologists. Management of recurrent early pregnancy loss. ACOG Practice Bulletin 2001;24. Branch DW, Khamashta MA. Antiphospholipid syndrome: obstetric diagnosis, management, and controversies. Obstet Gynecol 2002 ( in press). Branch DW, Silver RM, Pierangelli SS, et al. Antiphospholipid antibodies other than lupus anticoagulant and anticardiolipin antibodies in women with recurrent pregnancy loss, fertile controls, and antiphospholipid syndrome. Obstet Gynecol 1997;89:549. Brent RL, Beckman DA. The contribution of environmental teratogens to embryonic and fetal loss. Clin Obstet Gynecol 1994;37:646. Chung TK, Lee DT, Cheung LP, et al. Spontaneous abortion: a randomized, controlled trial comparing surgical evacuation with conservative management using misoprostil. Fertil Steril 1999;71:1054–1059. Clifford K, Rai R, Watson H, et al. Does suppressing luteinising hormone secretion reduce the miscarriage rate? Results of a randomised controlled trial. BM J 1996;312:1508–1511. Davis OK, Berkeley AS, Naus GJ, et al. The incidence of luteal phase defect in normal, fertile women determined by serial endometrial biopsies. Fertil Steril 1989;51:582–586. Daya S. Efficacy of progesterone support for pregnancy in women with recurrent miscarriage. A meta-analysis of controlled trials. Br J Obstet Gynaecol 1989;96:275–280. Erkan D, Merrill JT, Yazici Y, et al. High thrombosis rate after fetal loss in antiphospholipid syndrome: effective prophylaxis with aspirin. Arthritis Rheum 2001;44:1466–1467. Fantel AG, Shepard TH. Morphological analysis of spontaneous abortuses. In: Bennett MJ, Edmunds DK, eds. Spontaneous and recurrent abortion. Oxford: Blackwell Scientific Publications, 1987:8. Ginsberg JS, Greer I, Hirsh J. Use of antithrombotic agents during pregnancy. Chest 2001;119[Suppl 1]:122S–131S. Goldstein P, Berrier J, Rosen S, et al. A meta-analysis of randomized control trials of progestational agents in pregnancy. Br J Obstet Gynaecol 1989;96:265–274. Goldstein SR. Embryonic death in early pregnancy: a new look at the first trimester. Obstet Gynecol 1994;84:294. Holers VM, Girardi G, Mo L, et al. Complement C3 activation is required for antiphospholipid antibody-induced fetal loss. J Exp Med 2002;195:211–220. Hurd WW, Whitfield RR, Randolph JF, et al. Expectant management versus elective curettage for the treatment of spontaneous abortion. Fertil Steril 1997;68:601. Jeng GT, Scott JR, Burmeister LF. A comparison of meta-analytic results using literature vs. individual data. JAMA 1995;274:830. Kelly S, Pollock M, Maas B, et al. Early transvaginal ultrasonography versus early cerclage in women with an unclear history of incompetent cervix. Am J Obstet Gynecol 2001;184:1097–1099. Kutteh WH. Antiphospholipid antibody-associated recurrent pregnancy loss: treatment with heparin and low-dose aspirin is superior to low-dose aspirin alone. Am J Obstet Gynecol 1996;35:1584. Lanasa MC, Hogge WA, Kubic C, et al. Highly skewed X-chromosome inactivation is associated with idiopathic recurrent spontaneous abortion. Am J Hum Genet 1999;65:252–254.

Lanasa MC, Hogge WA, Kubik CJ, et al. A novel X chromosome-linked genetic cause of recurrent spontaneous abortion. Am J Obstet Gynecol 2001;185:563–568. Laskin CA, Bombardier C, Hannah ME, et al. Prednisone and aspirin in women with autoantibodies and unexplained recurrent fetal loss. N Engl J Med 1997;337:148–153. Levine JS, Branch DW, Rauch J. The antiphospholipid syndrome. N Engl J Med 2002;346:752–763. MacNaughton MC, Chalmers IG, Dubowitz V, et al. Final report of the Medical Research Council/Royal College of Obstetricians and Gynaecologists multicentered randomized trial of cervical cerclage. Br J Obstet Gynaecol 1993;100:516. Ober C, Karrison T, Odem RR, et al. Mononuclear-cell immunisation in the prevention of recurrent miscarriages: a randomised trial. Lancet 1999;354:365–369. Patton PE. Anatomic uterine defects. Clin Obstet Gynecol 1994;37:705. Preston FE, Rosendaal FR, Walker ID, et al. Increased fetal loss in women with heritable thrombophilia. Lancet 1996;348:913–916. Rai R, Cohen H, Dave M, Regan L. Randomized controlled trial of aspirin and aspirin plus heparin in pregnancy in women with recurrent miscarriage associated with phospholipid antibodies. BMJ 1997;314:253. Rai R, Backos M, Elgaddal S, et al. Factor V Leiden and recurrent miscarriage—prospective outcome of untreated pregnancies. Hum Reprod 2002;17:442–445. Rai R, Backos M, Rushworth F, Regan L. Polycystic ovaries and recurrent miscarriage—a reappraisal. Hum Reprod 2000:15;612–615. Scientific Advisory Committee of the Royal College of Obstetricians and Gynaecologists. The management of recurrent miscarriage. RCOG ‘Green-top’ Guideline No. 17, 2001. Wilcox AJ, Weinberg CR, O'Connor JF, et al. Incidence of early pregnancy loss. N Engl J Med 1988;319:189. Woelfer B, Salim R, Banerjee S, et al. Reproductive outcomes in women with congenital uterine anomalies detected by three-dimensional ultrasound screening. Obstet Gynecol 2001;98:1099–1103. Wood SL, Brain PH. Medical management of missed abortion: a randomized clinical trial. Obstet Gynecol 2002;99:563–566.

Chapter 5 Ectopic Pregnancy Danforth’s Obstetrics and Gynecology

Chapter 5 Michael J. Heard and John E. Buster

Ectopic Pregnancy

INCIDENCE PATHOGENESIS RISK FACTORS Tubal Damage and Infection Salpingitis Isthmica Nodosa Diethylstilbestrol Cigarette Smoking Contraception Evidence-based Recommendation SIGNS AND SYMPTOMS DIAGNOSIS Serial ß-Human Chorionic Gonadotropin Determinations Ultrasonography Uterine Curettage Evidence-based Recommendation TREATMENT FOR ECTOPIC PREGNANCY Medical Management Surgical Treatment PERSISTENT ECTOPIC PREGNANCY FOLLOWING SALPINGOSTOMY Evidence-based Recommendation ECTOPIC PREGNANCY AND ASSISTED REPRODUCTIVE TECHNOLOGY (ART) Incidence Location Tubal Pathology Ovulation Induction Embryo Transfer Heterotopic Pregnancy Evidence-based Recommendation EXPECTANT MANAGEMENT Evidence-based Recommendation COST ANALYSIS Evidence-based Recommendation RARE TYPES OF ECTOPIC PREGNANCY Abdominal Pregnancy Ovarian Pregnancy Cornual Pregnancy Cervical Pregnancy Heterotopic Pregnancy SUMMARY POINTS REFERENCES Incidence Pathogenesis Risk Factors Signs and Symptoms Diagnosis Treatment for Ectopic Pregnancy Ectopic Pregnancy and Assisted Reproductive Technology

Management of ectopic pregnancy is changing dramatically. Although ectopic pregnancy remains a leading cause of life-threatening first-trimester morbidity, informed clinical suspicion and modern diagnostic procedures now routinely lead to diagnosis and treatment before there are symptoms. Medical therapy with systemic methotrexate, an intervention targeted specifically toward proliferating trophoblasts, is now preferred to surgery as standard first-line therapy. Surgery remains the first choice for hemorrhage, medical failures, neglected cases, and cases where medical therapy is contraindicated. In the wake of these changes, the United States has seen a considerable drop in maternal morbidity and mortality from this disease. Optimal dosing for methotrexate remains controversial. Rules for timing and technique for surgical intervention during medical failures are empirical. Early diagnosis and selection of optimal therapy are key to prevention of complications, preservation of fertility, control of costs, and elimination of mortality. Using an evidence-based approach to diagnosis and treatment for ectopic pregnancy, this chapter provides a comprehensive examination of the standard of care for this serious gynecologic disease.

INCIDENCE The incidence of ectopic pregnancy in the United States is not known precisely. Recent attempts at the Centers for Disease Control and Prevention (CDC) to estimate the incidence of this disease have been thwarted, because many cases are treated medically in outpatient facilities and are not recorded in hospital registries. Where hospital records were used, a relentless increase in ectopic pregnancies from 4.5 per 1,000 in 1970 to 16.8 per 1,000 in 1989 to 19.7 per 1,000 (108,000 cases) in 1992 has been reported. The current incidence will probably prove much higher once the CDC is able to make estimates including outpatient records. Certain epidemiologic trends make this likely. First, there is a continued increase in risk factors ( Table 5.1) in a society with unprecedented sexual liberties. Second, there is increased ascertainment of ectopics from use of more sensitive and specific diagnostic methods that detect many cases that, in the past, would have resolved spontaneously without diagnosis. Third, with the increasing use of assisted reproductive technology (ART) for treatment of infertility, there is an increase risk of ectopics which comprise up to 5% of pregnancies achieved by using ART. Not surprisingly, heterotopics also are being reported with increasing frequency in ART pregnancies. Between 1979 and 1986, 13% of maternal deaths were secondary to ectopic pregnancy; by 1992, this dropped to 9%. However, ectopic pregnancies continue to be the leading cause of maternal death in the first trimester accounting for 5% to 6% of all maternal deaths in the United States. Ninety percent of these deaths were due to hemorrhagic complications.

TABLE 5.1. Risk factors associated with ectopic pregnancy

PATHOGENESIS Any event that impairs the ability of the tube to transport gametes or embryos will predispose to ectopic implantation. The clinical picture is determined by the site of ectopic implantation. The most common site of ectopic pregnancy is the fallopian tube, which accounts for 98.3% of all ectopic gestations. Implantation in the ampulla is observed in 79.6% of tubal ectopic pregnancies; 12.3% are in the isthmus, 6.2% are in the fimbrial end, and the remaining 1.9% occur in the interstitial (cornual) region. Ectopic nidation outside the fallopian tubes is rare; only 1.4% of ectopic pregnancies are abdominal pregnancies, 0.15% ovarian, and 0.15% cervical ( Fig. 5.1).

FIG. 5.1. Implantation sites for ectopic pregnancy following natural cycles and assisted reproductive technology.

In most tubal implantations, the proliferating trophoblasts invade the tubal wall. The degree of trophoblastic invasion of maternal tissues, the viability of the pregnancy, and the site of implantation determine the sequence of clinical events. As the trophoblasts proliferate, the growth may extend from the luminal mucosa, into the muscularis and lamina propria, into the serosa and, ultimately, full thickness even into large blood vessels in the broad ligament. With vascular invasion, bleeding takes place which distorts the tube, stretches the serosa, and causes pain. The embryo is abnormal and degenerates in about 80% of cases. Spontaneous tubal abortion occurs in about 50% of tubal ectopic pregnancies and is often clinically silent. Spontaneous tubal abortion with hemorrhage can occur with bleeding that is self-limited. Tubal rupture usually is associated with significant hemorrhage. This complication is most likely to occur in the isthmic part of the tube, which has limited distensibility. Chronic tubal rupture with extension into the broad ligament can produce a pelvic hematoma that can last for several weeks. Unruptured ectopic pregnancies can produce a chronic course, with persistently elevated ß-human chorionic gonadotropin (ß-hCG) levels that may last for weeks.

RISK FACTORS Ectopic pregnancy most often is associated with risk factors leading to tubal epithelial damage, which alters gamete and embryo transport. Meta-analyses identify the risk factors listed in Table 5.1 as the most influential. Tubal Damage and Infection Documented tubal pathology carries a 3.5-fold common adjusted odds ratio for ectopic pregnancy. Patients with a previous ectopic pregnancy are 6 to 8 times more likely to experience another ectopic pregnancy, and 8% to 14% of patients experience more than one ectopic pregnancy. Patients with a history of tubal surgery have a 21-fold common adjusted odds ratio of ectopic pregnancy. Tubal pathology frequently results from pelvic infections. Patients with a history of pelvic infections, including gonorrhea, serologically confirmed chlamydia, and nonspecific pelvic inflammatory disease, have a two-fold to four-fold higher risk of developing an ectopic pregnancy. The ectopic pregnancy rate is 4% in women with laparoscopically proven salpingitis, compared with 0.7% in women with normal tubes. In evaluating histologic specimens of ectopic pregnancy, microscopic evidence of inflammatory disease is present in 38% of cases. Recurrent episodes of pelvic infections increase the likelihood of tubal occlusions: 12.8% after one infection, 35.5% after two infections, and 75% in patients with three or more infections. Salpingitis Isthmica Nodosa Salpingitis isthmica nodosa is a disease defined by an anatomic thickening of the proximal portion of the fallopian tubes with multiple luminal diverticula. This pattern of tubal pathology increases the incidence of ectopic pregnancy by 52% in age- and race-matched controls. Diethylstilbestrol In utero exposure to diethylstilbestrol (DES) alters fallopian tubal morphology, resulting in absent or minimal fimbrial tissue, a small tubal os, and decreased length and caliber of the tube. Abnormal tubal anatomy accounts for the five-fold increase in the risk for ectopic pregnancy. Cigarette Smoking Patients who smoke cigarettes are at a slightly increased risk for ectopic pregnancy. It is difficult to conceptualize the link between ectopic pregnancy and cigarettes. Theories include impaired immunity in smokers predisposing them to pelvic infections, alterations in tubal motility, or a representation of certain lifestyles associated with increased risk of tubal injection. Contraception Intrauterine devices (IUDs) have been associated with ectopic pregnancy. A multicenter case-control study conducted by the World Health Organization in ten countries found an odds ratio of 6.4 for ectopic pregnancy in current IUD users compared with pregnant controls, whereas the odds ratio was only 0.5 when the comparison was made with nonpregnant controls. Similarly, in the Oxford Study of 17,032 contraceptive users, the proportion of unplanned pregnancies that were ectopic was higher in women using IUDs compared with women taking oral contraceptives. Thus, IUDs effectively prevent pregnancy, but if pregnancy occurs in a woman using an IUD, there is increased likelihood that the pregnancy will be ectopic. Tubal ligation carries a similar risk for ectopic pregnancy to what is observed with current IUD use. A meta-analysis using case-control studies found the odds ratio for tubal sterilization to be 9.3 when compared with pregnant controls and 0.52 when compared with nonpregnant controls, a finding confirmed by two additional multicenter case-control trials. As with the IUD, tubal ligations effectively prevent pregnancy, but if pregnancy does occur, the suspicion for an ectopic pregnancy should be high. Electrocoagulation procedures are associated with higher ectopic pregnancy risk than other forms of tubal sterilization, possibly resulting from tubal recanalization or uteroperitoneal fistula formation. Uteroperitoneal fistulas have been found in up to 75% of hysterectomy specimens from women with previous tubal ligations in which the tubes were cauterized flush with the uterus. Oral contraceptives are associated with a reduced risk of ectopic pregnancy when compared with nonpregnant controls but with elevated risk when compared with pregnant controls. This protection is presumably due to the suppression of ovulation by oral contraceptives. It is therefore not surprising that patients who take emergency contraception, such as oral contraceptives after fertilization, are at substantial risk for an ectopic pregnancy. This has been attributed to altered tubal motility. Barrier contraception (condoms, spermicides, and diaphragms) also reduces the odds ratio of ectopic pregnancy. An additional advantage may be attributed to the decreased risk of sexually transmitted diseases in women using barrier methods. Evidence-based Recommendation Women with a previous ectopic pregnancy, tubal surgery, tubal pathology, or in utero DES exposure are at high risk for ectopic pregnancy. Women who have experienced genital infections, infertility, or more than one sexual partner have a moderate risk of ectopic pregnancy. Previous pelvic or abdominal surgery, smoking,

vaginal douching, or an early age of first sexual intercourse have only a slightly increased risk of ectopic pregnancy. Contraception, if used properly, is an effective way of reducing pregnancy, both intrauterine and extrauterine. If pregnancy occurs in women with an IUD, after tubal ligation, or following emergency contraception, suspicion for ectopic pregnancy should be high. (Strength of recommendation: A.)

SIGNS AND SYMPTOMS Today, many ectopic pregnancies never produce symptoms; rather, they are timely diagnosed and treated because the patient is identified as high risk. Table 5.1 summarizes and weighs risk factors that should be examined in every woman who has just been identified as being pregnant. If the diagnosis is delayed, some patients may develop the classic triad of amenorrhea, irregular vaginal bleeding, and lower abdominal pain. Early diagnosis, unfortunately, is not always achievable. Sudden, severe, lower abdominal pain is the most common complaint in 90% to 100% of women with symptoms of an ectopic pregnancy. Pain radiating to the shoulder, syncope, and shock as a result of hemoperitoneum occur in up to 20% of patients. The most common signs are detected upon abdominal examination. Abdominal tenderness is present in 90% of patients and rebound tenderness in 70%. The pelvic examination is usually nonspecific; cervical motion tenderness is present in up to two thirds of patients, while a tender adnexal mass is present in 50%.

DIAGNOSIS Ectopic pregnancy can be diagnosed as early as 4.5 weeks gestation. Unfortunately, visualizing an ectopic pregnancy this early is frequently not possible. More importantly, traditional laparoscopic visualization ( Fig. 5.2, Fig. 5.3, Fig. 5.5) is now rarely necessary. Routine diagnostic tests are serial measurements of ß-hCG, ultrasonography, serum progesterone levels, and uterine curettage.

FIG. 5.2. Laparoscopic visualization of an isthmic ectopic pregnancy.

FIG. 5.3. Laparoscopic visualization of an ampullary ectopic pregnancy.

FIG. 5.5. Ultrasonogram of free fluid noted under the liver edge above the right kidney. Confirmed to be blood from a ruptured ectopic pregnancy at the time of surgery. (Courtesy of R. Mangal, M.D., Obstetrics/Gynecologic Associates, Houston, TX.)

Outpatient diagnosis of ectopic pregnancy using various algorithms has been shown to be safe and effective without need for hospitalization even when the diagnosis is equivocal. The clinical algorithm in Figure 5.4 is highly efficacious in diagnosing ectopic pregnancy.

FIG. 5.4. Diagnostic algorithm for ectopic pregnancy.

Serial ß-Human Chorionic Gonadotropin Determinations ß-hCG determinations used today are based on the enzyme-linked immunosorbent assay (ELISA), detecting low ß-hCG concentrations in urine and serum, 20 mIU/mL down to 1 mIU/mL, respectively. The ß-hCG, produced by trophoblastic cells in normal pregnancy, rises at least 66% and up to two-fold every 2 days. This generally applies to ß-hCG values below 10,000 mIU/mL. Eight-five percent of abnormal pregnancies, whether intrauterine or ectopic, have impaired ß-hCG production with a prolonged doubling time. ß-hCG levels that plateau or fail to rise normally along with a low serum progesterone value should be considered nonviable. If a viable intrauterine gestation is not visible by transvaginal ultrasonography when the ß-hCG is above 2,000 mIU/mL (First International Reference Preparation [IRP]) and no fetal heartbeat can be visualized in the adnexa, uterine curettage can be performed. In this situation, treatment of a nonviable intrauterine pregnancy is performed or ectopic pregnancy is diagnosed when the ß-hCG levels do not fall. These ß-hCG thresholds are not universal, and each institution must identify its own values to avoid terminating normal intrauterine pregnancies.

FIG. 5.6. Transvaginal ultrasonographic illustration of tubal ectopic gestation.

ß-hCG determinations are further employed for diagnosis after uterine curettage. If the ß-hCG fails to decline by 15% from a level drawn immediately before surgery, the pregnancy is presumed ectopic and treatment should be initiated. Ultrasonography Although the uterus and adnexa may be evaluated abdominally or vaginally, transvaginal ultrasonography reliably detects intrauterine gestations when the ß-hCG levels are between 1,000 and 2,000 mIU/mL (First IRP), as early as 1 week after missed menses. An intrauterine gestation should almost always be visualized when the ß-hCG level is greater than 2,000 mIU/mL. Diagnosis of an ectopic pregnancy can be made with 100% specificity but with low sensitivity (15%–20%) if an extrauterine gestational sac containing a yolk sac or embryo is identified. A complex adnexal mass without an intrauterine pregnancy improves sensitivity to 21% to 84% at the expense of lower specificity (93.0%–99.5%). In reviewing the literature, the presence of any noncystic, extraovarian adnexal mass in the absence of an intrauterine gestation was diagnostic of an ectopic pregnancy with 98.9% specificity, 96.3% positive predictive value, 84.4% sensitivity, and a 94.8% negative predictive value. Despite the high resolution of transvaginal ultrasonography, an adnexal mass will not be found in 15% to 35% of patients with an ectopic pregnancy, particularly in early stages. Some sonographic images, such as the pseudogestational sac, may mislead even an experienced examiner to falsely diagnose a gestational sac. Serial ß-hCG concentrations and transvaginal ultrasonography predict ectopic pregnancy with a positive predictive value of 95%. Diagnosis is made routinely by the absence of an intrauterine pregnancy (i.e., gestational sac) at a designated ß-hCG concentration. The vast majority of viable intrauterine pregnancies can be identified by ultrasonography when the ß-hCG is greater than 1,500 mIU/mL (First IRP). However, in those patients with an “indeterminate” ultrasonogram, one fourth have an ectopic pregnancy. Therefore, serial hCG and ultrasonography alone cannot diagnose all ectopic pregnancies. Uterine Curettage Uterine curettage is necessary when a transvaginal ultrasonogram and a rising or plateauing ß-hCG level below the cutoff value are not sufficient for diagnosis. A decrease in the ß-hCG level of 15% or more 8 to 12 hours after curettage is diagnostic of a complete abortion. If the ß-hCG titer plateaus or rises and the trophoblast was not removed by curettage, an ectopic pregnancy is likely. Evidence-based Recommendation Serial ß-hCG determinations, transvaginal ultrasonography, and uterine curettage allow for definitive diagnosis of ectopic pregnancy. A confirmatory laparoscopy is rarely necessary. (Strength of recommendation: A.)

TREATMENT FOR ECTOPIC PREGNANCY Medical Management Methotrexate therapy of ectopic pregnancy has been used successfully over the last two decades. A folic acid antagonist, methotrexate inhibits de novo synthesis of purines and pyrimidines, interfering with DNA synthesis and cell multiplication. Rapidly proliferating trophoblasts are particularly vulnerable to methotrexate and this differential sensitivity forms the basis of the therapy. When methotrexate is administered to pregnant women undergoing planned termination, a single dose of 50 mg/m 2 significantly blunts the hCG increment over the following 7 days and has been associated with a drop in circulating progesterone and 17a-hydroxyprogesterone concentrations prior to abortion. It appears that methotrexate directly impairs trophoblastic production of hCG with a secondary decrement of corpus luteum progestin secretion. Hemodynamically stable patients with unruptured ectopic pregnancy measuring less than or equal to 4 cm by ultrasonography are eligible for methotrexate therapy. Patients with larger masses or evidence of acute intraabdominal bleeding should undergo surgical treatment. The two commonly employed methotrexate treatment regimens are shown in Table 5.2.

TABLE 5.2. Methotrexate protocols: single versus multiple Multiple-dose Methotrexate Multiple-dose methotrexate therapy is tailored to the patient's weight and ectopic pregnancy responsiveness. Outcomes of 12 studies comparing multiple-dose systemic methotrexate with laparoscopic salpingostomy are presented in Table 5.3. Between 1982 and 1997, this tabulation shows 325 cases of ectopic pregnancy treated with multiple-dose methotrexate. Of these cases, 93.8% were treated successfully with multiple-dose systemic methotrexate (no subsequent therapy was required), and 78.9% of the women tested had patent fallopian tubes; in addition, of the women desiring pregnancy, 57.9% had a subsequent intrauterine pregnancy and 7.4% developed a repeat ectopic pregnancy. These rates all compare favorably with conservative surgical management.

TABLE 5.3. Outcome of different treatments for ectopic pregnancy There is one randomized clinical trial comparing laparoscopic salpingostomy with systemic multiple dose methotrexate. In it, 100 patients with laparoscopy-confirmed ectopic pregnancy were randomly treated with systemic methotrexate or laparoscopic salpingostomy. In the 51 patients treated with methotrexate, three (8%) required surgical intervention for active bleeding or tubal rupture. An additional course of methotrexate was required in two patients (5%) for persistent trophoblast, basal on ß-hCG secretion. Of the 44 patients in the salpingostomy group, two patients (5%) failed and required salpingectomies, and eight patients (22%) required treatment with methotrexate for persistent trophoblast. Tubal patency was present in 67% of the patients in the methotrexate group and in 61% in the salpingostomy group. This randomized study and previous meta-analysis have demonstrated the effectiveness of systemic methotrexate therapy as equal to laparoscopic salpingostomy. Single-dose Methotrexate Single-dose methotrexate, although more convenient, is not as efficacious as multiple-dose methotrexate. The high success rates in the initial studies using single-dose methotrexate was most likely due to the inclusion of spontaneously aborting intrauterine pregnancies. Subsequent studies of single-dose methotrexate therapy involving 304 patients are presented in Table 5.3. Although overall success of treatment, measured as no surgical intervention, is 87.2%, 11.5% additional patients required more than one dose of methotrexate. Of the patients considered successfully treated (with one or more doses), tubal patency was found in 81.3% of the women evaluated. The subsequent intrauterine pregnancy rate was 61%, and for ectopic pregnancies 7.8%, in the patients desiring future fertility in the same group (those treated with either one or more doses of methotrexate). Based on the clinical evidence presently available, the routine use of methotrexate as a single-dose intramuscular regimen is probably not as effective as multiple dosage. However, single-dose therapy remains a standard according to publications of the American College of Obstetricians and Gynecologists. With this background, a recent meta-analysis of 26 studies evaluating methotrexate dosing for ectopic pregnancy by Barnhart et al. showed an odds ratio of 1.96 higher likelihood of rupture with use of single-dose methotrexate over multidose therapy. This failure rate was even higher when controlling for baseline ß-hCG values. Safeguards and Counseling During methotrexate therapy, a patient should be examined by a single examiner only once. The physician and the patient must recognize that transient pain (“separating” or “tearing pain”) is common. Transient pelvic pain from resolution tubal bleeding of the ectopic pregnancy frequently occurs

3 to 7 days after the start of therapy, lasts 4 to 12 hours, and is presumably due to tubal abortion. Perhaps the most difficult aspect of methotrexate therapy is learning to distinguish the transient abdominal pain of successful therapy from that of a rupturing ectopic pregnancy. Isthmic ectopic pregnancies are probably at high risk for rupture, and there is simply no way to identify these in advance. A review of ectopic ruptures after methotrexate therapy revealed a high prevalence of isthmic (47%) ectopic pregnancies that was unexpected. These ectopic pregnancies have an early normal doubling time and rise in ß-hCG titers after methotrexate dosing. Physicians must therefore carefully observe for clinical indications that an operation is necessary ( Table 5.4 and Table 5.5). Thus, surgical intervention is required when pain is worsening and persistent beyond 12 hours. Orthostatic hypotension or a falling hematocrit should lead to immediate surgery. Sometimes it is necessary to hospitalize the patient with pain for observation (usually about 24 hours). In addition, colicky abdominal pain is common during the first 2 or 3 days of methotrexate therapy, and the woman should avoid gas-producing foods, such as leeks and cabbage. Finally, the patient should avoid exposure to the sun, because photosensitivity can be a complication of methotrexate.

TABLE 5.4. Treatment with multiple-dose methotrexate

TABLE 5.5. Methotrexate failure Methotrexate by Direct Injection In 1987, Feichtinger and Kemeter instilled 1 mL (10 mg) of methotrexate into an ectopic gestational sac under transvaginal ultrasonography, and resolution occurred within 2 weeks. Direct injection delivers concentrations of methotrexate to the site of implantation at higher concentrations than those achieved with systemic administration. Less systemic distribution of the drug should decrease the overall toxicity. However, this approach has the substantial disadvantage of requiring laparoscopic or ultrasound needle guidance. Outcomes in 21 studies involving direct injection of methotrexate with either laparoscopic or transvaginal ultrasound guidance are presented in Table 5.3. Between 1989 and 1997, 75.1% of 668 cases of ectopic pregnancy were treated successfully with methotrexate by direct injection, and some patients required more than one injection. Tubal patency and subsequent pregnancy rates were comparable to conservative laparoscopic surgery and systemic methotrexate: 80.2% of the women tested had patent oviducts and, of the women desiring pregnancy, 57.2% had a subsequent intrauterine pregnancy and 5.9% developed a recurrent ectopic pregnancy. Randomized, controlled trials have demonstrated successful treatment with methotrexate by direct injection in 86.2% of the patients. Again, successful therapy included some patients who received more than one injection. Tubal patency was present in 85.1% of the women evaluated, and intrauterine pregnancy occurred in 73.1% of the women desiring subsequent fertility. One of the earlier randomized, controlled trials was discontinued because three of seven patients assigned to laparoscopic injection of methotrexate required additional laparoscopic surgery. Even with the higher success rate in the randomized trials, this technique is more cumbersome than systemic methotrexate. Side Effects High doses of methotrexate can cause bone marrow suppression, hepatotoxicity, stomatitis, pulmonary fibrosis, alopecia, and photosensitivity. These side effects are infrequent in the short treatment schedules used in ectopic pregnancy and can be attenuated by the administration of leucovorin (citrovorum factor). The side effects of methotrexate resolve within 3 to 4 days after the therapy is discontinued. Impaired liver function is the most common side effect. Other side effects include stomatitis, gastritis and enteritis, and bone marrow suppression. Local therapy by direct injection of methotrexate into the ectopic gestation resulted in fewer side effects, likely because of less systemic absorption. Even with local injection, impaired liver function tests, gastritis and enteritis, and bone marrow suppression can occur. Additional case reports exist in the literature. Cases of life-threatening neutropenia and febrile morbidity can occur after single or multidose intramuscular methotrexate, requiring hospitalization. Cases of transient pneumonitis from methotrexate therapy for ectopic pregnancy have been observed. Reversible alopecia (a loss of 33–50% of the scalp hair) on two separate occasions following single-dose therapy for an ectopic pregnancy has also been reported. Rarely, hematosalpinx and pelvic hematoceles have been noted as late sequelae of methotrexate following the normalization of ß-hCG levels. These patients have pelvic pain, abnormal bleeding, and a pelvic mass, requiring surgical intervention, 3 to 5 months after therapy. Methotrexate remains the first choice before surgical therapy. Direct Injection of Cytotoxic Agents Prostaglandins, hyperosmolar glucose, potassium chloride, and saline by direct injection have been tried as therapeutic alternatives to methotrexate. The limited experience with prostaglandins and hypertonic glucose, poor success rates, and the need for laparoscopic or transvaginal aspiration makes these treatment alternatives unattractive. Evidence-based Recommendation Multiple-dose systemic methotrexate is the medical treatment of first choice for ectopic pregnancy. (Strength of recommendation: A.) Surgical Treatment Since the first successful salpingectomy performed by Tait in 1884, ectopic pregnancies traditionally have been treated by salpingectomy during laparotomy. Historically, ectopic pregnancies were diagnosed at the time of emergency surgery, when concern for the patient's life superseded any concerns for her future fertility. It was not until 1953, when Stromme performed the first conservative procedure (salpingostomy) for ectopic pregnancy, that subsequent successful pregnancy outcomes were reported, confirming the potential for fertility preservation after salpingostomy. Subsequently, other conservative surgical procedures, such as manual fimbrial expression and segmental tubal resection with later reanastomosis, have been performed to preserve fertility. These surgical techniques have been modified for the laparoscope. Ruptured Ectopic Pregnancy Early diagnosis and treatment of ectopic pregnancy avoids rupture in most cases. In the 1970s, 13.5% to 17.8% of patients with ectopic pregnancies arrived for treatment in hypovolemic shock, whereas in the early 1980s, only 4.4% of patients arrived in this condition. Today, either laparotomy or laparoscopy with salpingectomy is the first choice for rupture. Once contraindicated over concern of decreased venous return from intraperitoneal insufflation, laparoscopic salpingectomy is successful in patients in hypovolemic shock. Nearly all patients in hypovolemic shock require blood transfusions. In the hands of a skilled laparoscopist, with adequate cardiac monitoring and anesthesia, laparoscopic salpingectomy is an acceptable alternative to laparotomy. At present, it is the surgeon's choice of laparoscopy or laparotomy for ruptured ectopic pregnancy. Stable Ectopic Pregnancy If methotrexate is contraindicated, laparoscopic salpingostomy is the first surgical choice. Salpingectomy can be performed either during laparotomy or laparoscopy using cautery or sutures (laparoscopic or endo-loops). Subsequent to salpingostomy, 53% of patients have intrauterine pregnancies, compared with 49.3% after salpingectomy. Recurrent ectopic pregnancy rates were slightly higher after conservative surgery, 14.8% compared with 9.9%. Laparoscopic salpingectomy is preferred over salpingostomy in cases of uncontrollable bleeding not resolving with conservative measures when extensive tubal damage is present, if the ectopic pregnancy is in the same tube, and if sterilization is desired. The recommended conservative surgical procedure for an ampullary ectopic pregnancy is linear salpingostomy, because the ectopic nidation typically is located between the endosalpinx and serosa rather than in the tubal lumen. A linear salpingostomy is created through a longitudinal incision by electrocautery, scissors, or laser over the bulging antimesenteric border of the fallopian tube. The products of conception are removed with forceps or gentle flushing or suction. After maintaining hemostasis, the incision is closed primarily or left to heal by secondary intention. Isthmic pregnancies routinely are treated with segmental excision, followed by intraoperative or delayed microsurgical anastomosis. The tubal lumen is narrower and the muscularis is thicker in the isthmus than in the ampulla, predisposing the isthmus to greater damage after salpingostomy and greater rates of proximal tubal obstruction. Manual fimbrial expression, also known as milking, should be used only when the trophoblastic tissue is already aborting spontaneously through the fimbriae. Laparoscopy has several advantages over laparotomy, including less blood loss, decreased need for analgesia, and improved postoperative recovery, as measured by shorter hospitalizations and length of time to resume normal activity. In addition, cost analysis has demonstrated significant savings in randomized trials. Laparoscopic salpingostomy and fimbrial expression have been evaluated in 32 studies and are presented in Table 5.1. Of the 1,614 patients treated between 1980 and 1997, treatment was successful in 93.4% (required no additional therapy). Of the patients evaluated for tubal patency using either hysterosalpingography or laparoscopy, 77.8% had patent tubes. Of the women desiring subsequent fertility, 56.6% had an intrauterine pregnancy and 13.4% developed another ectopic pregnancy. When evaluating subsequent fertility, intrauterine pregnancy rates are comparable for laparoscopy and laparotomy, as are rates of recurrent ectopic pregnancy.

PERSISTENT ECTOPIC PREGNANCY FOLLOWING SALPINGOSTOMY Persistent ectopic pregnancy is diagnosed by a plateauing or rising ß-hCG concentration following conservative surgical therapy. Although the number of reported cases is small, women with persistent ectopic pregnancies are treated successfully using single-dose systemic methotrexate. The increased rate of persistent ectopic pregnancies has been a criticism of conservative laparoscopic therapy when compared with laparotomy. A decision analysis that compared prophylactic methotrexate with linear salpingostomy against no methotrexate in a group of 1,000 women concluded that prophylactic methotrexate at the time of surgery was preferable if certain conditions are met as follows: (a) the incidence of persistent ectopic pregnancy is greater than 9% with observation alone after salpingostomy, (b) the incidence of persistence is less than 5% when prophylactic methotrexate is given, (c) the probability of ectopic pregnancy rupture is greater than 7.3% with a persistent ectopic pregnancy, and (d) the complication rate associated with prophylactic methotrexate is less than 18%. Because the great majority of

clinical circumstances meet these recommendations, prophylactic methotrexate administration is recommended. Evidence-based Recommendation Due to lower morbidity and equal efficacy, laparoscopic surgery is preferable to laparotomy in the treatment of bleeding or complicated ectopic pregnancy. Salpingectomy by laparotomy is reserved for ectopic ruptures with a hemodynamically unstable patient. (Strength of recommendation: A.)

ECTOPIC PREGNANCY AND ASSISTED REPRODUCTIVE TECHNOLOGY (ART) Incidence The risk of ectopic pregnancy is increased in patients undergoing an ART procedure. This increased risk has been attributed to the cause of infertility for which most patients seek treatment, that is, tubal factor infertility. Information on ectopic pregnancies resulting from ART comes from data obtained from institutions in the United States and Canada reporting to the Society for Assisted Reproductive Technology. The rate of pregnancies that resulted in ectopic pregnancies after in vitro fertilization (IVF) in 1999 was 3%. This included outcome of ART cycles using fresh, nondonor eggs or embryos in approximately 53,000 embryo transfers. This lower percentage likely reflected the trend toward performing salpingectomies when hydrosalpinges are present to improve the success of ART. Location As in naturally occurring ectopic pregnancies, the fallopian tube is the most common site for ectopic pregnancies following IVF. Data obtained from three case-control studies reveal that 82.2% of ectopic pregnancies were tubal. When tubal location was specified, 92.7% were ampullary and 7.3% interstitial. Extratubal ectopic nidations were as follows: 4.6% ovarian or abdominal, 1.5% cervical, and 11.7% heterotopic pregnancies (see Fig. 5.1). Tubal Pathology The most important predisposing factor for ectopic pregnancy in patients undergoing IVF is tubal pathology. Ectopic pregnancies are 4 times higher in patients with tubal factor infertility compared with patients with normal tubes. Hydrosalpinges are associated more commonly with ectopic pregnancy than other types of tubal pathology. Prior tubal reconstructive surgery (salpingostomy) increases the risk of ectopic pregnancy by 10% above that in patients with tubal factor infertility without prior surgery. Thus, it is not surprising that patients with previous pelvic inflammatory disease have a six-fold increase in ectopic pregnancy after IVF. However, a history of prior ectopic pregnancy does not seem as important a risk factor in IVF cycles as in natural cycles. Salpingectomy, particularly with hydrosalpinx, has been shown to decrease risks of ectopic pregnancy while increasing pregnancy rates after IVF. Meta-analysis has demonstrated that the presence of hydrosalpinges decreases the chance for viable pregnancy by approximately 50% when compared with patients with tubal disease but without hydrosalpinges. The implantation rate was also noted to be 50% lower with a higher chance of miscarriage and ectopic gestation. The ultimate conclusion is that when a hydrosalpinx is present there is a decreased pregnancy rate with resultant decreased delivery rate following IVF. In addition, patients who undergo salpingectomy or proximal tubal occlusion prior to oocyte retrieval and transfer are at decreased risk for pelvic infection as well as future ectopic pregnancy. Ovulation Induction Hormone alterations during ovulation induction theoretically alter tubal function. In animal models, estrogen administration results in functional tubal blockage and embryo arrest in the fallopian tube. In humans, steroid hormones alter tubal function and contractility, thus affecting tubal peristalsis. There remains controversy as to whether ovulation-inducing agents, including clomiphene citrate, increase ectopic pregnancy rates. Embryo Transfer Knutzen et al. injected 50 µL of radiopaque fluid in mock embryo transfers and found that the material entered the tubes either partially or totally in 44% of subjects, suggesting misplacement of embryos into the fallopian tubes leads to ectopic pregnancy. Embryo catheter placement was implicated also in the increased risk of ectopic pregnancies, which occurred more frequently in patients who underwent deep fundal transfer versus midcavity placement. Although transfer techniques may increase the chances of embryos reaching the fallopian tubes, ectopic pregnancies may result from tubal pathology preventing the embryos from moving back into the uterus. Heterotopic Pregnancy Heterotopic pregnancies occur in 1% to 3% of pregnancies following an ART procedure and usually are diagnosed incidentally on routine follow-up ultrasonographic studies. This increased prevalence of heterotopic pregnancies following ART may be related to ovarian hyperstimulation and multiple ovum development. Of 111 reported heterotopic pregnancies following ART, 88.3% were tubal, 6.3% cornual, 2.7% abdominal, 1.8% cervical, and 0.9% ovarian. Evidence-based Recommendation Heterotopic and extratubal ectopic pregnancies are more frequent following ART than with natural cycles. Salpingectomy or proximal tubal occlusion of a preexisting hydrosalpinx prior to IVF helps prevent tubal ectopic pregnancies while increasing pregnancy rates following ART. (Strength of recommendation: B.)

EXPECTANT MANAGEMENT Ectopic pregnancies may resolve spontaneously. In a cavalier experiment in 1955, Lund hospitalized 119 women with ectopic pregnancy for observation. All were at least 6 weeks gestation. Some required multiple blood transfusions, and many were hemodynamically unstable. However, 68 resolved without surgery. Twelve additional studies reported in the literature since Lund's study found similar results ( Table 5.3). Of the ectopic pregnancies, 67.2% resolved without surgery. Thus, both conservative medical and surgical therapy overtreats at least 50% of women with ectopic pregnancy. Falling ß-hCG levels under 1,000 mIU/mL have been followed with conservative expectant management. Although patients with an equivocal diagnosis of ectopic pregnancy may be treated in this fashion, there are no data to support expectant management in clinical practice. Evidence-based Recommendation Expectant management of ectopic pregnancy may be considered an appropriate conservative therapy for some patients with low initial (1,000 mIU/mL) and falling hCG levels.

COST ANALYSIS In 1990, total costs for ectopic pregnancies were estimated to be $1.1 billion. Direct costs, expenditures for health care, accounted for 77% of the total costs, and the remainder were incurred as a result of lost wages or household responsibilities not performed due to illness (indirect costs). Direct costs from hospital charges were estimated at $6,079 per case, with hospital accommodations (mean length of stay, 3.47 days) and operating room charges accounting for the majority of the hospital expense, 36% and 40%, respectively. An additional $3,254 for professional fees increased inpatient charges to $9,333, and $149 for postoperative follow-up visits increased the total direct cost to $9,482 per case. Indirect costs for a 28-day disability were estimated at $250.5 million, 67% as a result of lost wages and the remainder from lost household duties. Laparoscopic surgery was less expensive than laparotomy. Surgical management generally has been directed toward laparoscopy. The cost of laparoscopic conservative surgery or salpingectomy was estimated at $2,125 and $1,872, respectively. Costs for conservative treatment or salpingectomy via laparotomy were

estimated to be higher, $3,420 and $3,490, respectively. Systemic methotrexate therapy is even more cost efficient. Charges for patients successfully treated with methotrexate average $1,563 (range, $1,169 to $2,300). Patients requiring laparotomy, due to hemodynamic instability, incur a higher average charge of $8,001 (range, $3,171 to $22,082). The average length of stay following laparotomy was 5.2 days. A study, undertaken to compare the costs of systemic methotrexate with surgery, concluded that there would be a reduction in overall costs if patients were treated without confirmatory laparoscopy when hCG levels were below 3,000 mIU/mL; otherwise, there was not a substantial cost saving over surgery. Because a confirmatory laparoscopy no longer is required for diagnosis, the lower cost for medical therapy is more realistic. Compared with the cost of a laparoscopic salpingostomy, methotrexate results in a 20% decrease in the cost of treatment. Evidence-based Recommendation Systemic methotrexate for unruptured ectopic pregnancy is less expensive than surgery, and direct costs are decreased substantially with methotrexate therapy. In addition to its cost effectiveness, systemic methotrexate does not subject patients to surgery and the complications associated with it. This cost benefit, however, diminishes with higher hCG titers and even disappears with levels greater than 3,000 mIU/mL, because of treatment failures and increased complications. (Strength of recommendation: B.)

RARE TYPES OF ECTOPIC PREGNANCY Abdominal Pregnancy The incidence of abdominal pregnancy is estimated at 1 in 8,000 births and represents 1.4% of all ectopic pregnancies. The prognosis is poor, with an estimated maternal mortality rate of 5.1 per 1,000 cases. The risk of dying from an abdominal pregnancy is 7.7 times higher than from other forms of ectopic pregnancy. The high rate of morbidity and mortality from abdominal pregnancy often results from a delay in diagnosis. Abdominal pregnancies can be categorized as primary or secondary. These ectopic pregnancies may become apparent throughout gestation, from the first trimester to fetal viability. Symptoms may vary from those considered normal for pregnancy to severe abdominal pain, intraabdominal hemorrhage, and hemodynamic instability. Primary abdominal pregnancies are rare and are thought to occur as a result of primary peritoneal implantation. They usually abort early in the first trimester due to hemorrhagic disruption of the implantation site and hemoperitoneum. Secondary abdominal pregnancies occur with reimplantation after a partial tubal abortion or intraligamentary extension following tubal rupture. Historical criteria to distinguish between primary and secondary abdominal pregnancies are moot, because treatment is directed by the clinical picture. Ultrasonography is the diagnostic tool of choice and usually can identify the empty uterus along with the extrauteral products of conception. If the fetus is near viability, hospitalization is recommended. If time permits, bowel preparation, administration of prophylactic antibiotics, and adequate blood replacement should be made available prior to an operative delivery. Unless the placenta is implanted on major vessels or vital structures, it should be removed. Although complications may occur, including sepsis, abscess formation, secondary hemorrhage, intestinal obstruction, wound dehiscence, amniotic fluid cyst formation, hypofibrinogenemia, and preeclampsia, the placenta can be left in place to prevent further hemorrhage at the time of surgery. In contrast to the typical tubal ectopic pregnancy, methotrexate is unlikely to accelerate retained placental absorption, because the trophoblastic cells are no longer actively dividing. Ovarian Pregnancy Ovarian pregnancy, the most common form of abdominal pregnancy, is rare, accounting for less than 3% of all ectopic gestations. Clinical findings are similar to those of tubal ectopic gestations: abdominal pain, amenorrhea, and abnormal vaginal bleeding. In addition, hemodynamic instability as a result of rupture occurs in 30% of patients. Women with ovarian pregnancies are usually young and multiparous, but the factors leading to ovarian pregnancies are not clear. The diagnosis usually is made by the pathologist, because many ovarian pregnancies are mistaken for a ruptured corpus luteum or other ovarian tumors. Only 28% of cases were diagnosed correctly at time of laparotomy. The recommended treatment is cystectomy, wedge resection, or oophorectomy during laparotomy, although laparoscopic removal has been successful. Cornual Pregnancy Cornual or interstitial pregnancy accounts for 4.7% of ectopic gestations and carries a 2.2% maternal mortality. Almost all cases are diagnosed after the patient is symptomatic. The most frequent symptoms are menstrual aberration, abdominal pain, abnormal vaginal bleeding, and shock, resulting from the brisk hemorrhage associated with uterine rupture. Due to myometrial distensibility, rupture is usually delayed, occurring at 9 to 12 weeks gestation. A unique risk factor for interstitial pregnancy is previous salpingectomy, present in about 25% of patients. Only a high index of suspicion and repeated ultrasonographic examination with Doppler flow studies allows early diagnosis. With a timely early diagnosis, alternatives to the traditional cornual resection during laparotomy have been performed successfully. These include laparoscopic cornual resection, systemic methotrexate administration, local injection of methotrexate, potassium chloride injection, and removal by hysteroscopy. Regardless of the initial treatment attempted, if uncontrolled hemorrhage occurs, immediate hysterectomy is warranted. Cervical Pregnancy The incidence of cervical pregnancy ranges from 1 in 2,500 to 1 in 12,422 pregnancies. The most common predisposing factor is a prior dilation and curettage, present in 68.6% of patients. Interestingly, 31% of these were performed for termination of pregnancy. Other predisposing factors implicated in cervical pregnancies are previous cesarean delivery and IVF. The most common initial symptom of cervical pregnancy is painless vaginal bleeding. These ectopics usually are diagnosed incidentally during routine ultrasonography or at the time of surgery for a suspected abortion in progress. In reported cases, 91% of patients sought treatment for vaginal bleeding, and 29.2% had massive bleeding. Not surprisingly, abdominal pain occurred with vaginal bleeding in only 25.8% of cases. The cervix is usually enlarged, globular, or distended. On occasion, it appears cyanotic, hyperemic, and soft in consistency. Sonography and magnetic resonance imaging have improved diagnosis of cervical pregnancy. Up to 81.8% of patients have been diagnosed correctly with ultrasonographic identification of the gestational sac in the cervix below a closed internal cervical os, with trophoblastic invasion into the endocervical tissue. When the patient is hemodynamically stable, conservative therapy commonly is employed. There are no large studies, only several case series. These have shown that use of methotrexate and uterine artery embolization are safe and effective for treatment in the stable patient with a cervical pregnancy. Systemic and local treatment with various agents carries an overall success rate of 81.3%. Unfortunately, massive hemorrhage may occur despite conservative measures, and hysterectomy is warranted. Heterotopic Pregnancy Heterotopic pregnancy is the coexistence of an intrauterine and ectopic gestation. In 1948, the spontaneous heterotopic pregnancy rate was calculated as 1 in 30,000 pregnancies, based on an ectopic pregnancy incidence of 0.37% and dizygous twinning rate of 0.8%. In the 1980s, the calculation rose to 1 in 10,000 due to an increased ectopic pregnancy rate. Today, heterotopic pregnancies occur in 1 in 3,889 to 1 in 6,778 pregnancies. In a review of 66 heterotopic pregnancies by Reece et al., 93.9% were tubal and 6.1% ovarian. Simultaneous existence of intra- and extrauterine pregnancies poses several diagnostic pitfalls. Heterotopic pregnancies are diagnosed in most cases after clinical signs and symptoms develop, and 50% of patients are admitted for emergency surgery following rupture. The delay in diagnosis is secondary to the finding of an intrauterine pregnancy, with the assumption that any symptoms will be self-limited. Similar to tubal ectopic pregnancies, the most common complaint is lower abdominal pain. Routine ultrasonography detects only about 50% of tubal heterotopic

pregnancies, and the remainder are diagnosed during laparoscopy or laparotomy when patients become symptomatic. Serial levels of the ß subunit of hCG are not helpful due to the effect of the intrauterine pregnancy. If patients are hemodynamically unstable, exploratory laparotomy is warranted. If the diagnosis is suspected or the patient is symptomatic but hemodynamically stable, laparoscopy can be performed. Expectant management is not recommended, because ß-hCG levels cannot be monitored adequately. Systemic methotrexate is contraindicated if a viable intrauterine pregnancy is present and desired. Local injection of methotrexate with potassium chloride has been noted successful in a small case series.

SUMMARY POINTS In most circumstances, ectopic pregnancy can be diagnosed before symptoms develop and treated definitively with few complications. Quantitative ß-hCG testing, ultrasonography, and curettage allow early diagnosis of ectopic pregnancy and use of medical therapy as the initial therapy option. Conservative surgical therapy and medical therapy for ectopic pregnancy are comparable in terms of success rates and subsequent fertility. Medical therapy is the preferred choice because of the freedom from surgical complications and lower cost. Surgery is the treatment of choice for hemorrhage, medical failures, neglected cases, and when medical therapy is contraindicated. Multiple-dose methotrexate is preferable to single-dose methotrexate, direct injection, or tubal cannulation and is the first choice for unruptured, uncomplicated ectopic pregnancy. Laparoscopic salpingostomy or salpingectomy is favored for cases of intraabdominal hemorrhage, medical failure, neglected cases, and complex cases when medical therapy is contraindicated. Prophylactic postoperative systemic methotrexate (a single dose) can prevent virtually all cases of persistent ectopic pregnancy following salpingostomy. Salpingectomy prior to IVF decreases ectopic pregnancy incidence while increasing pregnancy rates in select patients with preexisting tubal disease. REFERENCES Incidence Carson SA, Buster JE. Ectopic pregnancy. N Engl J Med 1993;329:1174–1181. Chow W, Daling JR, Cates W, et al. Epidemiology of ectopic pregnancy. Epidemiol Rev 1987;9:70–94. National Center for Health Statistics. Advance report of final mortality statistics, 1992. Hyattsville, MD: US Department of Health and Human Services, Public Health Service, CDC. Mon Vital Stat Rep 1994;43[Suppl].

Pathogenesis Bone NL, Greene RR. Histological study of uterine tube with tubal pregnancy: a search for evidence of previous injection. Am J Obstet Gynecol 1961;82:1166. Budowick M, Johnson TRB, Genadry R, et al. The histopathology of the developing tubal ectopic pregnancy. Fertil Steril 1980;34:169–171. Westrom L. Effect of acute pelvic inflammatory disease on fertility. Am J Obstet Gynecol 1975;121:707–713.

Risk Factors Ankum WM, Mol BWJ, Van der Veen F, et al. Risk factors for ectopic pregnancy: a meta-analysis. JAMA 1996;65:1093–1099. Brenner PF, Benedetti T, Mishell DR. Ectopic pregnancy following tubal sterilization surgery. Obstet Gynecol 1977;49:323–324. Dubuisson JB, Aubriot FX, Mathieu L, et al. Risk factors for ectopic pregnancy in 556 pregnancies after in vitro fertilization: implications for preventive management. Fertil Steril 1991;56:686–690. Kranz SG, Gray RH, Damewood MD, et al. Time trends in risk factors and clinical outcome of ectopic pregnancy. Fertil Steril 1990;54:42–46. Majmudar B, Henderson PH, Semple E. Salpingitis isthmica nodosa: a high-risk factor for tubal pregnancy. Obstet Gynecol 1983;62:73–78. Mol BWJ, Ankum WM, Bossuyt PMM, et al. Contraception and the risk of ectopic pregnancy: a meta-analysis. Contraception 1995;52:337–341.

Signs and Symptoms Barnhart K, Mennuti M, Benjamin I, et al. Prompt diagnosis of ectopic pregnancy in an emergency department setting. Obstet Gynecol 1994;84:1010–1015. Breen JL. A 21-year survey of 654 ectopic pregnancies. Am J Obstet Gynecol 1970;106:1004–1019. Grimes DA. The morbidity and mortality of pregnancy: still risky business. Am J Obstet Gynecol 1995;170:1489–1494.

Diagnosis Buster JE, Carson SA. Ectopic pregnancy: new advances in diagnosis and treatment. Curr Opin Obstet Gynecol 1995;7:168–176. Garcia CR, Barnhart KT. Diagnosing ectopic pregnancy: decision analysis comparing six strategies. Obstet Gynecol 2001;97:464–70. Goldstein SR, Snyder JR, Watson C, et al. Vaginal sonography versus serum human chorionic gonadotropin in early detection of pregnancy. Am J Obstet Gynecol 1988;158:608–612. Kim DS, Chung SR, Park MI, et al. Comparative review of diagnostic accuracy in tubal pregnancy: a 14-year survey of 1040 cases. Obstet Gynecol 1987;70:547–554. Lindblom B, Hahlin M, Sjoblom P. Serial human chorionic gonadotropin determinations by fluoroimmunoassay for differentiation between intrauterine and ectopic gestation. Am J Obstet Gynecol 1989;161:397–400. Shalev E, Yarom I, Bustan M, et al. Transvaginal sonography as the ultimate diagnostic tool for the management of ectopic pregnancy: experience with 840 cases. Fertil Steril 1998;69:62–65. Stovall TG, Ling FW, Cope BJ, et al. Preventing ruptured ectopic pregnancy with a single serum progesterone. Am J Obstet Gynecol 1989;160:1425–1431. Timor-Tritsch IE, Meh MN, Peisner DB, et al. The use of transvaginal ultrasonography in the diagnosis of ectopic pregnancy. Am J Obstet Gynecol 1989;161:157–161.

Treatment for Ectopic Pregnancy Medical Management Buster JE, Heard MJ. Current issues in medical management of ectopic pregnancy. Curr Opin Obstet Gynecol 2000;12:525–527. Fernandez H, Pauthier S, Sitbon D, et al. Role of conservative therapy and medical treatment in ectopic pregnancy: literature review and clinical trial comparing medical treatment and conservative laparoscopic treatment. Contracept Fertil Sex 1996;24:297–302. Hajenius PJ, Mol BW, Bossuyt PM, et al. Interventions for tubal ectopic pregnancy. Cochrane Database Syst Rev 2000;(2):CD000324. Kooi S, Kock HC. A review of the literature on nonsurgical treatment in tubal pregnancies. Obstet Gynecol Surv 1992;47:739–749. Medical Management of Tubal Pregnancy. ACOG Practice Bulletin, Clinical Management Guidelines for Obstetrician-Gynecologists December 1998;3:410–416. Shalev E, Peleg D, Bustan M, et al. Limited role for intratubal methotrexate treatment of ectopic pregnancy. Fertil Steril 1995;63:20–24.

Stovall TG, Ling FW, Gray LA, et al. Methotrexate treatment of unruptured ectopic pregnancy: a report of 100 cases. Obstet Gynecol 1991;77:749–753. Stovall TG, Ling FW. Single-dose methotrexate: an expanded clinical trial. Am J Obstet Gynecol 1993;168:759–772.

Surgical Treatment Brumsted J, Kessler C, Gibson C, et al. A comparison of laparoscopy and laparotomy for the treatment of ectopic pregnancy. Obstet Gynecol 1988;71:889–892. DeCherney AH, Kase N. The conservative surgical management of unruptured ectopic pregnancy. Obstet Gynecol 1979;54:451–455. Gracia C, Brown H, Barnhart K. Prophylactic methotrexate after linear salpingostomy: a decision analysis. Fertil Steril 2001;76:1191–1195. Lundorff P, Thorburn J, Hahlin M, et al. Laparoscopic surgery in ectopic pregnancy: a randomized trial versus laparotomy. Acta Obstet Gynecol Scand 1991;70:343–348. Murphy AA, Nager CW, Wujek JJ, et al. Operative laparoscopy versus laparotomy for the management of ectopic pregnancy: a prospective trial. Fertil Steril 1992;57:1180–1185. Pouly JL, Mahnes H, Mage G, et al. Conservative laparoscopic treatment of 321 ectopic pregnancies. Fertil Steril 1986;4:1093–1097. Seifer DB, Guttmann JN, Grant WD, et al. Comparison of persistent ectopic pregnancy after laparoscopic salpingostomy versus salpingostomy at laparotomy for ectopic pregnancy. Obstet Gynecol 1993;81:378–382. Silva PD, Schaper AM, Rooney B. Reproductive outcome after 143 laparoscopic procedures for ectopic pregnancy. Obstet Gynecol 1993;81:710–715. Vermesh M, Silva PD, Rosen GF, et al. Management of unruptured ectopic gestation by linear salpingostomy: a prospective, randomized clinical trial of laparoscopy versus laparotomy. Obstet Gynecol 1989;73:400–404.

Ectopic Pregnancy and Assisted Reproductive Technology Abusheikha N, Salha O, Brinsden P. Extra-uterine pregnancy. Assisted reproductive technology in the United States and Canada: 1994 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil Steril 1996;66:697–705. Camus E, Poncelet C, Goffinet F, et al. Pregnancy rates after in-vitro fertilization in cases of tubal infertility with and without hydrosalpinx: a meta-analysis of published comparative studies. Hum Reprod 1999;14:1243–1249. Abusheikha N, Salha O, Brinsden P. Extra-uterine pregnancy following assisted conception treatment. Hum Reprod Update 2000;6:80–92. Marcus SF, Brinsden PR. Analysis of the incidence and risk factors associated with ectopic pregnancy following in vitro fertilization and embryo transfer. Hum Reprod 1995;10:199–203. Martinez F, Trounson A. An analysis of factors associated with ectopic pregnancy in a human in vitro fertilization program. Fertil Steril 1986:45:79–87.

Expectant Management Fernandez H, Lelaidier C, Baton C, et al. Return of reproductive performance after expectant management and local treatment for ectopic pregnancy. Hum Reprod 1991;6:1474–1477. Garcia AJ, Aubert JM, Sama J, et al. Expectant management of presumed ectopic pregnancies. Fertil Steril 1987;48:395–400. Shalev E, Peleg D, Tsabari A, et al. Spontaneous resolution of ectopic tubal pregnancy: natural history. Fertil Steril 1995;63:15–19. Ylöstalo P, Cacciatore B, Korhonen J, et al. Expectant management of ectopic pregnancy. Eur J Obstet Gynecol Reprod Biol 1993;49:83–84.

Cost Analysis Alexander JM, Rouse DJ, Varner E, et al. Treatment of the small unruptured ectopic pregnancy: a cost analysis of methotrexate versus laparoscopy. Obstet Gynecol 1996;88:123–127. Mol BW, Hajenius PJ, Engelsbel S, et. al. Treatment of tubal pregnancy in the Netherlands: an economic comparison of systemic methotrexate administration and laparoscopic salpingostomy. Am J Obstet Gynecol 1999;181:945–951. Mol BWJ, Hajenius PJ, Engelsbel S, et al. An economic evaluation of laparoscopy and open surgery in the treatment of tubal pregnancy. Acta Obstet Gynecol Scand 1997;76:1–5. Stovall TG, Bradham DD, Ling FW, et al. Cost of treatment of ectopic pregnancy: single-dose methotrexate versus surgical treatment. J Womens Health 1994;3:445–450. Washington AE, Katz P. Ectopic pregnancy in the United States: economic consequences and payment source trends. Obstet Gynecol 1993;81:287–292. Yao M, Tulandi T, Kaplow M, et al. A comparison of methotrexate versus laparoscopic surgery for treatment of ectopic pregnancy: a cost analysis. Hum Reprod 1996;11:2762–2766.

Rare Types of Ectopic Pregnancy Atrash HK, Friede A, Hogue CJR. Abdominal pregnancy in the United States: frequency and maternal mortality. Obstet Gynecol 1987;69:333–337. Goldenberg M, Bider D, Oelsner G, et al. Treatment of interstitial pregnancy with methotrexate via hysteroscopy. Fertil Steril 1992;58:1234–1236. Grimes HG, Nosal RA, Gallagher JC. Ovarian pregnancy: a series of 24 cases. Obstet Gynecol 1983;61:174–180. Hallatt JG. Primary ovarian pregnancy: a report of twenty-five cases. Am J Obstet Gynecol 1982;143:55–60. Reece EA, Petrie RH, Sirmans MF, et al. Combined intrauterine and extrauterine gestations: a review. Am J Obstet Gynecol 1983;146:323–330. Timor-Tritsch IE, Monteagudo A, Mandeville EO, et al. Successful management of viable cervical pregnancy by local injection of methotrexate. Am J Obstet Gynecol 1994;170:737–739. Ushakov FB, Elchalal U, Aceman PJ, et al. Cervical pregnancy: past and future. Obstet Gynecol 1996;52:45–59.

Chapter 6 Genetics and Prenatal Diagnosis Danforth’s Obstetrics and Gynecology

Chapter 6 Kenneth Ward

Genetics and Prenatal Diagnosis

PATTERNS OF INHERITANCE Single Gene Disorders Polygenic, Multifactorial Disorders Cytogenetic Disorders Parent of Origin Effects THE MOLECULAR GENETICS REVOLUTION Deoxyribonucleic Acid The Genome Project GENETIC EVALUATION Genetic History and Physical Examination Genetic Counseling Pregnancy Termination Laboratory Screening PRENATAL DIAGNOSIS Maternal Serum Screening Fetal Imaging Fetal Sampling for Prenatal Diagnosis Prenatal Treatment and Gene Therapy GENETICS IN GYNECOLOGIC DISORDERS TRENDS SUMMARY POINTS REFERENCES

Genetic screening has become an important component of prenatal care. Recent discoveries have expanded the indications for cytogenetic or molecular genetic tests performed on chorionic villi, amniocytes, and fetal blood. Soon first-trimester DNA-based prenatal diagnosis will become possible for thousands of additional conditions. New technologies such as flow cytometry, fluorescent in situ hybridization, and DNA amplification with the polymerase chain reaction (PCR) enable testing of a single cell for chromosomal or Mendelian problems. This approach already has allowed preimplantation testing of embryos and minimally invasive diagnosis using the small population of fetal cells in the maternal circulation. Over the next decade, gene therapy will be applied prenatally for the treatment of inborn errors and postnatally for the treatment of genetic diseases. This will provide additional impetus for genetic evaluation. Genetic principles are central to understanding the pathophysiology of many of the conditions for which obstetrician-gynecologists provide care ( Table 6.1). More than 20 million Americans have a diagnosed genetic disease. One percent of all newborns have a recognizable Mendelian disorder, 0.5% have a chromosomal syndrome, and many more have a polygenic, multifactorial disorder. Most pregnancy losses and most congenital anomalies have genetic causes. Genes also play an important role in common gynecologic disorders such as leiomyomata, endometriosis, gynecologic cancers, and infertility. Genetic variants are responsible for tendencies to have multiple gestations, preeclampsia, gestational diabetes, and other pregnancy complications. Susceptibilities to infectious or teratogenic agents are determined genetically. Our rapidly expanded knowledge of the human genome will revolutionize obstetric and gynecologic practice.

TABLE 6.1. The burden of genetic disease

Recognizing that a condition is genetic enables us to find the gene responsible for the illness, which can lead to improved means of classification, diagnosis, prevention, and treatment. Correlation of the phenotype with the genotype often provides specific predictive insights. Because any DNA test can be performed prenatally, discovery of the gene that causes a particular disease can give at-risk couples the necessary information to prepare for having an affected child, to consider prenatal therapy if available, or to choose to end the pregnancy. Patients often have genetic illnesses that will affect pregnancy or their gynecologic care, illnesses that may remain undiagnosed unless physicians are thorough in their evaluation of unusual signs or symptoms. For example, a new obstetric patient has a wasted facial appearance, generalized weakness, and difficulty releasing her grip when shaking the physician's hand. Hopefully, this patient has already been evaluated and correctly diagnosed. If not, it is the obstetrician's responsibility to refer a patient with unusual signs and symptoms for evaluation. In this case, the patient has the classic signs of myotonic dystrophy, an autosomal dominant disorder. Formerly, myotonic dystrophy was a difficult diagnosis to establish in some patients, but the diagnosis can be made easily using DNA analysis of her blood. The physician and this woman need to know that she is at significant risk for developing polyhydramnios, which could result in preterm labor; her fetus may have positional deformities and is at risk for a severe, often fatal, neonatal form of myotonia; her labor is likely to be prolonged, and she may be unable to push during the second stage. If she requires cesarean section, she may have undiagnosed cardiac problems that place her at higher risk for anesthesia. If failure to recognize these risks results in a bad outcome, a lawsuit claiming negligence might be brought.

PATTERNS OF INHERITANCE Single Gene Disorders The observation that “like begets like” has been stated throughout recorded history, but current theories describing how genetic traits and illnesses are inherited are just over 100 years old. In the late 1800s, Mendel described how individual genetic traits were passed on from generation to generation. Single gene disorders (i.e., mendelian disorders) are conditions caused by a mutation at a single site in the DNA and inherited in the proportions predicted by Mendel's laws. These disorders can be dominant conditions in which the phenotype is expressed even when only one chromosome of a pair has a defect, or recessive conditions which are expressed only if the defect exists on both chromosomes. Classically, these different modes of inheritance are revealed by pedigree analysis ( Fig. 6.1 and Fig. 6.2).

FIG. 6.1. Symbols used to draw a pedigree.

FIG. 6.2. Patterns of inheritance.

As we learn more about the tremendous variation that occurs at every locus, the distinctions between dominant and recessive conditions have become blurred. Dominance and recessiveness are attributes of the phenotype, not attributes of the gene or allele. They are empirical terms, and they depend on the sensitivity of the method used to describe the phenotype. The researchers must specify particular phenotypic features when describing inheritance. For instance, sickle cell anemia is a recessive disease only if the full-blown disease is considered, but it is a codominant condition if the hemoglobin is being analyzed by electrophoresis. The ABO blood group is an example of a trait in which both codominant and recessive inheritance is seen. The retinoblastoma gene is a recessive tumor suppressor gene at the cellular level, but abnormalities in the gene are responsible for the autosomal dominant tendency to develop retinoblastomas and osteosarcomas. In autosomal dominant conditions, the disease is expressed in persons who are heterozygous for the disease-causing mutation. McKusick's catalog of mendelian disorders describes more than 3,000 dominant conditions. Marfan syndrome, myotonic dystrophy, neurofibromatosis, achondroplasia, and Huntington disease are examples of autosomal dominant disorders. The probability of an affected person transmitting the abnormal gene to the children is 50% with each pregnancy. Typically, autosomal dominant conditions have less than 100% penetrance, and fewer than 50% of the offspring show signs of the disorder. Male and female offspring usually are affected with equal frequency and severity. The trait passes through one parental line only, and father-to-son transmission can occur. For highly penetrant, autosomal dominant conditions, the gene is expressed in each generation (i.e., vertical transmission). New mutations are relatively common and, on average, paternal age is advanced when isolated, sporadic, or new mutation cases appear. Autosomal dominant phenotypes often involve isolated or multiple structural defects. They can be extremely variable and the onset of clinical features is often age dependent. Dominant disorders tend to be less severe than recessive diseases, but they are usually lethal in the rare persons who are homozygous for a dominant disease. Autosomal recessive conditions are expressed only in persons in whom both versions (i.e., alleles) of the involved gene are abnormal. More than 1,500 autosomal recessive conditions have been described. Cystic fibrosis (CF), sickle cell anemia, Tay-Sachs disease, and phenylketonuria are examples of autosomal recessive disorders. Male and female offspring are affected with equal frequency and severity. Each parent is a heterozygous carrier, and abnormal genes are inherited from both parents. Each offspring of two carrier parents has a 25% chance of being affected, a 50% chance of being a carrier, and a 25% chance of being neither a carrier nor affected. If the recessive phenotype is extremely rare, consanguinity usually is found in the pedigree. Affected persons rarely have affected children; autosomal recessive inheritance shows a “horizontal” pattern in a pedigree, with typically only a single generation of siblings affected. Affected persons who mate with unaffected persons who are not carriers have only unaffected, carrier offspring. Most autosomal recessive phenotypes are biochemical or enzymatic in nature, and they tend to be less variable and more severe than dominant conditions. X-linked inheritance occurs when a trait is carried on the X chromosome. Boys are hemizygous for X chromosome genes, but girls can be homozygous or heterozygous. Of the 300 X-linked recessive diseases that are recognized, the hemophilias and Duchenne muscular dystrophy are the best known. Characteristics of X-linked recessive inheritance include a higher incidence of the disorder in male than in female offspring. The mutant gene or disease never is transmitted directly from father to son, and all the daughters of an affected man are carriers. The trait is transmitted through carrier females, and affected males in the same kindred are related to one another through the females. X-linked dominant diseases are much rarer; examples include Alport syndrome, vitamin D-resistant rickets, and incontinentia pigmenti. They appear twice as often in female as in male offspring. All daughters of an affected man have the disorder, but no sons are affected. Heterozygous affected women transmit the mutant allele at a rate of 50% to progeny of both sexes. If the affected woman is homozygous, all of her children will be affected. Y-linked or holandric inheritance occurs when a trait is carried on the Y chromosome. Only male offspring are affected, and there is only male-to-male transmission. No known disease genes are inherited in this fashion, but genes for gender determination, tooth size, and height occur on the Y chromosome. Mitochondrial inheritance is a more recently described mode of inheritance, causing traits and disorders that are inherited through the mitochondrial chromosome. Unique patterns are seen in affected families. Mitochondria are inherited exclusively with the cytoplasm of the egg; a woman who carries a disease will pass the disease to 100% of her offspring. Male carriers will pass the disorder to none of their offspring. Leber optic atrophy and certain rare myopathies are inherited in this fashion. Polygenic, Multifactorial Disorders Multifactorial or polygenic inheritance is the most common form of inheritance. Even in the classic mendelian disorders described above, there can be tremendous quantitative and qualitative differences in the phenotype among persons who have the same allele or the same genetic mutation. This variability can be evident as nonpenetrance of certain features (or the entire phenotype) and as differences in the severity of features, the frequency of cyclic or episodic events, or the age of onset of the first clinical sign of the disorder. Genetic variability can be caused by the underlying genetic background of the affected person, including gender influences and limitations. The phenotype may be influenced further by maternal factors such as cytoplasmic inheritance, the intrauterine environment, or imprinting. X-linked disorders can be altered by variations in X inactivation or lionization. Each genotype undergoes subtle changes through somatic mutation, gene amplification, or transpositions and positional effects over time. Exogenous factors such as the environment, teratogens, medical intervention, and chance also influence variability. Most congenital anomalies reflect multifactorial inheritance ( Table 6.2). A common error some obstetricians make is to counsel a patient that rare conditions will not occur repetitively in her family. If the birth defect in question has a strong genetic component or if there is an identifiable environmental or teratogenic component which would recur in subsequent pregnancy, the risks may remain high for that patient ( Table 6.3). The rates may be even higher if a mendelian or chromosomal condition has gone unrecognized in the affected child. Before counseling patients about the recurrence risk of any birth defect, it is pertinent to review which syndromes are associated with that birth defect and ask whether any member of the family has those syndromes. When this requires skill or knowledge beyond the usual expertise of an obstetrician-gynecologist, referral to a medical geneticist is appropriate.

TABLE 6.2. Types of birth defects

TABLE 6.3. Empiric recurrence risks for common congenital anomalies

Multifactorial inheritance usually works according to a threshold model. Several factors must collaborate to cause a bodily function to go awry, and only after these factors reach some critical point is the phenotypic effect seen. Many different factors can affect the observed recurrence risk. The intrinsic heritability “or geneticness” of the condition is frequently the most important factor. This usually is determined by examining whether monozygotic twins are concordant for a particular condition relative to dizygotic twins. For instance, neonatal seizures show a very high heritability rate, with 85% to 90% concordance in monozygotic twins verses 10% to 15% concordance in dizygotic twins. The population incidence of the condition is another important variable. The recurrence risk is higher for common disorders or within populations with a high incidence of the disorder. In disorders with a relatively high heritability, the recurrence risk of the disorder approximates the square root of the population incidence. There can be marked variation in the population frequency of different disorders and different ethnic groups. For instance, cleft lip occurs commonly in Native Americans, but African Americans have a nine-fold lower incidence of cleft lip than the general population. If the incidence of a congenital anomaly shows a sex bias, the recurrence risk is higher in the offspring (and other relatives) if the parent is of the less frequently affected sex. For example, pyloric stenosis affects 5 times as many male as female offspring, and empirical data show that there is a 25% chance of producing an affected child if the mother had pyloric stenosis at birth and only a 4% risk if the father is the affected parent. Similarly, the recurrence risk is higher when the gender of the affected child is the less frequently affected. Again, the risk of having another child with pyloric stenosis is 3.2% if the first-affected sibling is a boy and 6.5% if the affected sibling is a girl. Hirschsprung disease, clubfoot, and cleft lip are examples of anomalies which are more common in male infants. Cleft palate, anencephaly, hip dysplasia, and scoliosis are more common in female infants. The number of affected individuals in a kindred can affect the recurrence risk. The greater the number of family members who have already been affected with a multifactorial condition, the more likely it is that the genetic background is favorable for expression of this condition. After a couple has one affected child with cleft lip and palate, there is a 4% empirical recurrence risk. After two affected children, the risk rises to 10%. Consanguinity also increases the risk of recurrence because of the greater likelihood of deleterious genes being shared, but a more distant relationship from an affected person decreases the risk of recurrence. The severity of the disorder often predicts the recurrence risk. An illustration of this is found in Hirschsprung disease. The recurrence risk is proportional to the length of the aganglionic segment of the colon. Neural tube defects are the most notable exception to this rule, because the recurrence risk for any neural tube defect appears to be the same whether the first-affected child had anencephaly or a small spina bifida lesion. Cytogenetic Disorders Cytogenetic disorders are changes in the genome visible under a light microscope. These gross lesions involve the loss or duplication of a large number of genes; multiple malformations and dysfunctions usually are observed clinically. Diagnostic clues for a cytogenetic disorder can range from subtle dysmorphic features to major structural malformations, particularly craniofacial, skeletal, cardiac, and genitourinary malformations. No individual anomaly is pathognomonic for a particular chromosomal syndrome, rather it is the pattern that can be distinctive. There is tremendous overlap between patterns and because nonchromosomal syndromes can mimic chromosomal abnormalities, obtaining a karyotype is always necessary to confirm the diagnosis. Cytogenetic disorders usually are associated with some degree of mental retardation and growth deficiency. Most have an increased rate of perinatal loss and premature mortality of live-born neonates. The rate of chromosome abnormalities is at least 40% to 60% in first-trimester abortuses, and the rates of abnormalities also are elevated in fetal deaths and preterm and post-term deliveries ( Table 6.4 and Table 6.5). About 1 in 160 babies is born with a genetic defect detectable by ordinary cytogenetic means ( Table 6.6).

TABLE 6.4. Incidence of chromosomal aberrations in pregnancy losses at various gestational ages

TABLE 6.5. Types of chromosomal abnormalities in spontaneous abortuses

TABLE 6.6. Incidence of chromosomal aberrations seen in newborn surveys

Cytogenetic studies have been used clinically for approximately 40 years. In the late 1950s, it was determined that humans have 46 chromosomes and that many of the recognized birth defect syndromes, such as Down syndrome, Turner syndrome, and Klinefelter syndrome, have abnormalities of chromosome number or structure. Normally, the nucleus of most human cells contains two sets of chromosomes, with one set contributed by each parent. Each set has 22 autosomes and either an X or a Y sex chromosome ( Fig. 6.3).

FIG. 6.3. Normal human karyotype (46,XY).

Metaphase chromosome preparations can be prepared from any cell undergoing mitosis. Typically, in order to obtain adequate numbers of cells, mitosis is induced artificially using a mitogenic chemical such as phytohemagglutinin. The cells are then incubated in a dilute solution of an agent that poisons the mitotic spindle. The chromosomes are swollen using a hypotonic salt solution, fixed on a slide, and dried for staining. The stained chromosomes can be observed by light microscopy. The dyes used to stain chromosome preparations reveal patterns of light and dark bands that reflect regional variations in the molecular composition of each chromosome. Giemsa is the most commonly used dye. Q-banding is a fluorescence technique that gives results similar to the G-banding (i.e., Giemsa banding), while R-banding (i.e., reverse banding) gives a pattern opposite to that with G- or Q-banding. T-banding specifically stains the telemetric regions of chromosomes, which can help to screen for missing regions at the ends of the chromosomes, and C-banding primarily stains the centromeric region of the chromosomes. Differences in the size of the chromosomes, the banding pattern, and the centromere position allow the 24 chromosomes to be differentiated from each other in an analysis called a karyotype. The most common features looked for on a karyotype include aneuploidy (i.e., abnormal number of chromosomes) or structural chromosome abnormalities such as deletions, inversions, insertions, or translocations ( Table 6.7) ( Fig. 6.4).

TABLE 6.7. Chromosomal nomenclature

FIG. 6.4. Types of chromosomal abnormalities. A, terminal deletion; B, interstitial deletion; C, paracentric inversion; D, pericentric inversion; E, translocation (additional material from a different chromosome); F, isochromosome; G, dicentric chromosome; H, ring chromosome. The arrows indicate the sites of chromosomal breaks where rearrangement occurs.

Because the technology to study cytogenetic disorders is well established, there is greater clinical experience with this than with other types of genetic testing. There are several well-defined indications for obtaining a fetal karyotype analysis ( Table 6.8). Pregnant women who are 35 years or older routinely are offered a fetal karyotype analysis, because trisomy tends to occur more commonly with advancing maternal age ( Table 6.9). Other indications for obtaining a fetal karyotype include having a previous child with an abnormal karyotype, parental chromosomal rearrangements, unexplained intrauterine growth retardation, and an abnormally low level of maternal serum a-fetoprotein. In addition, many fetal anomalies associated with karyotypic abnormalities can now be visualized using high-resolution ultrasonography ( Table 6.10). A fetal structural abnormality detected by ultrasonography is another frequent indication for obtaining a fetal karyotype.

TABLE 6.8. Indications for a fetal karyotype

TABLE 6.9. The risk of karyotypic abnormalities related to maternal age at delivery

TABLE 6.10. Risk of a chromosomal abnormality with selected sonographic findings

A variety of genetic defects, including the common trisomies and many chromosomal translocations, can be detected by routine karyotype analysis. Recent modifications, including chromosome painting that allows a particular chromosome to be identified directly, or fluorescent in situ hybridization (FISH) that allows specific sites along the chromosome to be identified, have greatly extended the capabilities of the cytogenetics laboratory. However, there are molecular and single gene rearrangements that cannot be observed by light microscopy and require molecular genetic technology for evaluation. FlSH is a cytogenetic technique in which a specific DNA probe with a fluorescent label is bound to homologous DNA in a clinical sample. FISH can be performed either on a metaphase chromosome spread to detect microdeletions and microduplications, and during interphase to detect a larger chromosomal region in a nondividing cell. Interphase FISH can be performed on cultured cells, tissue sections, and on cytologic smears. FISH has been used to detect common aneuploidies, such as trisomy 21, trisomy 18, trisomy 13, and the sex chromosome aneuploidies, in prenatal diagnosis. Because uncultured amniocytes can be used, the FISH technique can offer more rapid detection of chromosome aneuploidies. In one of the first large studies, FISH was performed as an adjunct to conventional cytogenetics in 4,500 patients. Region-specific DNA probes to chromosomes 13, 18, 21,X, and Y were used to determine ploidy by analysis of signal number in hybridized nuclei. A sample was considered to be euploid when all autosomal probes generated two hybridization signals and when a normal sex chromosome pattern was observed in greater than or equal to 80% of hybridized nuclei. A sample was considered to be aneuploid when 70% or more of hybridized nuclei displayed the same abnormal hybridization pattern for a specific probe. The accuracy of all informative FISH results, euploid and aneuploid, was 99.8%, and the specificity was 99.9%. Current prenatal FISH protocols are not designed to detect all chromosome abnormalities and should be used only as an adjunctive test to cytogenetics. FISH can provide rapid and accurate clinical information in pregnancies when fetal abnormalities have been observed by ultrasonography. Future improvements are likely. FISH protocols, which would allow the simultaneous and unequivocal discernment of all human chromosomes, are under development. Each chromosome is labeled in a unique way using several different colors. A “spectral karyotype” can be generated, which allows visualization of a unique, defined emission spectra for each human chromosome. Computerized analysis may allow automatic and rapid analysis, with resolution approaching routine G-banding. Parent of Origin Effects Genomic imprinting refers to the differential expression of genes based on the parent of origin of the gene. Imprinting usually is mediated by differential methylation of the alleles involved. Most experimental evidence regarding imprinting comes from animal studies, but some naturally occurring human analogs exist. The paternal genetic contribution appears to be essential for the development and function of the placenta and extraembryonic tissues, but the maternal contribution is required for embryonic development. Ovarian teratoma, the most common benign pelvic tumor in women of reproductive age, is characterized by a diploid karyotype, in which both haploid sets of chromosomes are maternal in origin. Complete hydatidiform moles, which show failure of normal embryonic and fetal development, are usually diploid with two paternal haploid chromosome sets and no maternally derived chromosomes. The differential function of parental chromosomal contributions in development is also evident when studying human triploidy. In an android conception (i.e., two paternal, one maternal chromosome set), the fetus is severely growth retarded, with a disproportionately large head and syndactyly of digits of the hand. The placenta is usually very large and hydropic. Survival into the second trimester or occasionally into the third trimester is possible but usually requires the presence of mosaicism with a diploid cell line. When the chromosomal constitution is gynoid (i.e., two maternal, one paternal set), the conceptus is underdeveloped and the placenta is small and cystic; such pregnancies rarely continue beyond the first trimester. In some persons with an apparently normal karyotype, both versions of a pair of homologous chromosomes were inherited from one parent, a phenomenon called uniparental disomy. This can give rise to abnormalities if genomic imprinting causes regions of both chromosomes to be inactivated or overexpressed.

THE MOLECULAR GENETICS REVOLUTION In the mid 1950s, there were only two “facts” known about the human genome. It was thought that humans had 48 chromosomes and that X-chromosome inactivation in humans occurred by the same mechanism as had been observed in fruit flies. Both of these observations have been proven to be in error. In the past few decades,

there has been an explosion of knowledge about the human genome, largely attributed to advances in molecular biology. Deoxyribonucleic Acid Genes are the instructions required for building structural proteins and enzymes and peptide hormones, and the complete set of genetic instructions for any organism is called its genome ( Table 6.11). The human genome has 46 chromosomes, including 22 pairs of autosomes and two sex chromosomes. The genome is made up of three billion base pairs, somewhere between 40,000 and 100,000 genes. The functions of approximately 10,000 human genes have been characterized, and in 2003 the first draft of the human genome sequence will be completed.

TABLE 6.11. The human genome

In 1944, Avery and colleagues demonstrated that DNA is the chemical that carries genetic instructions. Roughly equal parts of DNA and its supporting proteins make up the 46 chromosomes. If the strands of DNA in the nucleus of a single cell could be unwound and spliced together, the resulting DNA molecule would stretch more than 1.5 meters long, but it would be only 20 trillionths of a centimeter wide. The genetic code is spelled out with the four nitrogenous bases: adenine, thiamine, cytosine, and guanine ( Fig. 6.5). The purine and pyrimidine bases are arranged in a ladderlike, double helix arrangement that is very stable (i.e., theoretic dissociation constant = 10 - 23). During cell division, DNA is duplicated with extremely high fidelity by synthesis of a new strand of one side of the molecular ladder.

FIG. 6.5. Genetic code. The DNA code consists of four characters and is read three characters at a time. It is translated into an RNA message, which instructs cells in how to assemble proteins from amino acid building blocks.

The human genome consists of at least 40,000 genes, but the genes comprise only one tenth of the encoded information. Most of the genome is of unknown function, but it probably codes for the proper spacing, alignment, and punctuation of the genetic instructions. About 99.8% of the DNA sequence is identical from one person to the next. Stated another way, there are many minor differences between any two persons; on average, there is a variation of one nucleotide for every 200 to 500 base pairs. When these sequence differences occur within genes, they can lead to genetic diseases or genetic variation. Most of the minor differences have no observable effect because they occur in the noncoding regions of the genome, regions of DNA that do not contain genes. These otherwise unimportant differences have been the basis of the current explosion of genetic knowledge, because much of our ability to study genes or diagnose genetic illness exploits differences (i.e., DNA sequence polymorphisms) in these regions to track or find neighboring genes. The DNA sequence is read by cellular enzymes three bases at a time, and each triplet directs the positioning of a particular amino acid within the structure of a protein (see Fig. 6.5). The protein coding instructions are transmitted to the cellular machinery through messenger RNA, a transient, intermediary molecule that is similar to a single strand of DNA ( Fig. 6.6). The RNA strand is transcribed from the DNA template in the nucleus and has an opposite or complementary genetic sequence. Messenger RNA moves from the nucleus into the cytoplasm, where the protein manufacturing organelles build a protein. Analysis of messenger RNA molecules is extremely useful in the laboratory for detecting genes.

FIG. 6.6. Anatomy of a gene. Regulatory regions are present in the 5' region. Introns are spliced out of the final messenger RNA.

Several advances in molecular biology have enabled the molecular genetics revolution to take place. The first was the discovery of restriction enzymes, which are bacterial proteins that can cut DNA molecules at specific sites by recognizing the DNA sequence at those sites. Over 400 restriction enzymes have been discovered, many are commercially available, and about 25 are used commonly. Restriction fragment length polymorphisms (RFLPs) occur because of minor sequence changes (usually single base substitutions) that abolish or create a recognition site, altering the length of a digestion fragment. Restriction sites occur frequently, and several restriction sites can occur in the vicinity of any given gene. When these RFLPs are polymorphic, they become useful markers for linkage studies, diagnostic testing, and paternity testing ( Fig. 6.7). RFLPs and other DNA polymorphisms provide the landmarks for genetic maps.

FIG. 6.7. Linkage study using restriction fragment length polymorphisms. Each lane represents the genotype of one family member. M, mother; F, father; D, daughter; S, son. In this example, the disease allele is associated with the upper band passed from the mother to the son.

Scientists have gained a greater understanding of how to manipulate the physical conditions, such as pH, salt concentration, and temperature, of in vitro DNA reactions. These skills—combined with the use of restrictions enzymes—allowed the development of recombinant DNA or new combinations of DNA engineered in the laboratory. Recombinant DNA technology has made possible the development of gene probes (pieces of DNA usually radioactively labeled) that recognize and bind specifically to a homologous sequence in another sample of DNA. These technologies also underlie cloning—the copying of DNA segments in lower animals and the

manufacturing of human proteins using bacteria or cell cultures. Various blotting technologies are used commonly to study DNA. With blotting, biologically relevant molecules undergo electrophoresis and are transferred to a stable membrane for repeated experiments. Blots are called Southern blots if DNA is being analyzed, Northern blots if RNA is being analyzed, and Western blots if proteins are being analyzed. DNA testing is clinically applicable to many disorders and can be performed in one of several ways ( Table 6.12). When the molecular basis of a disease is known, direct mutation testing can provide a yes or no answer on any DNA sample. For instance, in CF, hundreds of mutations have been discovered. A battery of mutations can be tested for using various methods such as dot blots, which are simple to interpret ( Fig. 6.8).

TABLE 6.12. Common conditions for which DNA testing is available

FIG. 6.8. Direct mutation diagram. Direct detection of cystic fibrosis mutations using reverse dot blots. In this example, five mutations in exon 11 of the cystic fibrosis gene (G542X, S549N, G551D, R553X, R560T), are tested for using a simple YES/NO assay. Exon 11 is amplified using the polymerase chain reaction. The product of the reaction is labeled to allow its detection and is placed on a membrane. The membrane has been prepared with oligonucleotide probes, which detect either the normal or the abnormal sequence. A: results from a known cystic fibrosis carrier. B: results from a child with cystic fibrosis.

Similarly, fragile X syndrome is usually the result of an expansion of a triplet sequence within the gene. Normal persons usually have only 5 to 50 copies of this triplet repeat, but affected patients have hundreds or thousands of copies of the triplet repeat. Similar triplet expansions cause myotonic dystrophy, Huntington disease, and Kennedy disease. The region containing the triplet can be amplified using the PCR, which produces millions of copies of the small region of DNA from the X chromosome that contains the fragile X repeat. Specificity is achieved by directing the reaction using two complementary primers on either side of the region of interest. Once amplified, the size of the product can be measured to evaluate the number of triplets, determining whether the mutation exists ( Fig. 6.9).

FIG. 6.9. Fragile X mutation detection. In patient A, a shorter polymerase chain reaction product corresponds with a smaller number of triplet repeats. Patient B exhibits an expanded number of repeats. Affected patients typically have hundreds or even thousands of copies of the triplet.

For families with unusual mutations or with diseases for which the molecular basis is unknown, linkage testing can be performed. Linkage tests compare DNA polymorphisms close to the disease-causing gene in family members known to have or carry the disease with those of unaffected and at-risk family members. Indirect assessments can be made about whether at-risk persons have the disease allele. The accuracy of these predictions depends on correct diagnosis and relationships of the family members, and the genetic distance between the polymorphism tested and the disease allele. For some families, linkage testing an be uninformative ( Fig. 6.10).

FIG. 6.10. Informativeness of linkage testing for cystic fibrosis. Marker KM-19 in kindred 18 is “not informative”-the disease alleles can not be distinguished in the parents.

The Genome Project The Human Genome Project promises to be the single most important project in biology; genetics and genomics are now the central sciences of medicine. An understanding of the relationship between genetic variation and disease risk will alter the future prevention and treatment of common illnesses. The full human sequence will be completed in 2003, but a very accurate draft is available now for over 95% of the genome. Disease gene identifications that formerly required years of chromosome walking and jumping, cloning, physical mapping, sequencing, and sequence assembly can now be completed in weeks. Industrialized sequencing technologies using capillary electrophoresis, micro arrays, and others developed for the genome project are now widely used for genotyping and sequencing. We also now have a tremendous catalog of individual sequence variation in humans. Tens of thousands of micro-satellite markers are available for linkage analysis and hundreds of thousands of SNPs (single nucleotide polymorphisms) for genetic association studies. The tools are now well developed for doing these functional studies to find which variations and mutations cause individuals to be at risk for numerous medically important, genetically complex human diseases. The genome project has delivered improved cDNA resources, better predictive software, and additional knowledge about the non-protein coding regions of the genome. Remarkable technologies are commercially available for comprehensive analysis of gene expression in single cells, tissues, or whole organisms. Advances in gene knockout technology, antisense technology, gene transfer, and gene transfection allow greater in vitro insights using appropriate model systems, including both cell culture and whole organisms. The complete sequence of the Escherichia coli, yeast, nematode, fruit fly, and mouse genomes provide important evolutionary clues to gene function and extend the range of experiments possible.

At the same time there have been parallel improvements in the technology for global protein analysis. Gene expression is played out at the protein level—elegant techniques are now available to examine spatial and temporal patterns of protein expression, protein-ligand interactions, and protein modifications. Finally, the genome project occurred at the same time as the information technology revolution. Tremendous bioinformatics and computational software is now available for gene discovery, expression profiling, understanding gene-environment interactions, and so on. Suffice it to say, better tools are now available for making advances in women's health care than ever before in human history. At least 3% of the annual budget of the project is going to the Ethical, Legal, and Social Issues section of the enterprise. This amount of early attention to societal impact is unprecedented for a science and technology project. Grant-funded programs have examined privacy issues, genetic discrimination in insurance and employment, and the role of coercion. Genetic discoveries may challenge long-held beliefs about equality, predetermination, and free will as we learn about genes that have a major role in personality, creativity, intelligence, and mental illness. The safety, efficacy, and utility of new gene tests should be evaluated, especially before treatment is available.

GENETIC EVALUATION Genetic History and Physical Examination Important details are being learned about many rare disorders which practitioners may see only once in their careers. In the past, many practitioners have had a “laissez-faire” attitude about genetic disorders, because “you can't do anything about your genes.” However, it is important to detect genetic conditions so that the patient can have adequate counseling about the condition and the risk to offspring. For severe conditions, patients often are interested in prenatal diagnosis, so they can consider pregnancy termination or prepare for the birth of an affected child. Our new predictive powers have led to an expanded medicolegal duty to warn patients of risks of which they may not have been aware. Failure to provide accurate and timely reproductive counseling has resulted in a host of lawsuits. Genetic diagnosis becomes even more critical as more treatment options become available for the child with a severe genetic disease. In light of these expanding obligations to screen, what is expected of the general obstetrician-gynecologist? As with any medical diagnosis, the history is the most important part of the genetic evaluation. Important aspects of a genetic history, such as the patient's age, menstrual history, and obstetric history, are queried routinely as part of any routine obstetric or gynecologic history. In addition, it is important to ask about the patient and her partner's ethnicity ( Table 6.13). The family history should extend to third-degree relatives (i.e., cousins). A minimal familial history can usually be elicited using the following questions:

TABLE 6.13. Single gene disorders with an ethnic predilection

Do you have a family history of diabetes, hypertension, cancer, or twins? Are there any diseases that seem to run in your family? Is there a history of genetic disease like cystic fibrosis, hemophilia, or muscular dystrophy? Is there anyone with mental retardation or any kind of birth defect? Have any of your sisters, cousins, or other relatives had problems with their pregnancies? Are your parents alive? Are they healthy? As far as you know, are you and your husband or partner related by blood? Do you know your ethnic background? (Do you know where your relatives are originally from?) Is there any reason you are especially concerned that you might have trouble with your pregnancy or that your baby may be born with a birth defect or other medical problem? It takes only a few minutes to ask these questions of a new patient. The final, open-ended question is often the most revealing. Alternatively, many practitioners find it helpful to ask similar questions using a patient-completed questionnaire. I use a form similar to the one suggested by the American College of Obstetricians and Gynecologists (ACOG) ( Fig. 6.11). It is designed so that only “yes” responses need to be dealt with further. Including a form such as this one in the patient's chart clearly documents that a genetic history was obtained for medicolegal purposes. More than one fifth of healthy obstetric patients affirmatively answer at least one important question on the form.

FIG. 6.11. Genetic screening questionnaires.

The physician should suspect genetic factors if a patient has an unusual problem and other people in her family have the same disorder. Similarly, if a patient reports a positive family history, it is important to decide whether the patient unknowingly has the same disease. When a patient reports an unfamiliar genetic condition or a rare illness, the physician should seek information about the genetics of this condition. Is she at risk of passing the condition to her offspring? Is prenatal diagnosis available? Is prenatal treatment available? For instance, consider a patient whose first child died of methylmalonic aciduria. In a subsequent pregnancy, her obstetrician noted this history on his prenatal record but did not inform the patient that there was a one-in-four risk that her current fetus could be affected. The obstetrician was unaware that prenatal diagnosis is available and, more importantly, that a simple treatment (i.e., giving the mother supplemental vitamin B 12) could prevent much of the morbidity of this particular form of methylmalonic aciduria. The child was born severely damaged and died at 2 years of age. The patient was very angry that she was not informed about prenatal therapy, and a lawsuit was initiated. Obstetricians cannot be expected to be expert in every rare enzymopathy, but this case illustrates how important it is to seek out additional information about rare conditions through a literature search or by consultation with a genetics center. Fortunately, helpful computerized databases are available on the internet such as OMIM (Online Mendelian Inheritance in Man) and Gene Tests (a directory of DNA diagnostic laboratories). The general physical examination may reveal dysmorphic features or a distinctive physical finding that frequently is associated with genetic problems. For instance, a dislocated lens found on funduscopic examination of the eye is a distinctive feature consistent with Marfan syndrome or homocystinuria. If the patient has no history of severe ocular trauma, she probably has one of these conditions. Obstetricians cannot be as good at detecting such clues as an expert medical geneticist, because “the eye cannot see what the mind does not know.”

Genetic Counseling Genetic counseling is a communication process which deals with the occurrence or risk of occurrence of a genetic disorder in a family. As our abilities to learn about the fetus have increased, more couples have an indication for prenatal diagnosis or a need to discuss reproductive options. Although every obstetrician has a role in providing genetic counseling, many practitioners find that genetic counselors—persons with advanced degrees and who are specially trained in the educational, psychological, and administrative aspects of medical genetics—are helpful consultants. Genetic counselors are experienced in obtaining and interpreting a thorough family history; often counselors are involved in the establishment or confirmation of a diagnosis. When presented with a prenatal diagnosis, they can obtain and interpret the history of a current pregnancy, explaining fetal risks and discussing the options available. Genetic counselors can provide the detailed counseling that is necessary regarding fetal chromosomal abnormalities of consanguinity, recurrence risks of multifactorial disorders, fetal abnormalities identified by ultrasonography, or infertility and habitual abortion. They are trained extensively about genetic screening for diseases that are common in various ethnic groups. Genetic counselors play a central role in the discussions regarding the option of aborting a genetically abnormal fetus. This type of counseling is traditionally informational and nondirective. Pregnancy Termination Pregnancy termination for genetic reasons can be particularly heart wrenching for a couple because the pregnancy usually is a desired pregnancy. Patients should be encouraged to involve their doctors, genetic counselors, clergy, other support persons, and family in these difficult decisions. It is the physician's responsibility to explain the fetal diagnosis and prognosis. If a woman decides to have a pregnancy termination, the physician should explain the termination procedure, options if there are any, and the relative risks of the different procedures. The cost of the procedure is discussed and whether the procedure is covered by public funding or insurance. The physician should explain the benefits of diagnostic examination of the fetus by DNA, metabolic, or chromosomal analysis or by dysmorphologic examination. The disposition of the fetal remains should be discussed. The possibility that a fetus may live for a short period after induced labor termination is discussed. With late second-trimester or third-trimester inductions, it is often appropriate to encourage patients to see or hold the baby, and to name their baby. Patients are advised that lactation may occur after the delivery, and they are told about the options available to reduce lactation. With late terminations, the option of having a memorial service or in some way commemorating the baby's existence should be discussed. Physicians and counselors help couples decide what information to tell other children and family members, friends, and acquaintances. It is important to reinforce that the genetic defect is not caused by the patient. The woman who is carrying the pregnancy and undergoes the termination may grieve in different ways than may the father of the baby. Referral to local support groups and counselors is often appreciated. Six to eight weeks after the procedure, a follow-up visit should be scheduled to summarize the diagnostic findings, review recurrence risks, and discuss prenatal diagnosis or therapy options for future pregnancies. Laboratory Screening Laboratory studies play an important role in the diagnosis of genetic disorders. A genetic illness is sometimes first discovered as an incidental finding on blood studies or an ultrasonographic examination. For instance, a low mean corpuscular volume on an automated complete blood count suggests thalassemia. In some instances, a positive family history prompts laboratory studies that clarify a patient's risk. Some programs have evolved to screen entire populations for genetic conditions using laboratory assays. Just as we currently perform a history and physical examination or a cholesterol screen to identify disease risk, soon there will be a DNA screen to detect mutations in dozens of important genes involved in cancer, cardiovascular disorders, and metabolic disease. Population screening is appropriate when a defined subset of the population is at risk, and an accurate and inexpensive heterozygote test is available ( Table 6.14). It is optimal if prenatal diagnosis is available, as well (e.g., sickle cell anemia, Tay–Sachs disease, thalassemia). The goals of screening programs are early diagnosis to allow better treatment of affected persons and identification of at-risk matings between persons who are heterozygotes or carriers of recessive disease. Neonatal screening programs for phenylketonuria, galactosemia, and hypothyroidism are carried out in most states. Successful carrier screening for Tay–Sachs disease has been achieved in several Jewish populations. The cost effectiveness of the screening program is often a primary concern in deciding whether to proceed with population screening. Equally important issues include the ability to manage minor variants which do not require action, stigmatization of carriers, and responsibility for decisions not to screen.

TABLE 6.14. Population screening

PRENATAL DIAGNOSIS Limited but important information about the fetus can be gained using the traditional diagnostic techniques of history, auscultation, and palpation. It is important to consider the onset of fetal movement and the assumption of the vertex position as developmental milestones that the fetus does or does not achieve. Experienced examiners can assess fetal size, size or dates discordance, fetal positioning, and fetal heart rate abnormalities. Maternal Serum Screening Screening for fetal genetic conditions can be achieved by testing maternal serum. The first such program involved the use of maternal serum a-fetoprotein (MSAFP) levels to test for neural tube defects, an etiologically heterogenous group of conditions characterized by failure of embryonic closure of the neural tube. A cause for a neural tube defect can be identified in only 5% to 20% of cases, and most cases are thought to be polygenic or multifactorial. Between 90% and 95% of all infants with neural tube defects are born to women with no history of a child with neural tube defect. There is substantial evidence for genetic predispositions to neural tube defects, including racial and ethnic variations in incidence, the increased incidence when a couple is consanguineous, gender bias, and increased monozygotic twin concordance. There is also strong evidence for environmental factors including maternal folate deficiency, previous spontaneous abortion or stillbirth, and the seasonal incidence. Mendelian disorders associated with neural tube defects include Meckel syndrome, in which affected individuals have a posterior encephalocele. Chromosomal syndromes, such as trisomy 18, trisomy 13, and triploidy, and sporadic syndromes such as OEIS complex (i.e., omphalocele, extrophy, imperforate anus, spinal defect) can result in neural tube defects. Other well-described environmental causes of neural tube defects include amniotic band disruption sequence, maternal diabetes, maternal use of valproic acid, and hyperthermia. Forty-five percent of fetuses with neural tube defects have anencephaly, 45% have spina bifida, 5% have an encephalocele, and the remaining 5% have iniencephaly or exencephaly. The incidence of neural tube defects is high (approximately 1%) in Ireland, Wales, Alexandria, and the Punjab. The rate is between 1 in 1,000 and 2,000 in the United States. Across the United States, the incidence is higher in the east than in the west, and highest in the Appalachian region. In the United States, if a person has previously had one child with neural tube defect, the recurrence risk is 2% to 3%. If there have been two affected children, it is 6.4%, and with three affected children, it may be as high as 25%. For patients at high risk for neural tube defect, prenatal diagnosis can be performed by targeted ultrasonography and an amniocentesis for amniotic fluid a-fetoprotein and acetylcholinesterase at approximately 16 weeks of gestation. The peak concentration of a-fetoprotein in the amniotic fluid occurs between 12 and 14 weeks, the widest margin between abnormal and normal distributions at approximately 16 to 18 weeks. A cutoff of 2.5 multiples of the median again yields a 98% detection rate, with a 0.8% false-positive rate. Acetylcholinesterase level determinations in amniotic fluid do not depend on gestational age. Amniotic fluid a-fetoprotein and acetylcholinesterase levels are normal in the 5% to 10% of cases of neural tube defects that are closed. Other open fetal defects, such as omphalocele and gastroschisis, can cause a rise in amniotic fluid a-fetoprotein. MSAFP screening was introduced for assessing fetuses of women with no known risk factors for neural tube defects in the 1980s. Like all screening tests, the predictive value of the test depends on the population prevalence and particular cutoff used for setting the limits of normal and abnormal. In the United States, a cutoff of 2.5 multiples of the median frequently is used, meaning that 5% of those tested will have positive results. With this cutoff, more than 95% of anencephalic fetuses, 80% of fetuses with open spina bifida, and approximately 5% of fetuses with closed spina bifida are detected, for an overall detection rate of approximately 64%. MSAFP screening is most accurate from week 16 to 18. MSAFP starts to increase at approximately 13 weeks and peaks at 32 weeks gestation. An inaccurate gestational age determination is the most common reason for an abnormal MSAFP result.

It is important to correct MSAFP values for maternal weight, race, diabetes, and multiple gestation. There is a negative correlation between maternal weight and MSAFP. Blacks have approximately 1.1 times the MSAFP level of Caucasians, and Asians have an intermediate level between blacks and Caucasians. In insulin-dependent diabetics, the MSAFP level is approximately 60% of nondiabetic controls, and it is inversely correlated with the hemoglobin A 1C levels. Between 1% and 2% of infants of diabetic mothers have babies with neural tube defects. In multiple gestation, the median twin MSAFP level from 16 to 20 weeks is about 2.5 multiples of the median for a singleton pregnancy. Low MSAFP levels have been associated with Down syndrome. One fifth to one third of these fetuses' mothers exhibit low MSAFP levels, with a median MSAFP of 0.7. Additional assays, such as unconjugated estriol, can provide more information about risk. Estradiol levels are low in cases of trisomy 21, very low in trisomy 18, and normal with spina bifida. Human chorionic gonadotropin levels are high in trisomy 21, very low in trisomy 18, and low in anencephaly. For some couples, MSAFP screening raises anxiety, because the results are available around the time they feel that miscarriage is not going to occur and after the pregnant woman already feels fetal movement. When counseling patients about MSAFP screening, it is important to stress that it is a screening rather than a diagnostic test. The physician should explain the possible reasons for a high or low result, discuss the evaluation that would be recommended in that case, and stress that most babies of mothers with an abnormal screening result are normal. Frequently, discussions about a-fetoprotein screening bring out other issues the couple are worried about with respect to birth defects. This discussion is also an opportunity to educate the couple about the background incidence of birth defects. The search is on for new biochemical markers which would improve the sensitivity and specificity of maternal serum screening. Retrospective studies suggest that maternal serum levels of dimeric inhibin A may be highly predictive. When the fetus is affected by Down syndrome, the maternal serum inhibin A concentrations are 2.1 times the median value in controls. Serum concentrations of inhibin A in Down syndrome pregnancies do not rise above normal until the end of the first trimester. The levels were not significantly different in the women with fetuses affected by trisomy 18. Small retrospective studies also have examined the feasibility of first-trimester screening for Down syndrome. Earlier screening for Down syndrome would allow more time for intervention in the event of a positive test result. For women who choose pregnancy termination, the procedure can be carried out at a time when it is medically, psychologically, and perhaps morally less problematic. On the other hand, earlier tests will find many Down syndrome fetuses that would have aborted spontaneously. The follow-up diagnostic tests carry a greater risk of miscarriage (of normal pregnancies) at these earlier gestational ages. Serum a-fetoprotein screening for neural tube defects (NTDs) is impossible during the first trimester. Initial studies show that maternal serum free ß-human chorionic gonadotropin and pregnancy-associated plasma protein A are useful markers. The “free-ß” chorionic gonadotropin levels are approximately one half the median, and pregnancy-associated plasma protein A levels are twice the median control levels in Down syndrome pregnancies. Urinary markers are being evaluated also. Large, prospective studies are needed to further assess first-trimester screening. Eventually, sorting fetal cells from maternal blood may prove to be the most sensitive and specific screening test. Fetal Imaging For many years, the fetus could be seen before birth only by using x-ray films. Radiographic examinations have limited prenatal indications because of concerns about fetal radiation exposure and because the information obtained by radiographs is limited to inspection of the calcified structures. Today, fetal radiography is used mostly for the differential diagnosis of skeletal dysplasias in the third trimester. In the past, attempts to gain information about the fetal soft tissues involved injecting into the amniotic fluid a water-soluble dye to outline the fetal gastrointestinal tract or a fat-soluble dye to outline the fetal skin. Modern high-resolution ultrasonography has revolutionized fetal imaging, giving clinicians a noninvasive way to get information about the internal and external features of the fetus. Newer imaging methods serve as a useful adjunct to ultrasonographic examination for the prenatal diagnosis of certain conditions. Computed tomography (CT) uses low doses of radiation and computerized processing to obtain cross-sectional images. Magnetic resonance imaging (MRI) is based on detection of moving hydrogen atoms when tissues are subjected to a strong magnetic field. Both methods are expensive, but they are noninvasive and do not exhibit the shadowing phenomena seen with ultrasonography. CT and MRI are most useful for suspected central nervous system anomalies, particularly if ultrasonographic imaging is limited by reverberation artifacts caused by the fetal skull. They also are useful for cases of oligohydramnios; decreased amniotic fluid makes ultrasonographic imaging difficult, but the condition holds the fetus still for CT or MRI. Fetal movement is a major limiting factor with CT and MRI, but the newer ultrafast scanners can produce an image within fractions of a second. Prenatal studies do not necessarily require maternal sedation or fetal paralysis with an intrauterine, intramuscular injection of a muscle relaxant such as curare. Fast CT scans have slightly lower resolution but an even lower dose of radiation compared with conventional CT. Another advantage of CT is the ability to use contrast agents. For instance, CT amniography can differentiate cyst adenomatoid malformation of the lung and diaphragmatic hernia by demonstrating the location of fetal stomach and small bowel. MRI allows differentiation in tissue densities and is exceptionally useful for differentiating white and gray matter in the central nervous system, fat, and flowing blood. MRI computers can construct images in any plane desired. Although there are no known biologic hazards with MRI, there are also no clear indications for use of MRI in the first trimester. Because the teratogenic risk is unknown, MRI use should be limited to the second and third trimesters. Direct visualization of the fetus is indicated only in certain clinical situations and can be performed using a small-bore, fiberoptic endoscope. The trocar for the most commonly used fetoscope is 2.2 mm in diameter; the scope, itself, is 1.7 mm in diameter. The narrow field of view and the short focal length give a limited view of a small portion of the fetus. Fetoscopy may reemerge as an important adjunct to amniocentesis and fetal blood sampling as narrower scopes are developed. Scopes small enough to fit through the shaft of a 20-gauge needle have been developed, but it is not yet possible to get enough light inside the uterus to allow visualization with such a narrow scope. Embryoscopy has been used during the first trimester to visualize the embryo or early fetus through the membranes. This is accomplished by passing the endoscope through the cervix and up against the membranes. Neither fetoscopy nor embryoscopy may be possible in many cases because of the placental position or cloudy amniotic fluid. Fetal Sampling for Prenatal Diagnosis Amniocentesis Amniocentesis was introduced to the United States in the 1960s, and it is the most extensively used fetal sampling technique. Genetic amniocentesis is performed routinely at approximately 15 weeks of gestation when the amniotic fluid volume is approximately 200 mL. At this gestational age, ultrasonographic examination cannot detail all of the fetal anatomy, but it can reliably ascertain dates or rule out multiple gestations. Typically, 20 mL of fluid is removed with a 20- to 22-gauge needle using a transabdominal approach with ultrasound guidance ( Fig. 6.12). Biochemical testing can be performed on the fluid as indicated. Amniotic fluid a-fetoprotein levels are obtained routinely to screen for open fetal defects, and fetal cells can be grown for karyotype determination or for DNA assays. DNA assays that use the PCR to amplify small amounts of DNA allow direct analysis of amniotic fluid. Roughly one third of the amniocenteses must be performed transplacentally. In most operators' experience, this has not been associated with substantially increased risk if care is taken to avoid major fetal vessels. The transplacental approach is associated with a slightly higher incidence of Rh sensitization. With either approach, it is imperative that Rh-negative women who may be carrying an Rh-positive fetus receive RhoGAM.

FIG. 6.12. Amniocentesis. With multiple gestations, it is usually possible to sample each of the gestations. Indigo carmine dye can be placed in the sac after the amniocentesis is completed to prevent tapping the same sac twice. Biochemical assays of amniotic fluid are somewhat harder to interpret in multiple gestations, because many biologic molecules can diffuse from one sac into the other. Few women describe amniocentesis as terribly painful. Those who do frequently experience a uterine contraction at the time the needle is inserted. It is not unusual to have some cramping or a bruised feeling at the site after the procedure. Vaginal spotting or amniotic fluid leakage occurs in 1% to 2% of cases. After a routine amniocentesis, the fluid usually stops leaking within 2 to 3 days. Even when the amniotic fluid volume becomes markedly decreased, miscarriage is not inevitable, because the membranes usually seal and the amniotic fluid can reaccumulate within a week, allowing the pregnancy to progress normally. As ultrasonographic equipment has improved, the maternal risk incurred with amniocentesis has decreased. Symptomatic amnionitis occurs in fewer than 1 of 1,000

patients. Serious maternal bowel or vascular injuries are extremely rare. The procedure-related rate of fetal loss after amniocentesis generally is quoted as 0.5% (1 in 200), but many centers are reporting lower rates. Chorionic Villus Sampling Chorionic villus sampling (CVS) is a diagnostic technique that was introduced to the United States in the mid-1980s. With this technique, a small sample of the chorionic villi is taken for examination of chromosomal status, biochemical assays, or DNA tests. Assays depending on analysis of amniotic fluid such as a-fetoprotein cannot be performed on a chorionic villus sample. CVS usually is accomplished by the transcervical or transabdominal route. Occasionally a transvaginal CVS is performed with the uterus extremely retroflexed. A transcervical CVS usually is performed between 9 and 12 weeks of gestation, at which time a plastic catheter, approximately 1.5 mm in diameter, is passed through the cervix and then directed toward the placental mass under continuous ultrasound guidance ( Fig. 6.13). Between 10 and 20 mg of villi are aspirated through this catheter by negative pressure using a syringe. Transabdominal CVS is performed using an 18- to 20-gauge spinal needle passed into the thickest portion of the placenta that is readily assessable ( Fig. 6.14). Villi are aspirated into a syringe. This procedure can be performed throughout gestation.

FIG. 6.13. Transvaginal chorionic villus sampling.

FIG. 6.14. Transabdominal chorionic villus sampling. Transabdominal CVS is considered easier to learn and safer, but patient acceptance appears to be lower. Theoretically, transcervical CVS would have a greater risk of infection, although this has not been borne out by large surveys. In most series, a larger sample is obtained with transcervical CVS, but more passes are required to obtain this sample. Transcervical procedures require more uterine manipulation, and bleeding or leakage of fluid is more common during and after the procedure. Most laboratories report a greater level of maternal cell contamination with transcervical CVS, although this is rarely a clinically important issue. Those performing CVS should be facile with both techniques, because patient anatomy frequently dictates which is the optimal technique. CVS compares favorably with amniocentesis with regard to safety. Two large National Institutes of Health cooperative trials found a procedure-related loss rate of approximately 0.8%. In approximately 2% of first-trimester chorionic villus samples, a discrepancy is found between the cytogenetic analysis of the placenta and that of the fetus. Frequently, a second invasive procedure, usually amniocentesis or fetal blood sampling, is required to determine whether the fetus is affected. This phenomenon is called confined placental mosaicism. Pregnancies in which confined placental mosaicism is found by CVS may be at risk for spontaneous abortion, perinatal loss, or intrauterine growth retardation. The reported rates of loss have ranged from 3.6% to 16.7%. In chromosomally abnormal conceptuses, a mosaic normal cell line in the placenta may be the factor that allows prolonged survival of aneuploid fetuses. Kalousek studied 14 placentas from live-born or from terminated pregnancies with trisomy 13 or 18, and found the placentas were all mosaic for, or contained only, diploid cells. There have been several reports of increased incidents of limb anomalies when fetuses have undergone very early CVS using the transabdominal approach. These reports are of concern, because the affected children have a relatively distinctive pattern of malformation, and it is biologically plausible that their anomalies may be related to CVS. The absolute number of fetuses with this problem is small, but the publicity regarding these findings has caused many women to avoid CVS. The Centers for Disease Control performed a multistate case-control study to assess and quantify the risk for specific limb deficiencies associated with CVS. Between 1988 and 1992, 131 infants with nonsyndromic limb deficiency, born to mothers 34 years of age or older, were reported in seven population-based birth defect surveillance programs. Control subjects were 131 infants with other birth defects. They found that exposure to CVS was associated with a six-fold increase in risk for transverse digital deficiency (odds ratio = 6.4; 95% confidence interval, 1.1 to 38.6). The data showed a significant trend toward increased risk with earlier gestational exposure. The CDC estimates that the absolute risk for transverse digital deficiency in infants after CVS is approximately 1 per 3,000. Most feel that the actual risk is significantly lower and that there is minimal to no risk when CVS is performed after 70 days of gestation. Further studies are needed to determine whether the problem is specific to CVS or whether the same risk affects other invasive first-trimester diagnostic manipulations. Fetal Blood Sampling and Fetal Biopsy Originally, fetal blood was sampled by inserting a needle into the placenta; the blood obtained was usually a mixture of fetal and maternal blood, which limited its usefulness. In 1977, fetoscopy came into use, allowing a needle to be placed in a cord vessel under direct sonographic visualization. Pure fetal samples could be obtained, but fetoscopy requires special equipment and expertise, and the procedure-related loss rate was between 3% and 7%. With improvement in ultrasonographic imaging during the mid-1980s, cordocentesis can be performed in a manner similar to amniocentesis ( Fig. 6.15).

FIG. 6.15. Cordocentesis (fetal blood sampling). A transplacental route usually is preferred, and a spinal needle is advanced under ultrasound guidance into a vessel with cord insertion into the placenta. Cordocentesis usually is not performed until after 17 weeks gestation. Depending on the indication, procedure-related loss rates of as low as 1% have been reported with no observed increase in the rate of preterm delivery. Typically, no maternal sedation or antibiotics are required. In most cases, a fetal sample can be obtained on the first attempt, usually in less than 10 minutes, and any substance measurable in adult blood can be assayed in fetal blood. Most cordocenteses are performed to obtain a fetal karyotype because of fetal anomalies or to determine the fetal hematocrit to assess isoimmunization or severe fetal anemia. Fetal platelet counts, acid-base status, antibody levels, and blood chemistries can be assayed as indicated. Hematologic values are checked routinely (particularly the mean corpuscular volume which is higher in fetus than mother) to be certain that the blood obtained is fetal. A Kleihauer test can be performed to check for maternal blood contamination. A variety of other tissues, particularly fetal skin, liver, and muscle, have been sampled prenatally to diagnose a genetic disorder using either electron microscopy or biochemical analysis. Such biopsies are necessary if a genetic abnormality is expressed only in certain tissues and the causative gene is unknown. Now that the molecular basis of many of these disorders is known, simpler DNA assays using villi, amniocytes, or blood usually obviate the need for tissue biopsy. Initially, fetal

biopsies were performed using fetoscopy, but they now are performed under ultrasound guidance. Typically, local anesthesia, maternal sedation (with or without fetal paralysis), and prophylactic antibiotics are used for tissue biopsy procedures. As with any invasive procedure, RhoGAM is necessary for the nonsensitized Rh-negative mother who may be carrying an Rh-positive fetus. Early Amniocentesis Early amniocentesis is similar to amniocentesis at the “traditional” gestational age, except that the procedure is performed at 10 to 12 weeks of gestation. Early amniocentesis was proposed as an alternative first-trimester prenatal diagnostic technique, but a large Canadian prospective study demonstrated an unacceptable rate of procedure-related malformations in the exposed pregnancies. Preimplantation Diagnosis Various methods of diagnosis before a pregnancy is formally established are becoming available in the 1990s. Someday these extremely early diagnoses may be necessary to allow initiation of genetic therapy treatments. For many couples, preimplantation diagnosis provides an alternative to selective pregnancy termination, allowing them to avoid the moral issues and the psychological trauma that accompanies termination of a wanted pregnancy. For most genetic conditions, carriers have both normal and abnormal gametes, and chance determines whether an abnormal gamete is incorporated into the conceptus. Aided by the rapid progress in assisted reproductive technologies and molecular diagnostic techniques, it has become possible to test gametes in vitro and select healthy gametes for fertilization. Sperm sorting has been accomplished using molecular probes tagged with laser-activated dyes and using separation techniques such as flow cytometry, but current approaches usually cause unacceptable damage to the sperm. Greater success has been possible in genotyping oocytes, which are larger and more resistant to damage. Oocyte diagnosis takes advantage of the unique properties of female meiosis; unlike sperm, oocytes conveniently discard their unused genetic material in the form of polar bodies. For a heterozygous woman, the discarded genetic material can be tested to see whether it contains the abnormal allele. If it does, then the oocyte must contain the normal allele. Conversely, if the polar body tests positive for the normal allele, the oocyte must contain the abnormal allele. Only the normal oocytes are then fertilized. Contamination with cumulus cells adherent to the exterior surface of the zona pellucida can lead to errors. Polar body biopsy can be performed only in conjunction with in vitro fertilization. The first polar body has no essential function and contains no embryonic material; it is small and relatively easy to remove using micromanipulation techniques ( Fig. 6.16). Theoretically, DNA testing results can be obtained before fertilization, leaving a long period for confirmatory studies before implantation. Aneuploidy screening can be incorporated as well, at least for maternal meiosis I nondisjunction. The main disadvantages are that polar body biopsy is an indirect assay of the oocyte and that it can be used only for maternal carriers. In addition, the further away from the centromere the gene is located, the more likely recombination is to occur, making polar body biopsy results indeterminate. Polar bodies that are heterozygous need to be discarded or the oocytes biopsied again to remove the second polar body.

FIG. 6.16. Polar body biopsy. Initial success has been greater with “selective implantation” protocols in which the diagnosis is made during the first week after fertilization but before implantation. Preimplantation embryos can be grown in vitro and biopsies obtained after the first few cleavage divisions. There is much experience with this technique in animal research, and embryo splitting at this stage has been used extensively for diagnosis in the cattle and sheep industry. The eight-cell preembryo is probably at the ideal stage for biopsy. Cells are still independent and totipotent, they have not developed gap junctions which will make them adherent, and damage is tolerated relatively well. By this stage, some of the embryonic genes have begun to function, and it may become possible to perform some biochemical microassays. This technique is limited by the difficulty in removing blastomeres, which are larger than polar bodies, and the short time to work with the sample before the chance of successful implantation begins to lessen. It has proved difficult to freeze spare embryos after biopsies and, as with other preimplantation techniques, there is the possibility of sperm or cumulus cell contamination. Attempts have also been made to perform testing 5 days after fertilization, when the preembryo has reached the blastocyst stage. The blastocyst consists of roughly 120 cells that are mostly trophoblastic tissue, but the inner cell mass that eventually becomes the embryo is clearly visible. Only the best laboratories have had any success culturing human preembryos to this developmental stage in vitro. As an alternative, investigators have tried to lavage naturally conceived blastocysts from the uterus during the 2 to 3 days when the conceptus is normally free-floating in the endometrial cavity before implantation. Unfortunately, it has proved exceedingly difficult to obtain multiple blastocysts by lavage after superovulation. Various methods of sampling blastocyst cells have shown success in animal models including bisection, aspiration of the cavity, and excision of cells herniating through the zona pellucida. The latter is achieved after mechanical disruption of the zona to cause premature herniation or after spontaneous hatching. The advantages of blastocyst biopsy include the relative differentiation of the cells, greater cell number, better transfer efficiency, and self-selection of the healthiest embryos. Lavage of naturally conceived blastocysts offers a potentially “low-tech” approach to preimplantation diagnosis. Perhaps contrary to expectations, disruption of zona pellucida may enhance hatching. Technical limitations include the small number of blastocysts available and their limited incubation time. The cells are very adherent to each other, increasing the risk of damage to the inner cell mass. As with CVS mosaicism, there is the possibility that the trophectoderm does not reflect the fetal karyotype or biochemical status. Fetal Cells in Maternal Circulation The newest fetal sampling technique that shows promise involves separating fetal cells that occur naturally in the maternal circulation. Nucleated erythrocytes, fetal leukocytes, and syncytiotrophoblast cells are found in the maternal circulation during most pregnancies, from as early as 6 weeks gestation. It is possible to use a separation technique such as flow cytometry to establish an enriched population of fetal cells and then assay these cells for fetal mutations using a technique such as DNA amplification. Certain cell types have a long life span in the maternal circulation, allowing persistence of cells from prior pregnancies, and confusing results are possible. “Vanished” abnormal co-twins can lead to diagnostic errors. Ultimately, this method may provide a noninvasive, reliable screen for aneuploidy that is inexpensive enough to use in low-risk populations. Cystic Fibrosis Screening The CFTR gene was discovered in 1989, and more than 1,000 mutations which can lead to CF have been identified. Couples who both carry the CF gene would have a 1 in 4 chance of delivering a child with CF. Cystic fibrosis causes pulmonary and gastrointestinal disease of varying severity. Most patients with CF have substantial illness and shortened life span and require lifelong medical care. Screening is now widely available for the most frequent CF mutations. Recently ACOG advised that DNA screening for CF should be made available to all couples seeking preconception or prenatal care. The ACOG/American College of Medical Genetics publication entitled Cystic Fibrosis Carrier Testing: the Decision is Yours can be helpful for this purpose. Because Caucasians have a higher rate of CF (particularly European or Ashkenazi Jewish), obstetric care providers are advised to offer screening specifically to these couples and to record in the medical record the couple's decision on whether to be screened ( Table 6.15.).

TABLE 6.15. Cystic Fibrosis: carrier rates in the United States

Prenatal Treatment and Gene Therapy As we begin to understand the molecular mechanisms by which genes cause disease, we will have the opportunity to design and apply preventive, health maintaining measures. Some genetic conditions can be treated by giving patients the protein they are missing or by stimulating a function that is not performing properly (e.g., growth hormone deficiency, diabetes). More sophisticated gene therapies are being tested. Finding the CF gene enabled the disease to be “cured” in the test tube and improved in animal models. Scientists may be able to design proteins or antisense RNAs that can be used as a drug to block the effect of abnormal genes or kill cancer cells. Modified viruses may be used to insert corrected genetic instructions. Prenatal gene therapy with stem cell transplantation has been attempted for severe combined immunodeficiency with some preliminary success. Many genetic conditions need to be treated prenatally. Although in utero treatment is experimental, prenatal treatment of disorders with vitamin-dependent or responsive cofactors has been successful. For instance, prenatal treatment of the vitamin B 12-responsive form of methylmalonic acidemia by administration of 10 mg per day of vitamin B 12 has improved the biochemical defect. Similarly, infants with biotin-responsive multiple carboxylase deficiency have been aided by maternal biotin supplementation. In these rare disorders, prenatal treatment can mean the difference between life and death for affected infants. Dietary restriction of galactose in mothers who are at risk of delivering a galactosemic infant, and dietary restriction of phenylalanine in mothers who themselves have phenylketonuria, are helpful means of preventing the devastating effects of these metabolic conditions. Gene therapy and prenatal tissue transplantation are likely to rapidly expand the therapeutic

options for treating and preventing metabolic disease.

GENETICS IN GYNECOLOGIC DISORDERS Genes play an important role in the pathogenesis of many common gynecologic disorders. Molecular genetic investigations of persons with gonadal dysgenesis and pseudohermaphroditism have defined many aspects of human sexual differentiation. Because it is the smallest chromosome, the Y chromosome became the first human chromosome to be mapped completely. Genetic testing for susceptibility to ovarian cancer is rapidly becoming part of routine practice. Most cancer is clonal in origin, meaning it arises from a single aberrant cell. Cytogenetic or molecular alterations are observed uniformly in malignant cells. Although some of these changes appear to be random events occurring in rapidly dividing cells, other specific genetic changes play an etiologic role in development of certain cancers. Particular mutations may be either germinal (i.e., inherited) or somatic (i.e., acquired). Either can be seen in familial cancer clusters: germinal because of segregation within the family of a cancer-causing mutation and somatic because of shared environmental exposures to carcinogens. Mendelian transmission of cancer predisposition usually is observed as multifocal and early-onset disease. Typically, cancer-predisposing mutations are found to overexpress protooncogenes that normally drive important cell functions or to inactivate tumor suppressor genes that normally exert a protective effect. Genetic testing for BRCA1 and BRCA2 mutations is now recommended to most women with invasive ovarian cancer. Approximately 10% of patients with ovarian cancer will have a positive test result, including 4% of women without a family history of ovarian cancer. Women with the BRCA mutation have better ovarian cancer survival rates than women without the mutation, possibly due to enhanced susceptibility to chemotherapy. A variety of strategies for prevention of ovarian cancer in relatives at risk, including chemoprevention and prophylactic oophorectomy, have shown some efficacy. Common gynecologic diseases such as endometriosis and polycystic ovary syndrome are familial, and genes involved in these conditions are likely to be discovered over the next few years. Age at menopause, susceptibility to hot flushes and osteoporosis, susceptibility to pelvic relaxation, and susceptibility to chronic vaginitis are likely to have genetic components, as well. Disease gene discoveries related to these conditions may suggest novel diagnostic and therapeutic approaches.

TRENDS The genome project promises to provide us with the most important information in human biology. Technologies developed for the genome project and the genomic sequence, itself, will provide the basis for much of biomedical research in the next century. The possibilities for understanding normal development, disease predisposition, and cancer are staggering. The ability to obtain an accurate prenatal diagnosis will expand exponentially over the next few decades. Gene therapy is becoming a reality faster than anyone thought possible. The challenge is for the obstetrician-gynecologist to stay abreast of all these developments and to educate patients about developments that can influence their care.

SUMMARY POINTS Most obstetric and gynecologic diseases show a polygenic, multifactorial pattern of inheritance. One in twenty newborns has a diagnosable genetic disorder. A thorough family history is currently the most important part of a genetic evaluation, but laboratory screening of the general population is becoming available for an increasing number of genetic conditions. Accurate prenatal diagnosis is now possible for hundreds of genetic conditions through ultrasonographic and genetic testing. The Human Genome Project is providing new information each month, making it difficult for care providers to keep abreast of all the new developments, especially for rarer diseases. Reference to current on-line data and liberal referral to genetic counselors and geneticists is necessary when encountering rare conditions. Over the next decade, genetic testing will continue to become less invasive, and there will be greater opportunities to prevent the morbidity of genetic disease through prenatal and presymptomatic treatments. REFERENCES ACOG Practice Bulletin. Clinical Management Guidelines for Obstetrician-Gynecologists. Prenatal diagnosis of fetal chromosomal abnormalities. Obstet Gynecol 2001;97(5 Pt 1)[Suppl]:1–12. Aitken DA, Wallace EM, Crossley JA, et al. Dimeric inhibin A as a marker for Down's syndrome in early pregnancy. N Engl J Med 1996;334:1231–1236. Alfirevic Z. Early amniocentesis versus transabdominal chorion villus sampling for prenatal diagnosis. Cochrane Database Syst Rev 2000;(2):CD000077. Cohen DP, Layman LC. Relevance of molecular medicine to clinical obstetrics and gynecology. Obstet Gynecol Surv 1997;52:73–80. Evans MI, Levy HL. The future of newborn screening belongs to obstetricians. Acta Paediatr Suppl 1999;88:55–57. Forest JC, Masse J, Moutquin JM. Screening for Down syndrome during first trimester: a prospective study using free ß-human chorionic gonadotropin and pregnancy-associated plasma protein A. Clin Biochem 1997;30:333–338. Genetic testing for cystic fibrosis. National Institutes of Health Consensus Development Conference Statement on genetic testing for cystic fibrosis. Arch Intern Med 1999;159:1529–1539. Gupta GK, Bianchi DW. DNA diagnosis for the practicing obstetrician. Obstet Gynecol Clin North Am 1997;24:123–142. Hall JG. Genomic imprinting and its clinical implications. New Engl J Med 1992;326:827–829. Jenkins TM, Wapner RJ. First trimester prenatal diagnosis: chorionic villus sampling. Semin Perinatol 1999;23:403–413. Krantz DA, Larsen JW, Buchanan PD, et al. First-trimester Down syndrome screening: free ß-human chorionic gonadotropin and pregnancy-associated plasma protein A. Am J Obstet Gynecol 1996;174:612–616. Milunsky JM, Milunsky A. Genetic counseling in perinatal medicine. Obstet Gynecol Clin North Am 1997;24:1–17. Miny P, Tercanli S, Holzgreve W. Developments in laboratory techniques for prenatal diagnosis. Curr Opin Obstet Gynecol 2002;14:161–168. Narod SA, Boyd J. Current understanding of the epidemiology and clinical implications of BRCA1 and BRCA2 mutations for ovarian cancer. Curr Opin Obstet Gynecol 2002;14:19–26. Nyberg DA, Souter VL. Sonographic markers of fetal aneuploidy. Clin Perinatol 2000;27:761–789. Pergament E. The application of fluorescence in-situ hybridization to prenatal diagnosis. Curr Opin Obstet Gynecol 2000;12:73–76. Pertl B, Bianchi DW. Fetal DNA in maternal plasma: emerging clinical applications. Obstet Gynecol 2001;98:483–490. Sanders RC. Prenatal diagnosis of structural abnormalities. Curr Opin Obstet Gynecol 1991;3:259–265. Schimpf MO, Domino SE. Implications of the Human Genome Project for obstetrics and gynecology. Obstet Gynecol Surv 2001;56:437–443. Schrock E, du Manoir S, Veldman T, et al. Multicolor spectral karyotyping of human chromosomes. Science 1996;273:494–497. Simpson JL, Golbus MS. Genetics in obstetrics & gynecology, second ed. Philadelphia: WB Saunders Company, 1992. Souter VL, Nyberg DA. Sonographic screening for fetal aneuploidy: first trimester. J Ultrasound Med 2001;20:775–790. Swisher E. Hereditary cancers in obstetrics and gynecology. Clin Obstet Gynecol 2001;44:450–463. Verlinsky Y, Handyside A, Grifo J, et al. Preimplantation diagnosis of genetic and chromosomal disorders. J Assist Reprod Genet 1994;11:236–243. Wilson RD. Prenatal evaluation for fetal surgery. Curr Opin Obstet Gynecol 2002;14:187–193. Yang EY, Flake AW, Adzick NS. Prospects for fetal gene therapy. Semin Perinatol 1999;23:524–534.

Chapter 7 Drugs in Pregnancy Danforth’s Obstetrics and Gynecology

Chapter 7 Jerome Yankowitz

Drugs in Pregnancy

PRINCIPLES OF TERATOLOGY ANTIBIOTICS AND OTHER ANTIINFECTIVE AGENTS Antibiotics UPPER RESPIRATORY TRACT COMPLAINTS Antihistamines Decongestants Cough Suppressants ASTHMA TREATMENT ß-Sympathomimetic Agents Glucocorticoids Theophylline Cromolyn Sodium Leukotriene Receptor Antagonists and Lipoxygenase Inhibitors GASTROINTESTINAL DISORDERS Antihistamines Antidopaminergic Agents Other Agents ANALGESIC USE Nonsteroidal Antiinflammatory Drugs Opioid Analgesics PSYCHIATRIC DISORDERS VITAMIN AND MINERAL USE RECREATIONAL DRUG USE ANTICONVULSANT AGENTS HEADACHES ANTINEOPLASTIC AGENTS ANTICOAGULATION RADIOLOGIC EXAMINATIONS COMPLEMENTARY AND ALTERNATIVE THERAPY SUMMARY POINTS USEFUL DATABASES AND WEB RESOURCES RECOMMENDED READINGS

PRINCIPLES OF TERATOLOGY Anything a pregnant woman ingests or is exposed to could affect her fetus. This is problematic for the health care provider who must treat a wide variety of illnesses during pregnancy. In fact, over 60% of American women receive a prescription for at least one medication during pregnancy. This figure is about 99% for women in France and over 70% for Hungarian women. The most common conditions for which medications are prescribed include gastrointestinal, dermatologic, psychological, and psychiatric disorders and pain. Generally, whatever medication would be given to a nonpregnant women is the appropriate choice in pregnancy. Awareness of the few exceptions and how to choose from available options requires a knowledge of teratology and alterations in drug metabolism related to pregnancy. Teratology is the study of abnormal development or the production of defects in the fetus. Birth defects affect 2% to 3% of all neonates. With longer follow-up, at least 5% of individuals are found to be affected by a birth defect. Exogenous causes of birth defects, including drugs or chemical exposures, account for almost 10% of birth defects. Thus, at least 0.2% to 0.3% of pregnancies are affected by teratogenesis. The Food and Drug Administration (FDA) introduced a drug classification system in 1979 to discourage nonessential use of medication during pregnancy. Drugs are classified as either A, B, C, D, or X, with the latter being the most teratogenic. There has been growing perception that the FDA classification has led to excessive maternal anxiety and unnecessary pregnancy termination. The FDA is evaluating a revised labeling system for drugs and biologic agents that will include a description of the drugs based on clinical management, summary of risk assessment, and discussion of data. How much drug the fetus will be exposed to is determined by a complex interaction of many factors, including how the agent is absorbed, the volume of distribution, metabolism, and excretion. Absorption is via the gastrointestinal tract, skin, lungs, or after parenteral administration. Pregnancy alters absorption in a variety of ways, including prolongation of gastric emptying time by the increased progesterone. The volume of distribution is generally increased during pregnancy. Estrogen and progesterone alter hepatic enzyme activity with varying effects on drug metabolism and clearance, depending on the precise pathway. Renal excretion is generally increased during pregnancy. Other factors affect precisely how much drug crosses the placenta. Passage of drugs across the placenta is influenced by several factors. Lipid-soluble substances readily cross the placenta, and water-soluble substances pass less well. Those with greater molecular weight also cross the placenta less easily. The degree to which a drug is bound to plasma protein influences the amount of drug that is free to cross. Virtually all drugs cross the placenta to some degree, with the exception of large organic ions such as heparin and insulin. Active placental transfer must be considered, also. Other concepts related to teratology include specificity, timing, dosage, maternal physiology, embryology, and genetics. Specificity indicates that a substance may be teratogenic in some species but not others. For example, thalidomide produces phocomelia in primates but not rodents. Often, animal data of safety or teratogenic effect is not necessarily applicable to humans. Timing is also critical. When administered between 35 and 37 days gestation, thalidomide produces ear malformations, but between 41 and 44 days, amelia or phocomelia. Dosage is also important. In most cases, administration of a low dose will result in no effect while malformations occur at intermediate doses and death at higher doses. Death may cause organ-specific teratogenic action to go unnoticed. The route of administration, possibly secondary to absorption, is also important. Small doses over several days may have an effect different than the same total dose given at once. Sequential dosing, as opposed to a bolus, may induce an enzyme to metabolize the substance that potentially causes less damage. Constant exposure may destroy cells that would have catabolized the drug if administered in periodic doses. As noted above for thalidomide, timing of exposure relative to embryologic events is important. Teratogen exposure in the first 2 to 3 weeks after conception is generally thought to have no effect or to result in spontaneous loss (all-or-nothing phenomenon). The period of susceptibility to teratogenic agents is during the period of organogenesis, which occurs primarily at 3 to 8 weeks postconception (35-70 days after the last menstrual period or LMP) or to 10 weeks from the LMP. After this period, embryonic development is characterized primarily by increasing organ size (10-12 weeks); thus, the principal effect of exposure will be growth restriction or effects on the nervous system and gonadal tissue. These systems continue to develop throughout pregnancy. During organogenesis each organ system will have different critical periods of sensitivity. A teratogen can act by causing cell death, altering tissue growth (hyperplasia, hypoplasia, or asynchronous growth), or interfering with cellular differentiation or other basic morphogenic processes. The genetic make-up of the mother and fetus can affect individual susceptibility to a drug. Fetuses with low levels of the enzyme epoxide hydrolase may be more likely to manifest the fetal hydantoin syndrome than those with normal levels of epoxide hydrolase. Combinations of agents may produce different degrees of malformation or growth restriction than if given individually. Fetuses whose mothers are on combination antiepileptic agents are at the highest risk for malformations, including neural tube defects and facial dysmorphic features. Most drug therapy does not require cessation of nursing because the amount excreted into breast milk is small enough to be pharmacologically insignificant.

ANTIBIOTICS AND OTHER ANTIINFECTIVE AGENTS Antibiotics are used widely during pregnancy to treat a variety of disorders including upper respiratory tract infections, urinary tract infections, and others ( Table 7.1). Pregnant patients are particularly susceptible to vaginal yeast infections. This is one reason that antibiotics should be used only when clearly indicated. Therapy with antifungal agents may be necessary after the course of antibiotic therapy.

TABLE 7.1. Antibiotics used for common infections in pregnancy

Antibiotics Penicillins Penicillin derivatives, including amoxicillin and ampicillin, have a wide margin of safety and lack toxicity for both the woman and her fetus. Penicillin is a ß-lactam that inhibits bacterial cell wall synthesis and can be administered orally, intramuscularly, and intravenously. It is the drug of choice for the treatment of a wide variety of bacterial infections including group A streptococcal pharyngitis, otitis media, and mild Streptococcus pneumoniae pneumonia. Penicillin is the drug of choice to treat syphilis. Pregnant women with allergy to penicillin should be desensitized to receive their full course when being treated for a syphilis infection. Ampicillin and amoxicillin are good choices for enterococcal urinary tract infections, but many other pathogens are resistant, so they should be used selectively. Amoxicillin-clavulanate (Augmentin) combines the ß-lactam with a ß-lactamase inhibitor that expands the spectrum of activity. This combination can be used for sinusitis and urinary tract infections. The extended spectrum penicillins are also safe, but they are much more expensive and generally not used as first-line agents for most disorders during pregnancy. The cephalosporins are safe and used for urinary tract infections, including pyelonephritis, and for gonorrhea. The penicillins can be used safely during breast-feeding. Clindamycin Clindamycin is a macrolide and acts on the bacterial ribosome preventing transcription. It can be used to treat bacterial vaginosis, although metronidazole is the first-line medication. It generally is reserved for anaerobic infections that are not sensitive to other agents. Up to 10% of patients will develop pseudomembranous colitis. Clindamycin is safe during breast-feeding. Metronidazole Metronidazole inhibits bacterial protein synthesis. It is used to treat trichomonas and bacterial vaginosis. This agent was found to be positive in the Ames test but has not been proven to be carcinogenic in humans, nor has it been shown to produce birth defects. Although some authorities suggest deferring use past the first trimester, there are no data supporting this suggestion. This medication is safe in breast-feeding, although the American Academy of Pediatrics recommends interrupting breast-feeding for 12 to 24 hours following a 2-gram dose. Aminoglycosides Aminoglycosides inhibit bacterial protein synthesis. They can be used to treat pyelonephritis but should be used only when serious gram-negative infection is suspected. Maternal administration has been associated with ototoxicity in the fetus leading to hearing loss. Breast-feeding is safe, because little drug passes to the neonate via the breast milk. Trimethoprim-sulfamethoxazole This combination (Bactrim or Septra) inhibits folic acid metabolism and is very active against many organisms that cause urinary tract infections. No definite teratogenic effects have been described, but in 2,296 Michigan Medicaid recipients, first-trimester trimethoprim exposure was associated with a slightly increased risk of birth defects, particularly cardiovascular, and in a retrospective study, the odds ratio was 2.3. Given this and the mechanism of action via the folate pathway, avoidance in the first trimester is prudent. It displaces bilirubin from its protein binding sites in the neonate, potentially contributing to an increased risk of hyperbilirubinemia or kernicterus in newborns. Therefore, it should not be used close to delivery. This theoretic effect has not been substantiated in clinical trials. Trimethoprim-sulfamethoxazole has been used to treat otitis, sinusitis, Shigella colitis, and Pneumocystis carinii infections, in addition to both asymptomatic bacteriuria and acute cystitis. Use of this medication is safe during breast-feeding. Nitrofurantoin Nitrofurantoin inhibits bacterial protein and cell wall synthesis. It is eliminated by excretion, and this bactericidal activity makes it highly effective in treating uncomplicated lower urinary tract infections. It can induce hemolytic anemia in patients with glucose 6-phosphate dehydrogenase deficiency and, because the newborn's red blood cells are deficient in reduced glutathione, the label carries a warning against use of the drug at term. However, hemolytic anemia in the newborn after exposure before birth has not been reported. This medication is compatible with breast-feeding. Erythromycin Erythromycin and azithromycin inhibit bacterial protein synthesis. They often are used as an alternative to the penicillins and are first-line treatment for mycoplasma and chlamydia infections. These medications also are useful in treating community-acquired pneumonia or severe bronchitis. Use of both erythromycin and azithromycin are compatible with breast-feeding. Antiviral Agents The emergence of the human immunodeficiency virus and acquired immunodeficiency syndrome has resulted in development of many antiviral agents. Previously, herpes was one of the few viral infections for which pregnant women might be exposed to treatment ( Table 7.2).

TABLE 7.2. Antiviral agents in pregnancy Acyclovir and Valacyclovir Acyclovir (Zovirax) has resulted in no fetal abnormalities in 601 exposures reported. The Centers for Disease Control and Prevention recommends that pregnant women with disseminated infection (e.g., herpes, hepatitis, or varicella pneumonia) be treated with acyclovir. No human studies during pregnancy have been carried out with famciclovir. Human Immunodeficiency Virus Treatment A variety of agents may be used to treat patients with human immunodeficiency virus infection. The medications generally fall into three categories, the nucleoside reverse transcriptase inhibitors (nRTIs), the nonnucleoside analog reverse transcriptase inhibitors (NNRTIs), and the protease inhibitors (PIs). Zidovudine is the most widely studied nRTI, and no adverse effects have been seen over 4 years of follow-up. In the same class of medications, didanosine, stavudine, and lamivudine also seem to be safe. Nevirapine, an NNRTI, has had no reported adverse effects. The PI class has not been as extensively studied.

UPPER RESPIRATORY TRACT COMPLAINTS The common cold is the most frequent acute illness, and most are self-diagnosed and treated. Medicine used to treat symptoms associated with the common cold are among the most commonly used drugs in pregnancy. Most patients complain of fatigue, malaise, rhinorrhea, nasal congestion, cough, and sore throat. The cold can be caused by a variety of viruses, rhinoviruses, coronaviruses, respiratory syncytial virus, adenovirus, parainfluenza and influenza viruses, and others. Therefore, in the absence of a complicating superinfection with bacteria, antibiotic treatment is not appropriate. The most common treatments are used to alleviate the symptoms listed above and include antihistamines, decongestants, and cough remedies. Antihistamines Most antihistamines are safe during pregnancy. Brompheniramine (Bromfed) was associated with an increased relative risk of malformations in the Collaborative Perinatal Project but not the Boston Collaborative Drug Surveillance Program. Other safe antihistamines include chlorpheniramine, clemastine, diphenhydramine, and doxylamine. There are newer antihistamines with little or no data in pregnancy and are best used as second-line therapy. These include astemizole (Hismanal), cetirizine (Zyrtec), loratadine (Claritin), and fexofenadine (Allegra). Decongestants The most common oral decongestants are all sympathomimetic agents and include pseudoephedrine, phenylephrine, and phenylpropanolamine. There have been some reports of an association with gastroschisis and first-trimester maternal exposure to pseudoephedrine. In the first trimester, an alternative would be to try use of topical preparations, including the nasal decongestants oxymetazoline (Afrin) and phenylephrine (Neo-Synephrine). Cough Suppressants Codeine and dextromethorphan are the most common cough suppressants. Neither has been associated with a teratogenic effect. Most cold treatments including the antihistamines, decongestants, and cough suppressants are safe during breast-feeding.

ASTHMA TREATMENT Although the cold is the most common acute illness during pregnancy, asthma is the most common chronic respiratory condition. About 5% of pregnancies are complicated by asthma, and it may cause increases in preterm birth, low birth weight, and other complications. Whether aggressive and active management reduces these risks to the background level is a controversial issue. Asthma is characterized by airway inflammation and hyperreactivity. Treatment of the asthmatic should start with reduction of environmental factors that worsen the disease. All patients should receive the influenza vaccination yearly. Allergens should be avoided, as should both active and passive exposures to cigarette smoke. For patients who do not respond optimally to these environmental alterations, a variety of pharmacologic treatments are available ( Table 7.3).

TABLE 7.3. Medications to treat asthma

ß-Sympathomimetic Agents The short-acting ß-sympathomimetic agents are the first-line treatment for acute asthma exacerbations. Albuterol inhalers (Proventil, Ventolin) are commonly used. Terbutaline and metaproterenol inhalers are acceptable alternatives. No teratogenic risks have been ascribed to these medications and all are compatible with breast-feeding. For longer term treatment salmeterol, a long-acting ß-sympathomimetic agent, is available and safe. All of the ß-sympathomimetic agents can cause tachycardia and other cardiovascular effects. These are usually mild and self-limited. Glucocorticoids Inhaled glucocorticoids are also first-line therapy. They act by reducing inflammation. Agents include beclomethasone, fluticasone, and others. No teratogenicity for inhaled steroids has been seen, and they are compatible with lactation. Systemic glucocorticoids also can be used for acute exacerbations but may increase the risk for cleft lip and palate up to five-fold. Theophylline Theophylline was at one time a first-line agent for treatment of asthma, but with the emergence of the ß-agonists and inhaled glucocorticoids, its role has been reduced markedly. This agent can be administered intravenously for acute asthma or orally for chronic suppression. The narrow therapeutic window has contributed to this medication falling out of favor. Used for many years, theophylline has shown no evidence of teratogenicity, and it can be used during breast-feeding. Cromolyn Sodium Cromolyn sodium is used for long-term therapy in patients with atopy, functioning as a mast cell stabilizer. Administered by inhaler, there are no known teratogenic or lactation concerns. Leukotriene Receptor Antagonists and Lipoxygenase Inhibitors The leukotriene receptor antagonists (zafirlukast, montelukast) and the 5-lipoxygenase inhibitors affect the inflammatory pathways. These agents are new, and human data are few to none. Given sparse data, if these agents can be avoided during pregnancy and lactation, other drugs should be used.

GASTROINTESTINAL DISORDERS Gastrointestinal problems are extremely common in pregnancy and include nausea, vomiting, hyperemesis gravidarum, gastroesophageal reflux, intrahepatic cholestasis of pregnancy, and inflammatory bowel disease. The clinical picture of several serious disorders may be altered or overlooked during pregnancy and include appendicitis, cholecystitis, pancreatitis, hepatitis, and carcinoma of the gastrointestinal tract. Nausea and vomiting, or morning sickness, occurs in as many as 90% of pregnancies. Nonpharmacologic treatment can include acupressure at the Neiguan point (2 inches proximal to the wrist crease between the tendons of the flexor carpi radialis and palmaris longus muscles), which may be of benefit. Ingestion of ginger appears to reduce nausea and vomiting in pregnancy. Pyridoxine, vitamin B 6 , also appears to reduce symptoms. Pharmacologic treatment can include use of antihistamines, antidopaminergics, and other agents ( Table 7.4).

TABLE 7.4. Drugs commonly used for management of nausea, vomiting, and hyperemesis gravidarum

Antihistamines Doxylamine (Unisom) is an antihistamine that was a component of Bendectin (doxylamine and pyridoxine). An effective treatment for the nausea and vomiting of pregnancy, Bendectin was withdrawn from the American market in 1983 because of unproved allegations that it increased the risk of malformations. Doxylamine can be combined with pyridoxine, reconstituting the two active ingredients of Bendectin. Other commonly used antihistamines include dimenhydrinate (Dramamine); diphenhydramine (Benadryl); hydroxyzine (Vistaril, Atarax), which has both antianxiety and antihistamine properties; and promethazine (Phenergan), which has a central cholinergic blocking activity. Antidopaminergic Agents Several antidopaminergic agents have been used in pregnancy and are probably safe. The list includes prochlorperazine (Compazine), metoclopramide (Reglan), chlorpromazine (Thorazine), perphenazine (Trilafon), droperidol (Inapsine), and haloperidol (Haldol). There tend to me more maternal side effects with these medications than with the antihistaminic agents. There are also some conflicting data about possible minimal increased risks of birth defects with the latter members of

the group. I tend to suggest use of the prochlorperazine or metoclopramide as first-choice agents in this class. These agents are probably safe in breast-feeding, but some require observing the neonate for sedation. Other Agents Trimethobenzamide (Tigan), which provides nausea inhibition at the chemoreceptor level, and ondansetron (Zofran) have been used, but there is less experience with these agents, and certainly the latter is much more expensive with no greater efficacy than standard antihistamines. Reflux or heartburn is a common complaint during pregnancy, particularly in later gestation. Up to 80% of women may have some symptoms of reflux or heartburn. As with hyperemesis, starting with the least invasive environmental or lifestyle changes is prudent. Advice includes elevation of the head of the bed while sleeping, wearing loose clothing, eating frequent small meals that are low in fat, and cessation of smoking. Antacids would then be the first-line therapy ( Table 7.5) and are not known to be associated with any fetal risk. The next class of agents is the histamine receptor antagonists, which can have a reduced bioavailability following antacid use. Therefore, the antacids and the histamine receptor antagonists should be given at least 1 hour apart. The histamine antagonists include cimetidine (Tagamet), famotidine (Pepcid), ranitidine (Zantac), and nizatidine (Axid). There are limited data concerning nizatidine, so the first three would be initial choices. All are compatible with breast-feeding. Promotility drugs can be used and include metoclopramide (described above) and cisapride (Propulsid), about which little has been reported. Sucralfate (Carafate) inhibits pepsin activity and may improve symptoms. The proton pump inhibitors are relatively new agents and, therefore, I would not recommend them. Misoprostol, a prostaglandin E 1 analog, is contraindicated in pregnancy.

TABLE 7.5. Drugs commonly used for gastroesophageal reflux in pregnancy

Intrahepatic cholestasis of pregnancy causes itching of the extremities, trunk, palms, and soles. The pruritus can worsen at night and become severe. In the past, cholestyramine (Questran) was the treatment of choice. Cholestyramine is a nonabsorbable anion exchange resin that binds bile acids, which are elevated in cholestasis. There is now some evidence that ursodeoxycholic acid (UDCA, ursodiol), a minor, naturally occurring hydrophilic bile salt, both reduces maternal pruritus and improves biochemical abnormalities without obvious adverse effects on the newborn. Evidence is accumulating from controlled clinical trials that UDCA is safe. When intrahepatic cholestasis is diagnosed, UDCA coupled with close maternal fetal surveillance is indicated due to potential increases in spontaneous preterm delivery, fetal distress with meconium staining of amniotic fluid, and fetal death. Inflammatory bowel disease in the form of ulcerative colitis and Crohn disease occur commonly during the reproductive years. Sulfasalazine (Azulfidine) is used for the treatment of both ulcerative colitis and Crohn disease. It is composed of 5-aminosalicylic acid and sulfapyridine. It is poorly absorbed from the gastrointestinal tract and is therefore safe in pregnancy. There is a potential for side effects if used during breast-feeding, so the American Academy of Pediatrics cautions women about use of sulfasalazine during lactation. Mesalamine (Asacol) also can be used during pregnancy and may have fewer maternal side effects than the sulfasalazine. Azathioprine is an immunosuppressant and appears safe to use during pregnancy.

ANALGESIC USE Analgesics, both by prescription and over-the-counter purchase, are among the most commonly used medicines in pregnancy. This class of drugs basically falls into two categories, the nonsteroidal antiinflammatory agents and the opioid family. Nonsteroidal Antiinflammatory Drugs Aspirin is one of the nonsteroidal antiinflammatory drugs (NSAIDs) and acts by irreversible inhibition of enzymes in the prostaglandin synthesis pathway. Caution certainly should be advised in use of aspirin beyond the lowest daily dosages, because this drug readily crosses the placenta. First-trimester use has been associated with an increased risk of gastroschisis. Although dosages at or below 100 mg per day have been studied to determine whether there is a reduction in preeclampsia or intrauterine growth restriction without complications, higher dosages have been associated with increased risk of placental abruption. The World Health Organization Working Group on Human Lactation and the American Academy of Pediatrics Committee on Drugs both raise concern about maternal use of aspirin while breast-feeding. Indomethacin and ibuprofen are commonly used NSAIDs that cause competitive and reversible inhibition of prostaglandin synthesis. These NSAIDs can cause constriction of the fetal ductus arteriosus as gestational age progresses and, therefore, their use is not suggested after about 32 weeks gestation. They have not been shown to cause malformations, but use beyond the first trimester can cause oligo- or anhydramnios secondary to direct renal effects. Both indomethacin and ibuprofen are considered compatible with breast-feeding. Acetaminophen is widely used in pregnancy. It crosses the placenta but is considered safe in the usual dosages. It can be used routinely in all trimesters to relieve pain and lower fevers. It is usually the analgesic of choice for a wide variety of aches, pains, and headaches. It is compatible with breast-feeding. Opioid Analgesics Many narcotic preparations are available and are used during pregnancy. They all cross the placenta but have not been associated with malformations when used in usual dosages. Use close to delivery can result in neonatal depression. The common narcotics, codeine, meperidine, and oxycodone, are all compatible with breast-feeding.

PSYCHIATRIC DISORDERS Major depression (15% incidence) and schizophrenia (8%-10% incidence) are very common during the reproductive years. As with any medication, there are concerns about teratogenesis, neonatal withdrawal, or long-term neurobehavioral effects, but these issues must be balanced by the risks to the mother and fetus or infant of withdrawing the necessary medication ( Table 7.6).

TABLE 7.6. Starting dosages for psychotropic medications in pregnancy

The tricyclic agents have been widely used to treat depression and also anxiety, obsessive compulsive disorders, migraines, and other problems. None of the tricyclics has been associated with malformations. If needed, it would be prudent to use the agents with the most accumulated experience, including nortriptyline, desipramine, amitriptyline, and imipramine. There is no evidence to date of clear adverse neonatal effects during breast-feeding, but the American Academy of Pediatrics considers the effects of amitriptyline and imipramine unknown but of concern for the nursing infant. The selective serotonin reuptake inhibitors (SSRIs) include fluoxetine and the newer agents, fluvoxamine, paroxetine, and sertraline. Extensive experience with

fluoxetine shows no clear increased risk of malformations. There is much less experience with the newer agents, but none of the data indicates a teratogenic effect. The effects on the infant of the group of drugs that includes fluoxetine are unknown but may be of concern in relation to breast-feeding. The patient should weigh the strength of her desire to breast feed and the benefits of breast-feeding against the potential effects of continued SSRI use. Other agents are prescribed for depression. Monoamine oxidase inhibitors have not been studied sufficiently to draw a conclusion about their safety. The psychostimulants may cause problems following in utero exposure and are best avoided. St. John's wort, an extract of the plant Hypericum perforatum, has been touted for antidepressant properties, but recent studies have not proven efficacy, so this medication should also be avoided. The mood stabilizers, specifically lithium, valproic acid, and carbamazepine, have all been identified as teratogens. The data on lithium causing malformations have weakened with additional studies. Initial reports found a 5% to 10% risk of malformations with a markedly increased risk of Ebstein anomaly. More recent reports show little, if any, increased risk of malformation and have failed to confirm the specific association with Ebstein anomaly. For the patient with a first-trimester exposure to lithium, targeted ultrasonography in the second trimester is warranted. Lithium also has been associated with hydramnios, possibly secondary to fetal diabetes insipidus. Although there is little data, the American Academy of Pediatrics considers lithium to be contraindicated during lactation. Valproic acid and carbamazepine are both associated with an increased risk of neural tube defects, so second-trimester serum screening for a-fetoprotein and targeted ultrasonography are warranted. Yet many of the neural tube defects are closed and difficult to detect. The antianxiety agents generally fall into the benzodiazepine family. There have been reports of an increased risk of cleft lip after exposure not substantiated in other reports. Although the odds ratio may have been increased, the absolute rate of clefting would still be low given a rate of only 0.06% in the normal population. The American Academy of Pediatrics states that the effects on the neonate during lactation are unknown but may be of concern. The antipsychotic agents include the butyrophenones (including haloperidol) and phenothiazines. No clear teratogenic effect has been seen with either group. Haloperidol is classified as having an unknown but possibly concerning effect if used during lactation.

VITAMIN AND MINERAL USE Prenatal use of multivitamins, often with additional iron, is commonly advocated by many health care providers for pregnant women. There is clear evidence that the folate component reduces the first occurrence and recurrence rate of neural tube defects. Other studies point to possible reduction in cardiac and urinary tract abnormalities. For the woman who eats a balanced diet, most of the recommended daily allowances should be obtained except for folate and iron. However, there are multiple at-risk populations that may not attain the goals, including those patients with eating disorders, vegetarians, the poor, substance abusers, women carrying multiple gestations, and so on. The only clear teratogenic vitamin is vitamin A which, when used at over 10,000 IU per day, can cause cranial neural crest anomalies. It would be prudent to not exceed 5,000 IU per day in supplementation. Supplementation of iron can improve hematocrit at the time of delivery and 6 weeks postpartum. Some studies have shown a benefit of calcium in reducing the risk of gestational hypertension and preeclampsia, although other studies fail to confirm this finding. Zinc may increase birth weight and head circumference in populations with zinc deficiency, but probably not in the United States.

RECREATIONAL DRUG USE Recreational drug use is a major problem in the United States. Anonymous testing studies have found rates of positivity from 7.5% to over 13% in a variety of populations, the drugs found including opiates, cocaine, or cannabinoids. As concerning as the illicit drugs are, the effects of tobacco also are well described and of great impact. Some studies suggest that one sixth of low birth weight incidence could be prevented if women stopped smoking during pregnancy. Nicotine reduces uteroplacental blood flow and increases the risk of preterm birth, low birth weight, and sudden infant death syndrome. Maternal alcohol use can cause the fetal alcohol syndrome characterized by craniofacial changes and impaired cognitive development. Although the full-blown picture can be seen with excessive consumption, no safe level of alcohol use has been established. The effects of recreational opioid use appear to be confined to an increased risk of growth restriction, intrauterine fetal death, and neonatal withdrawal. No clear long-term effects have been proven. Marijuana use during pregnancy has not been linked with any clear teratogenic effect or long-term developmental consequences. Cocaine use has been associated with an increased risk of placental abruption, preterm premature rupture of membranes, and low birth weight. A variety of congenital abnormalities have been described with maternal cocaine use, but no definite connection has been established. A variety of neurobehavioral effects have been described also, but long-term problems are not clear. Amphetamine, described as a stimulant for depressed patients, is becoming a commonly abused drug. Although no clear pattern of malformations has been seen, increases in cleft lip and palate have been seen in some, but not all, studies. Concern has been raised about impact on long-term physical growth and intellectual and behavorial development.

ANTICONVULSANT AGENTS Epilepsy is the most common neurologic disorder in pregnancy. Five percent of the population report having had a seizure at some point in their lives. All antiepileptic drugs (AEDs) cross the placenta and, therefore, have potential for teratogenicity ( Table 7.7). Given the incidence of epilepsy, 1 in 250 fetuses is exposed to an AED. Recent studies are making it clear that the AEDs are responsible for the congenital malformations found in the offspring of pregnant women with epilepsy, not the epilepsy, itself, as has been conjectured in the past.

TABLE 7.7. Anticonvulsants

Phenytoin (Dilantin) is a hydantoin AED. The fetal hydantoin syndrome includes a constellation of anomalies including craniofacial, limb, and neonatal growth and performance delays. The risk of teratogenicity is about 2 times the background risk. Phenytoin is compatible with breast-feeding. Carbamazepine (Tegretol) causes a group of defects similar to the fetal hydantoin syndrome, in addition to the increased risk of spina bifida. The risk of spina bifida is about 0.5% to 1.0%. Carbamazepine is compatible with breast-feeding. Phenobarbital is in the barbiturate class. It has been associated with findings similar to the hydantoin syndrome, as well as with congenital heart defects and orofacial

clefting. Breast-feeding is acceptable unless the infant becomes sedated, in which case breast-feeding should be discontinued. Valproic acid (Depakote) has a 1% to 2% risk of causing spina bifida. These neural tube defects tend to be in the lumbosacral area. Valproic acid use also has been associated with cardiac defects, orofacial clefting, and genitourinary anomalies. There is a fetal valproate syndrome that includes facial, central nervous system, and limb anomalies. Breast-feeding is permissible while using this medication. Several newer AEDs have been developed and include felbamate, gabapentin, lamotrigine, and others. Many patients make their first visits to prenatal clinics while taking these medications and having completed the first trimester. We counsel them that, at present, there is no evidence of teratogenicity but that little information is available.

HEADACHES Headache is a very common problem in pregnancy. Evaluation of headaches requires categorizing the headache as primary or secondary. Secondary headache is due to another condition such as flu, while in the primary category the headache, itself, is the disorder. Primary headaches include migraine, tension-type headache, and cluster headache. Criteria have been developed for differentiating these categories and include frequency, degree of pain, and location of pain. Sumatriptan is a selective serotonin receptor agonist. There is no evidence that it is a human teratogen, based on limited numbers of patients reported to the Sumatriptan Pregnancy Registry maintained by GlaxoWellcome and patients who contacted a teratogen hotline. Even less data is available about three newer drugs in this class: naratriptan, zolmitriptan, and rizatriptan. ß-Adrenergic blockers, such as propranolol have been used as preventive therapy. ß-Blockers, calcium channel blockers, and many of the antidepressants discussed earlier have been used as preventive therapy and appear safe in pregnancy and lactation.

ANTINEOPLASTIC AGENTS Cancer is relatively common during pregnancy, complicating 1 in 1,000 to 1 in 1,500 pregnancies. The most common malignancies include carcinoma of the cervix and breast, lymphoma, melanoma, leukemia, and carcinoma of the ovary and colon. Chemotherapeutic agents act on rapidly dividing cells and are, therefore, potentially harmful to the fetal tissue. The trimester of exposure is, of course, critically important to what potential effect the drug may have. Teratogenesis is a concern in the first trimester, but impact on the continued development of the brain in the second and third trimester is of import. Delivery planning is also important, because both maternal and neonatal blood counts may be adversely affected by the chemotherapy. For the most part, chemotherapeutic drugs are secreted into the breast milk, making breast-feeding contraindicated for the woman receiving this therapy. Many classes of drugs can be used as chemotherapy. The alkylating agents cross-link DNA. These drugs, including busulfan, chlorambucil, cyclophosphamide, and nitrogen mustard, are all considered teratogens in the first trimester. The antimetabolites are also teratogenic, possibly due to their effects on folic acid metabolism. This class includes aminopterin and methotrexate. Other antimetabolites that do not affect folic acid metabolism, such as the pyrimidine antagonist 5-fluorouracil, 6-mercaptopurine, and cytarabine, have much less frequently been associated with birth defects. The taxanes, including paclitaxel, have not been used in pregnancy enough to comment on safety. Cisplatin has been used in pregnancy, but most experience is in the second and third trimesters. Growth restriction is common. The therapy of a pregnant woman with cancer must be individualized and must be based on collaboration among the primary care provider, perinatologist, oncologist, and neonatologist.

ANTICOAGULATION Thromboembolism is a leading cause of morbidity and mortality in pregnancy and postpartum. In is the second most common cause of pregnancy-related maternal mortality in the United States. Anticoagulation is used for thromboembolism, valvular heart disease, inherited thrombophilias, and acquired thrombophilias such as antiphospholipid antibody syndrome. The agents available ( Table 7.8) are the coumarin derivatives, unfractionated heparin, and low-molecular-weight heparin (LMWH).

TABLE 7.8. Anticoagulation in pregnancy

Thromboembolism occurs in 0.5 to 3.0 of every 1,000 pregnancies. Treatment of acute deep vein thrombosis includes bed rest, elevation of the extremity to promote venous return, and heparin. Heparin is used with a target activated partial thromboplastin time (aPTT) of 1.5 to 2.5 times control. For pregnant patients, a switch to subcutaneous heparin takes place after 3 to 5 days, as opposed to the initiation of warfarin in the nonpregnant state. Anticoagulation in a patient with artificial heart valves can be a difficult dilemma. For those with particularly thrombogenic valves it may be necessary to consider oral anticoagulation. For the inherited and acquired thrombophilias, use of unfractionated or LMWH probably is acceptable. Use of the oral anticoagulant warfarin sodium (coumadin) is problematic, because there is a known teratogenic effect. This agent easily crosses the placenta. Warfarin depresses the vitamin K-dependent clotting factors (II, VII, IX, and X). Its effectiveness is measured by the prothrombin time (PT), expressed as an international normalized ratio (INR). The first case of teratogenicity of warfarin was reported in 1966, and this infant had nasal hypoplasia, bilateral optic atrophy, blindness, and mental retardation. In a review of all published cases up to 1980, one sixth had abnormalities, and another one sixth ended in stillbirth or spontaneous abortion. Two thirds of exposed pregnancies had a normal outcome. Some reports cite a teratogenic effect in up to two thirds of fetuses exposed between 6 and 12 weeks gestation. Other reports state that the embryopathy affects as few as 5% to 10% of fetuses exposed in the first trimester. The problem is in differing dosages and nature of neonatal evaluation. It appears that neither unfractionated heparin nor LMWH crosses the placenta to any appreciable degree and is, therefore, not teratogenic. Patients can develop heparin-induced thrombocytopenia at about 2 weeks after initiation of therapy. Osteopenia is a problem with prolonged use. Both of the latter complications are more common with unfractionated heparin. All three classes of anticoagulants are compatible with breast-feeding.

RADIOLOGIC EXAMINATIONS Many women require diagnostic imaging studies during pregnancy. The two main concerns are exposure to ionizing radiation and effects of contrast agents. The vast majority of data derives from relatively few pregnant women exposed to high doses of radiation from the atomic bombs dropped over Hiroshima and Nagasaki. From these few individuals it has been extrapolated that there is a threshold dose of at least 5 rads required to cause concerning effect on the developing fetus. The particular effect depends on gestational age at exposure and radiation dose ( Table 7.9). Lethality may be possible during preimplantation with doses as low as 5 rads, but by 9 days postconception at least 25 to 50 rads is needed. Malformations occur only between about 9 and 60 days postconception, and the threshold is at least 10

rads. Mental retardation is possible between 61 and 104 days postconception, with a threshold of 12 rads.

TABLE 7.9. Risks of irradiation per gestational age

Due to increasing concern about the escalating number of rads of exposure inherent in newer computed tomography (CT) technology, the University of Iowa Department of Radiology has developed a protocol to avoid unnecessary exposures and adequately counsel the pregnant patient. Counseling includes a discussion of a 1.5-fold to 2-fold increased risk in childhood leukemias for a 1-rad to 2-rad exposure in the midtrimester and into the third trimester. It is important to be aware of typical radiation exposures of commonly used examinations ( Table 7.10) but also to enlist the expertise of a radiation physicist to assist in calculation of doses in specific cases.

TABLE 7.10. Typical radiation exposure to the fetus for selected diagnostic studies (1 rad = 1,000 millirads)

Nuclear medicine studies involve exposure to ionizing radiation from a variety of isotopes. In most cases the fetal dose will be less than 1 rad except for sodium iodide, Ga67, iodinated red blood cells, or Ti201. Iodine-131 can cause fetal thyroid damage, goiter, and local effects when concentrated by the fetal thyroid after 70 days post-LMP. Use of iodine-131 to treat hyperthyroidism or ablate the thyroid can result in substantial fetal dosage. The most widely used radioisotope in pregnancy is technetium-99. It can cross the placenta, but the amount of radiation from any routine study is small. Magnetic resonance imaging (MRI) involves exposure to magnetic fields rather than ionizing radiation. There are few studies evaluating the effects of MRI on fetuses. There are also virtually no data on gadolinium, a nonionic contrast agent typically used for MRI evaluations. For this reason, MRI is reserved for situations when it clearly is clinically warranted.

COMPLEMENTARY AND ALTERNATIVE THERAPY Complementary and alternative therapy includes acupuncture, acupressure, massage, aromatherapy, and herbal preparations (phytomedicine). A wide variety of herbs are used but they have received limited scientific evaluation. Thus, safety and efficacy cannot be addressed clearly. Certain ingredients should be avoided altogether and include anthraquinone and berberine, which may stimulate uterine contractions. Many other herbs are thought to have similar properties and should be avoided. Patients should be queried about use of complementary medicines. An effort should be made to determine the constituents of a preparation and then evaluate the safety of each component.

SUMMARY POINTS Disease treatment in the pregnant patient should be similar to that given to other patients. Rather than using the FDA Drug Classification System, references and information in databases should be sought for particular medications in specific clinical situations. Most drugs are safe to use during pregnancy, including most antibiotics and medications to treat common conditions, such as upper respiratory tract and gastrointestinal complaints. A few medications are known teratogens, and the list includes coumadin, lithium, the anticonvulsive medications, several antineoplastic drugs, and vitamin A and its derivatives. Most drugs are safe during lactation because the amounts that appear in breast milk are subtherapeutic, approximately 1% to 2% of the maternal dose. One notable exception is lithium. USEFUL DATABASES AND WEB RESOURCES REPROTEXT, REPROTOX, TERIS, and Shepard's Catalog of teratogenic agents are useful resources. These and other databases can be purchased and installed in personal computers.

http://www.perinatology.com/exposures/druglhttp://www.reprotox.org/.%20Accessed%20December%206,%202002.http://depts.washington.edu/terisweb/.%20Accessed%20December%206,%202002. http://www.rxlist.com/.%20Accessed%20December%206,%202002.ist.htm Organization of Teratology Information Services, http://orpheus.ucsd.edu/otis/links.htm American Academy of Pediatrics. The transfer of drugs and other chemicals into human milk. Pediatrics 1994;93:137–150. http://www.aap.org/policy/0036.html The Organization of Teratology Information Services has a collection of fact sheets on exposure during pregnancy to a variety of diseases, medications, and herbal remedies. http://www.OTISpregnancy.org/fact_sheet.htm. Accessed December 6, 2002. General information regarding drugs and medications during pregnancy: http://www.vh.org/Patients/IHB/ObGyn/MedicinesPregnancy.html. Accessed December 6, 2002. http://www.perinatology.com/exposures/druglist.htm. Accessed December 6, 2002.

http://www.motherisk.org/. Accessed December 6, 2002. The National Library of Medicine, http://www.ncbi.nlm.m.nih.gov/PubMed/ The National Center for Complementary and Alternative Medicine, http://nccam.nih.gov/ The NIH Office of Dietary Supplements, http://dietary-supplements.info.nih.gov/ The American Botanical Council, http://www.herbalgram.org/ Accessed December 6, 2002. The Richard and Hinda Rosenthal Center for Complementary and Alternative Medicine, http://www.rosenthal.hs.columbia.edu/ A drug name and “pregnancy” or “lactation” can be entered into a search engine such as Dogpile ( http://www.dogpile.com/) or Google ( http://www.google.com/).

RECOMMENDED READINGS Altshuler LL, Cohen L, Szuba M, et al. Pharmacologic management of psychiatric illness during pregnancy: dilemmas and guidelines. Am J Psychiatry 1996;153:592–606. American Academy of Pediatrics, Committee on Drugs. The transfer of drugs and other chemicals into human milk. Pediatrics 1994;93:137. American College of Obstetricians and Gynecologists. Complementary and alternative medicine. ACOG Committee Opinion No. 227, Nov 1999. American College of Obstetricians and Gynecologists. Guidelines for diagnostic imaging during pregnancy. ACOG Committee Opinion No. 158, Sept 1995. American College of Obstetricians and Gynecologists. Management of herpes in pregnancy. ACOG Practice Bulletin No. 8, Oct 1999. Bennet PN, ed. Drugs and human lactation. New York: Elsevier, 1988. Berg CJ, Atrash HK, Koonin LM, et al. Pregnancy-related mortality in the United States, 1987–1990. Obstet Gynecol 1996;88:161–167. Buehler BA, Delimont D, van Waes M, et al. Prenatal prediction of risk of the fetal hydantoin syndrome. N Engl J Med 1990;322:1567–1572. Creizel A. Reduction of urinary tract and cardiovascular defects by periconceptional multivitamin supplementation. Am J Med Genet 1996;62:179–183. Delgado-Escueta AV, Janz D. Consensus guidelines: preconceptional counseling, management, and care of the pregnant woman with epilepsy. Neurology 1992;42[Suppl 5]:149–160. Hansen WF, Yankowitz J. Pharmacologic therapy for medical disorders during pregnancy. Clin Obstet Gynecol 2002;45:136–152. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgia, and facial pain. Cephalalgia 1988;8:1–96. Heinonen OP, Slone D, Shapiro S. Birth defects and drugs in pregnancy. Littleton, MA: Publishing Sciences Group, 1977. Hooton TM, Scholes D, Hughes JP, et al. A prospective study of risk factors for symptomatic urinary tract infection in young women. N Engl J Med 1996;335:468–474. Institute of Medicine: Committee on Nutritional Status During Pregnancy and Lactation. Food and Nutrition Board. Nutrition during pregnancy: Part I, weight gain: Part II, nutrition supplements. Washington, DC: National Academy Press, 1990. National Asthma Education Program. Management of Asthma during Pregnancy. Report of the Working Group on Asthma and Pregnancy, 1993. Pub No. NIH 93-3279. Schatz M, Zeiger RS, Harden K, et al. The safety of asthma and allergy medications during pregnancy. J Allergy Clin Immunol 1997;100:301–306. Schoenfeld A, Bar Y, Merlob P, et al. NSAIDs: Maternal and fetal considerations. Am J Reprod Immunol 1992;28:141–147. Shepard TH. “Proof” of human teratogenicity. Teratology 1994;50:97–98. Teratology Society Public Affairs Committee. FDA Classification of drugs for teratogenic risk. Teratology 1994;49:446–447. The North American Pregnancy and Epilepsy Registry. A North American Registry for Epilepsy and Pregnancy, a unique public/private partnership of health surveillance. Epilepsia 1998;39:793–798. Watts DH. Management of human immunodeficiency virus infection in pregnancy. N Engl J Med 2002;346:1879–1891. Yankowitz J, Niebyl JR, eds. Drug therapy in pregnancy, third ed. New York: Lippincott Williams & Wilkins, 2001.

Chapter 8 Ultrasound in Obstetrics Danforth’s Obstetrics and Gynecology

Chapter 8 Keith H. Nelson and Lewis H. Nelson, III

Ultrasound in Obstetrics

GENERAL PRINCIPLES INDICATIONS FIRST TRIMESTER SECOND TRIMESTER DOCUMENTATION AND THE BASIC ULTRASONOGRAPHIC EXAMINATION THIRD TRIMESTER TECHNIQUES OF FETAL TESTING INTERVENTIONAL ULTRASONOGRAPHY FUTURE DIRECTIONS CONCLUSION SUMMARY POINTS RECOMMENDED READINGS Introduction General Principles Documentation and the Basic Ultrasonographic Examination Techniques of Fetal Testing

The use of ultrasonography in obstetrics and gynecology has made sweeping changes in the management of pregnant patients. The purpose of this chapter is to review existing recommendations regarding the use of ultrasonography in pregnancy. The background setting of a patient's progress in pregnancy—from her initial visit confirming an intrauterine pregnancy through the anatomic survey to her third-trimester testing—was chosen. Many excellent textbooks dedicated to obstetric ultrasonography already exist. This chapter is intended to provide information about current ultrasonographic practice, guidance for performing the examination, and assistance in recognizing common anomalies. Future ideas and new technology will be reviewed at the end of the chapter. Most ultrasound providers recognize two categories of examination: a basic ultrasonographic examination that includes a routine anatomic survey, estimate of gestational age, evaluation of placenta and amniotic fluid, and evaluation of maternal anatomy; and a more comprehensive examination that may be targeted based on the abnormal findings in a basic examination. In the past, the terms “level I” and “level II” examination were used erroneously to describe these two categories. Such classification schemes must be avoided to prevent misguided patient and physician expectations for the purposes of the examination. Every ultrasonographic examination should be as complete as possible within the limits of the provider and equipment. An ultrasonogram performed simply for the purpose of sex determination that fails to identify an anencephalic fetus has certainly done only harm. A limited ultrasonographic examination for a specific purpose, such as evaluation of fetal presentation, is appropriate in cases in which the pregnancy has been examined previously or in a clinical emergency.

GENERAL PRINCIPLES Ultrasound images are generated using the same principle as sonar. High-frequency sound waves (3.5-7.5 MHz) pass from the surface of the ultrasound transducer through the tissues of the imaging target. Echoes of the transmitted sound waves return to the transducer. Because the average speed of ultrasound in human tissue is known (1,540 m/s), the distance of the reflector from the transmitter can be calculated using the time required for the echo to return. The resulting depth and signal strength data are plotted on a display screen. The position on the screen corresponds to the depth in the tissue. The brightness of the picture element corresponds to the strength of the return signal. Ultrasound systems display a black background and use increasingly bright shades of gray and white to represent stronger echoes. The acoustic impedance for a tissue determines the degree to which that tissue reflects or transmits ultrasound energy. Because the acoustic impedance of tissue varies by density and consistency, interfaces between dissimilar tissues and within heterogeneous tissues act as reflectors that echo part of the transmitted signal back to the transducer. The degree of dissimilarity between one tissue type and the next determines the strength of the return echo received. Simple fluids such as water or amniotic fluid generate very few echoes, because they are homogeneous. Bone generates very strong echoes because the adjacent soft tissues have very dissimilar acoustic properties, producing a marked acoustic impedance mismatch. The strength of the return echo also depends on the angle between the face of the reflector and the ultrasound beam. As the reflector becomes closer to perpendicular with the transmitting source, a greater proportion of the echo is received. This property is well illustrated by the midline structures of the fetal brain, which are imaged best when perpendicular to the incident beam. The strength of the transmitted signal, which is controlled by the ultrasound operator, also contributes to the strength of the return echo. Transducers spend only a small fraction of the time generating the ultrasound signal compared with the time spent detecting return echoes. The duty factor, expressed as the time spent transmitting divided by the time spent listening, is on the order of 0.01 for modern equipment. When a patient has an ultrasonographic examination lasting 20 minutes, the maternal and fetal total exposure to ultrasound energy is only 12 seconds. This exposure increases when Doppler sonography is employed due to the longer energy pulses. The American College of Obstetricians and Gynecologists (ACOG) and American Institute of Ultrasound in Medicine (AIUM) recommend that ultrasound systems used for obstetrics display the thermal index (TI) and mechanical index (MI) of acoustic output on the screen. The TI is an estimate of the increase in temperature in tissue due to conversion of mechanical energy from the ultrasound beam to heat. Values of the TI less than 1.0 are not of concern for generating significant temperature changes. Temperature increases are most likely to occur during Doppler ultrasonography because of the larger duty factors. The MI estimates the compressive and decompressive forces caused by the ultrasound beam. If these mechanical forces are excessive, microscopic bubbles might form in the tissue being imaged. This bubble formation is called cavitation. As long as the MI values are less than 1.0, cavitation is not a concern. The Food and Drug Administration has set the upper limit of acoustic intensity at 94 mW/cm 2 (spatial peak temporal average) for obstetric ultrasonography. For other applications, ultrasound systems are limited to a maximum output of 720 mW/cm 2 . Furthermore, the systems must display the TI and MI if either or both of these indices can exceed 1.0. Because many systems can be used for both obstetric and nonobstetric ultrasonography, the provider should be aware of the acoustic output and TI and MI of the system in use. Overuse of power Doppler and color Doppler imaging is especially noteworthy as settings in which the recommended maximums could be exceeded. The ALARA principle ( as low as reasonably achievable) has been suggested to govern the amount of fetal exposure to ultrasound energy while obtaining useful image data. The ultrasound provider should also remember that although there are no known bioeffects of diagnostic ultrasonography, future discoveries of adverse effects are certainly possible. As such, obstetric ultrasonography should be used only when indicated ( Table 8.1).

TABLE 8.1. Indications for ultrasonography in pregnancy

Ultrasonographic images are most correctly described using terms that acknowledge sound as the source of the image. Highly sound-reflective structures that produce large reflections and correspondingly bright screen images are called hyperechoic or hyperechogenic structures. Regions that produce very few echoes are termed hypoechoic or hypoechogenic. Regions that are echo free are termed anechoic. The reader is encouraged to avoid terms such as translucent or sonolucent, which are derived from the Latin root lucere, meaning “to shine.”

INDICATIONS The currently accepted ACOG recommendations for ultrasonographic examination in pregnancy are based on the opinion of a National Institutes of Health consensus

panel and are summarized in Table 8.1. Several studies have addressed the question of whether routine ultrasonographic screening for all pregnancies should be performed. Recommendations from ACOG, based on an analysis of existing literature, suggest that women with low-risk pregnancies should not receive screening ultrasonography unless medically indicated. Thus far the evidence indicates that there is no reduction in perinatal morbidity or mortality from routine ultrasonographic screening in the low-risk population. The rate of unnecessary interventions in pregnancies screened with routine ultrasonography is also unchanged. There are not yet enough data to conclude whether fetuses with life-threatening anomalies have an improved outcome because of routine ultrasound use. Although these studies demonstrate that routine ultrasonography in the low-risk pregnancy is not effective in improving perinatal mortality or morbidity, there are articles and editorials challenging those conclusions. Many of these studies offered routine screening to all patients except those who would stand to benefit most. Furthermore, a population agreeing to participate in a randomized study of routine screening does not necessarily constitute an accurate cross-section of the intended study population. The ability to perform a physical examination on the second patient—the fetus—is important. When performed with adequate equipment by properly trained personnel, prenatal ultrasonographic examinations can and do offer important information to the provider and patient. Good evidence exists to demonstrate that fetal anatomic surveys are over 99% specific in detecting anomalies; that is, when ultrasonographic examination findings are normal, the fetus will indeed be normal over 99% of the time. The sensitivity of the anatomic survey, or the probability that an ultrasonogram will demonstrate an anomaly if one is present, varies widely depending on the ultrasound provider, the quality and age of the ultrasound system, the anomaly in question, and the frequency of the anomaly in the practitioner's own population. For this reason, it is recommended that ultrasound specialists perform examinations for genetic screening or evaluation of abnormal scan results.

FIRST TRIMESTER With the advent of extremely sensitive home pregnancy tests which can detect human chorionic gonadotropin (hCG) levels down to 20 to 30 mIU/mL, the early diagnosis of pregnancy is now within reach of the typical consumer. Initial visits to the obstetrician can occur earlier instead of later, sometimes even before the first missed menstrual period. Ultrasonographic examinations in the first trimester should be performed with transvaginal imaging and augmented with transabdominal imaging, if necessary. Although first-trimester detection of an intrauterine gestation is possible, it still is limited by the size of the gestation. Corresponding levels of serum ß-hCG help to determine a threshold level at which sonographic evidence of an intrauterine pregnancy should be detected. Various sources place this level between 1,500 and 2,000 mIU/mL for transvaginal sonography and 3,000 and 5,000 mIU/mL for the transabdominal approach. Protocols have been developed for the management of patients with a positive pregnancy test but no ultrasonographic evidence of intrauterine gestation. This scenario may be of concern for ectopic pregnancy, especially when there is pain or vaginal bleeding. In the clinically stable patient, a repeat ultrasonogram in 3 to 7 days may yield more useful information and help avoid errors in management. Two methods are commonly used for determining the gestational age in the first trimester. These are the mean sac diameter (MSD) and the more accurate crown-rump length (CRL). The MSD is determined by two methods. The method selected depends on the reference table used to determine the gestational age. The first method involves imaging the sac and measuring the longest axis (length) and the largest axis perpendicular to the long axis (width). These two orthogonal measurements are averaged and the value used to estimate a gestational age. The second technique requires the addition of the depth of the sac from an image that is transverse and perpendicular to the length. The three values are averaged and an appropriate table used ( Fig. 8.1). The CRL is measured from an image of the fetus that includes the maximum distance from the cephalic pole to the caudal pole. Although the CRL is regarded as highly accurate in determining the gestational age, the biparietal diameter (BPD) in the late first trimester and before 20 weeks of gestation offers comparable accuracy. The MSD is more useful if no embryo is present in a gestational sac. The identification of a yolk sac or an embryo determines that a gestational sac is present. Using transvaginal ultrasonography, an embryo should be seen by 5 to 6 weeks of gestation as measured from the first day of the last menstrual period. By the time the CRL equals 5 millimeters, cardiac activity should be seen. If the CRL is less than 5 millimeters, the examination should be repeated. Table 8.2 lists the signs of an abnormal gestation. If the patient is at 6 weeks of gestation with a positive pregnancy test and there is no gestational sac visualized in the uterus, an ectopic pregnancy is highly suspect.

FIG. 8.1. Measurement of the mean sac diameter (MSD). The distances measured by the calipers in the longitudinal axis ( A, number one) and the depth ( A, number two) and the transverse diameter measured in (B) are added together and divided by three to produce the MSD. Some authors use only the sum of the longitudinal and depth measurements divided by two.

TABLE 8.2. Signs of an abnormal gestation

The obvious diagnosis of an ectopic pregnancy, in which a small gestational sac with cardiac activity present is seen outside the uterus, can be made even when serum ß-hCG levels are not at the threshold for determination of intrauterine pregnancy. The more difficult ectopic pregnancy cases usually involve abnormal adnexal findings such as ovarian or paratubal cysts, free peritoneal or pelvic fluid, but no definite pregnancy sac. The pseudogestational sac may be seen when the endometrium responds to chemical changes of pregnancy and may resemble a fluid-filled structure within the endometrial cavity. Some authors describe a ring of fire that may be seen around ectopic pregnancies developing in the adnexa when evaluated with color Doppler interrogation or power Doppler imaging ( Fig. 8.2). The ring represents increased blood flow about the developing embryo, but it is possible for a corpus luteal cyst or even normal ovarian tissue to demonstrate such a ring. It is recommended that further diagnostic information be gathered, using sequential determination of serum ß-hCG levels and clinical findings, as long as the clinical situation is stable.

FIG. 8.2. Ring of fire. A corpus luteum (A) can be confused with an ectopic pregnancy (B). The arrow indicates blood flow in the embryo (B). (From Diana M. Strickland, RDMS, RDCS; with permission.) See color figure 8.2.

When the serum ß-hCG levels are abnormally high (greater than 100,000 mIU/mL) for the estimated gestational age by last menstrual period dating or by the uterine size, the diagnosis of molar gestation should be entertained. The classic “snowstorm” description of ultrasonographic imaging in molar gestation has been discarded to reflect the improved resolution of ultrasound transducers. A complete molar pregnancy, in which no fetal tissue is identified, may demonstrate clusters of vesicles on ultrasonographic examination. Identification of a partial molar pregnancy is more difficult, but it may be suggested by the presence of a fetus with a cystic or multicystic placenta ( Fig. 8.3). The fetus might demonstrate numerous anomalies, including findings consistent with triploidy.

FIG. 8.3. Partial mole. The placenta appears cystic and enlarged for gestation. No fetal heartbeat was noted and, although fetal tissue was found at pathology, no chromosome studies were obtained.

During imaging of the early fetus several important features should be visualized. The number of fetuses in the uterus should be reported. A multiple gestation should not be diagnosed until the yolk sacs or embryos can be counted. This will avoid counting the empty gestational sac of an anembryonic pregnancy or blighted ovum as a viable pregnancy. Early gestation may be the best possible time to determine the amnionicity and chorionicity of multiple gestations, information of crucial importance in the later management of these pregnancies. A thick rim of tissue surrounding each fetus is more suggestive of dichorionic, diamniotic gestations. If the intervening tissue is thin or absent, monochorionic, diamniotic, or even monoamniotic gestations are more probable. Conjoined twins may also be detected in the first trimester ( Fig. 8.4). Determination of zygosity is possible only if a monoamniotic with monochorionic or diamniotic with monochorionic configuration is found, in which case the twins are monozygotic. Placentation in multiple gestations is often an important clue in determining zygosity and the risk of twin-to-twin transfusion syndrome. The single placenta associated with monochorionic monoamniotic twinning can be identified during the first and second trimesters. A fused placenta will appear as a single placenta but with the additional finding of dichorionic diamniotic membranes ( Fig. 8.5). Determination of fetal sex also may be helpful in determining zygosity.

FIG. 8.4. Conjoined twins. In (A) the twins are joined at the abdomen (abdominophagus) and separated at the chest (B).

FIG. 8.5. Fused placenta with diamniotic, dichorionic membranes. The fused chorions (open arrow) are noted between the amnions (closed arrow) of each twin. The twins could be either mono- or dizygotic.

The fusion of the expanding amnion with the chorion, obliterating the intervening space, occurs around 12 weeks. Prior to this union, the unfused amnion may be seen as a wispy echogenic membrane surrounding the developing fetus and should not be confused with a gestational sac ( Fig. 8.6). It is theorized that rupture of the unfused amnion followed by settling of the fragmented ends onto the surface of the developing embryo is the source of amniotic band syndrome. These disruptions of normally developing fetal anatomy may lead to amputations of random fetal parts or even to fetal death.

FIG. 8.6. Incomplete fusion of the amnion in the first trimester (A). The amnion in the singleton gestation resembles the membrane in a multiple gestation. However, in a different plane (B) the amnion surrounds the embryo in a more characteristic manner.

Later in the first trimester it may be possible to identify the normal rotation of the fetal gut, which occurs in the proximal part of the umbilical cord ( Fig. 8.7). Such normal development may be confused with an omphalocele, an abnormality of the abdominal wall at the umbilical ring. Completion of gut rotation occurs by week 14, with subsequent resolution of the normal developmental hernia.

FIG. 8.7. Normal rotation of the fetal gut. The double arrow indicates the portion of the umbilical cord transiently occupied by the fetal gut during migration and rotation.

One of the earliest markers for aneuploidy in the developing fetus is an abnormally thick but hypoechoic ridge of skin over the nuchal portion of the skull. This may be detected in the first trimester and is suggestive of, but not diagnostic of, Down syndrome (trisomy 21). Some authors believe that this thickened region represents a resolving cystic hygroma that is characteristic of many fetuses with Down syndrome. Further discussion of nuchal hypoechogenicity can be found in the “Second Trimester” section. The placenta is due more than the typical cursory glance during the first trimester. Early detection of placenta previa is possible but should be interpreted with caution. As the lower uterine segment develops in pregnancy, placentas that were previously low lying or covering the internal cervical os may no longer occupy that precarious location. The placenta may seem to migrate away from the os. The change actually occurs due to preferential placental development in areas of improved blood flow

and regression in poorly supplied regions. Blood flow in the lower uterine segment is less marked than in the fundus, encouraging placental development away from the lower uterine segment. Other abnormalities such as succenturiate lobes and circumvallate or circummarginate placentas may also be seen in the first trimester ( Fig. 8.8). In cases of suspected placenta previa, transvaginal imaging can be performed to confirm the placental location relative to the cervix. Doppler flow studies may also reveal a vasa previa, in which the fetal vessels lie across the internal cervical os. This phenomenon may occur secondary to a succenturiate lobe or velamentous umbilical cord insertion.

FIG. 8.8. Succenturiate lobe. The succenturiate lobe (S) is connected to the placenta (P) by the bridge of tissue and vessels.

Careful imaging of maternal anatomy should not be neglected in the excitement of finding a new pregnancy. Abnormal uterine masses such as leiomyomata ( Fig. 8.9) or maternal embryologic abnormalities such as a bicornuate or septate uterus ( Fig. 8.10) may be identified most easily when the pregnancy is not large enough to occlude good visualization. More unusual developments such as anterior and posterior sacculation of the uterus may be visualized, but these findings are extremely rare. Abnormalities of the adnexal structures in pregnancy may include any of the typically encountered abnormalities in the nonpregnant state. The most commonly encountered ovarian neoplasm in pregnancy is the mature teratoma or dermoid. Hemorrhagic cysts of the corpus luteum have the sonographic appearance of complex adnexal masses, but they usually will resolve over time ( Fig. 8.11).

FIG. 8.9. The fibroid is located anteriorly in the uterine wall and distorts the outer contour of the uterus, which helps to differentiate this from a uterine contraction.

FIG. 8.10. Transverse section of a bicornuate uterus in the first trimester. The gestational sac is in the right horn and the left horn shows decidual changes.

FIG. 8.11. Hemorrhagic corpus luteum cyst. In February (Feb) the cyst is complex, but by April (Apr) it has resolved and a normal ovary is present.

SECOND TRIMESTER A complete fetal anatomic survey can be performed most comprehensively in the late second trimester. This head-to-toe examination of the fetus identifies normal fetal anatomy, many but not all structural abnormalities, and markers for aneuploidy. As previously mentioned, there is not a consensus on the best use of ultrasonography in screening for abnormal pregnancies in low-risk populations. Additional information should be used to identify patients who will benefit from ultrasonographic examination. This information includes a family history of abnormal pregnancies or genetic anomalies, teratogenic exposures, abnormal laboratory findings including the maternal serum a-fetoprotein-human chorionic gonadotropin-estriol triple test, and abnormal clinical findings in the developing pregnancy. In some cases, the second trimester marks the first prenatal visit for patients. Such patients often will depend on ultrasonography for an accurate estimation of the gestational age. This estimation is based on ultrasonographic measurements of the BPD, head circumference (HC), abdominal circumference (AC), and femur length (FL). These values are averaged to generate a composite score that is compared with standards. Some equipment calculates the average automatically. Additional measurements that may improve the accuracy of age determination include other long bone lengths, foot length, cerebellar width, and intraocular distance. The physician must ascertain that only normal measurements are averaged. For example, adding the abnormal measurement of a shortened femur in skeletal dysplasia to the head and abdominal measurements will produce an incorrect gestational age. Two approaches can be used to calculate the gestational age if the anatomy is abnormal. One is to omit the data from the abnormal part. For example, obtaining the estimated fetal weight of an anencephalic fetus using the ultrasound system may require the BPD or HC. Because cranial measurements will be abnormal in this case, an equation that uses only the AC and FL can be used. The other option is to substitute a normal measurement from a table containing such data for the same point in the gestation as the patient. In the case of the anencephalic fetus, the appropriate head measurements for gestational age can be found in a reference table and substituted for the abnormal measurements. The anatomic survey seeks to identify major structures in the developing fetus. During this survey, many abnormal ultrasonographic findings may be discovered. Such findings will lead to referral or subsequent diagnosis of abnormal conditions. Table 8.3 contains only a few of the more commonly encountered findings and anomalies that warrant further investigation.

TABLE 8.3. Abnormal ultrasonographic findings by location and potentially associated diagnoses

DOCUMENTATION AND THE BASIC ULTRASONOGRAPHIC EXAMINATION Standards for performing an ultrasonographic examination can be found on the Web sites of the national organizations involved in ultrasound practice: the American College of Obstetricians and Gynecologists ( ACOG.org), the American Institute of Ultrasound in Medicine ( AIUM.org), and the American College of Radiology ( ACR.org). Because these standards are being updated continually, they are not repeated here. These standards are developed with the intention of improving patient care and assisting the physicians and sonographers performing the examinations. The flow of the basic examination is important because, with a disciplined approach, the examiner will have the opportunity to recognize a majority of congenital defects while obtaining information about the normal pregnancy. The examiner should develop a technique for performing the examination that is logical and consistent with published standards. The technique that follows is not the only method but is one that has proved helpful to the initiate and the experienced. Standard orientation should be used. On a transverse scan, the patient's right is on the left of the image. On a longitudinal scan, the patient's head is placed on the left of the image. If standard orientation is not used, the images should be labeled to provide orientation to the interpreter. Any measurement should be accompanied by an image. Documentation can be by any combination of thermal photographs, videotape, digital images, or radiographic film. Thorough documentation of the ultrasonographic examination is imperative for providing a historical reference for future examinations, images for referrals, and a permanent record for the patient's file. After fetal heart activity is confirmed, begin imaging in the long axis of the uterus and move to the maternal adnexae to look for ovarian and adnexal masses. The large venous complexes in the adnexae during pregnancy should not be confused with pathology. Return to the midline of the uterus, and locate and document the lowest edge of the placenta. Although this image will not help with the lateral location of the placenta, it will document anterior or posterior orientation and the extent to which the edge approaches the cervical os. A transvaginal or translabial ultrasonographic examination may be necessary to locate the lower edge of the very low-lying placenta ( Fig. 8.12). Document the area of the cervix unless the bladder is overly filled. Document the placental umbilical cord insertion site in order to diagnose possible marginal or velamentous cord insertions and to assess for vasa previa. The number of vessels in the umbilical cord should be documented in a cross-section of the cord, at the fetal cord insertion, or by visualization of the umbilical arteries around the bladder ( Fig. 8.13). By this point in the examination, there should be a general impression of amniotic fluid volume. An amniotic fluid index (AFI) is not required in most standards; a subjective determination is acceptable. If more examinations are anticipated during the pregnancy or there is a question about the trend in fluid volume, the AFI provides consistency among examiners.

FIG. 8.12. Low-lying, posterior placenta. In (A) the placenta (P) could be mistaken for a posterior placenta previa. Transvaginal ultrasound (B) demonstrates the lower border of the placenta (P) separate from the internal os (OS) by the distance indicated between the plus signs (+).

FIG. 8.13. The umbilical arteries around the fetal bladder frequently can be seen without the use of color Doppler as in (B).

Document the lie and presentation of the fetus with a longitudinal scan of the fetal axis. Subsequent imaging of the fetus transverse to the longitudinal fetal axis is usually most helpful. Begin with the fetal head. This provides an opportunity to diagnose cranial defects, which are both common and serious, and to locate the anatomy for the head measurements (BPD and HC) used in determining gestational age ( Fig. 8.14). A normal cavum septum pellucidum eliminates the diagnoses of agenesis of the corpus callosum, hydranencephaly, hydrocephaly, holoprosencephaly, and anencephaly ( Fig. 8.15). The plane used for measurements should be documented. The choroid plexuses are near the anatomic plane for the BPD. Choroid plexus cysts will appear as smooth-walled anechoic spaces within the echogenic choroid plexus ( Fig. 8.16). The risk of a chromosomal anomaly with the isolated finding of choroid plexus cysts during a routine ultrasonographic examination is less than the risk associated with amniocentesis, regardless of the number or persistence of the cysts.

FIG. 8.14. Reference plane for the biparietal diameter and head circumference. The cavum septum pellucidum is a major landmark for this reference plane and for anomaly screening.

FIG. 8.15. The cavum septum pellucidum is absent in anencephaly (A) because the cranium and majority of the brain are absent. In holoprosencephaly (B) it is absent because the normal brain midline is not developed.

FIG. 8.16. Unilateral choroid plexus cyst. In the absence of other ultrasonographic findings or risk factors, the chances for trisomy 18 are less than the risk from an amniocentesis. This is also the case even if there are multiple or bilateral cysts (inset).

Posterior and inferior to the BPD plane is the cerebellum. Documentation of a normal configuration will provide assurance that there is a minimal chance of an open neural tube defect ( Fig. 8.17). Slightly more inferior, the nape of the neck is visualized. Abnormalities in this area include cystic hygromas or posterior encephaloceles ( Fig. 8.18). Identification of nuchal translucency, more properly termed nuchal hypoechogenicity, is accomplished in the same plane. Although the finding of nuchal hypoechogenicity in the first trimester appears to be a promising tool to screen for aneuploidy, it should be regarded as investigational. The most recent ACOG recommendation is that nuchal skin measurements should not be used clinically until better standardization is developed and more convincing data support such a practice. The fetal neck is smaller than the fetal head, so any enlargement should be evident. Fetal goiters or teratomas may be seen anteriorly ( Fig. 8.19).

FIG. 8.17. The normal cerebellum (A) has a characteristic “doughnut” shape compared with the elongated cerebellum (B) described as the “banana sign” associated with an open neural tube defect.

FIG. 8.18. A cystic hygroma (A) is more fluid filled than a cephalocele (B), which contains brain tissue and generally contains a septum. Usually the bony defect in the skull can be also be detected in the latter case.

FIG. 8.19. An anterior cervical teratoma is seen in longitudinal section (A) and transverse (B). Although the diagnosis was made prenatally, intubation at delivery was unsuccessful. Tracheal compression is a significant risk with cervical tumors in the fetus and generally is the cause of death.

The upper part of the fetal chest is symmetric and consists mostly of lung tissue and the fetal thymus, which is difficult to image. The normal orientation of the fetal heart lies along a 45-degree left axis relative to the anterior-posterior thoracic diameter. The apex of the heart overlying the fetal stomach is an important relationship. Malrotations and situs inversus can be missed if careful correlation with the fetal lie is not included ( Fig. 8.20). A lack of homogeneous lung tissue with or without displacement of the heart suggests cystic adenomatoid malformations, diaphragmatic hernias, teratomas, or other lesions. The fetal heart can still maintain a normal 45-degree axis but be displaced laterally by these and other anomalies ( Fig. 8.21). Documentation of a normal four-chamber view eliminates the majority of fetal heart defects ( Fig. 8.22). A short-axis view is not currently a requirement of a basic examination but increases the sensitivity for detecting cardiac defects.

FIG. 8.20. Situs inversus. Using standard orientation, in a transverse image (A), the maternal right (R) and left (L) appear on the left and right of the image, and in the longitudinal image (B), the maternal feet are to the right of the image. This indicates a vertex presentation with the fetal back to the left side of the mother, which means the apex of the fetal heart points to the right side of the fetus in (A).

FIG. 8.21. The heart is displaced from a more central position by the cystic adenomatoid malformation in the right side of the chest, but the normal 45-degree axis of the heart is preserved. The line from the fetal spine to the sternum normally should pass through the atria.

FIG. 8.22. A normal four-chamber view (A) compared with the same view in Ebstein anomaly (B) and in transposition of the great vessels (C). RV, right ventricle; LV, left ventricle; RA, right atrium; LA, left atrium; AO, aorta; PA, pulmonary artery. (From Dr. Alfred Abuhamad; with permission.)

With the transducer held in the same orientation for the four-chamber heart view, move inferiorly in the fetus to locate and document the plane for the AC measurement. The fetal gall bladder, which may be seen during this transition, is one of the three anechoic areas normally seen in the fetal abdomen. The other two are the stomach and urinary bladder. The observation of additional areas requires explanation. The fetal liver occupies most of the abdominal cavity in the plane used for the circumference measurement, which is why that measurement is sensitive for fetal weight and disturbances in growth. Below the AC plane, the abdomen consists mostly of fetal bowel and the kidneys posteriorly. An image showing both kidneys is useful for eliminating unilateral agenesis, horseshoe kidney, and multicystic or polycystic kidneys ( Fig. 8.23). The fetal umbilical cord insertion site should be documented. If a gastroschisis or omphalocele is present, further investigation is needed ( Fig. 8.24).

FIG. 8.23. The cysts (c) of the multicystic kidney (A) involve the left kidney. A normal contralateral kidney with normal amniotic fluid volume portends a good prognosis. In (B) the polycystic horseshoe-shaped kidney is accompanied by oligohydramnios, a poor prognosis.

FIG. 8.24. Although the umbilical cord insertion site is normal, a gastroschisis will appear as a mass to the right of the insertion site. B: The umbilical cord inserts into the sac of an omphalocele which, in this case, contains fetal liver (L). Isolated gastroschisis is not associated with an increased risk of chromosomal anomalies, contrary to an omphalocele which has significant associated risk. A smaller omphalocele has further increased risks.

Moving caudally in the fetus, document the fetal urinary bladder. Following that image, rotate the transducer to image the femur. Although documentation of the FL suggests that a portion of the lower extremity has formed, an additional image of the fetal feet is helpful. Be aware of the orientation of the foot with respect to the tibia in order to diagnose clubfoot ( Fig. 8.25). Images of the fetal hands, although difficult and not required in standards, are useful in finding soft markers of aneuploidy such as shortened phalanges, characteristic posturing of the fetal hands, or polydactyly.

FIG. 8.25. The clue to the diagnosis of clubfoot is that the sole of the foot and the toes are in the same plane as the longitudinal axis of the leg.

THIRD TRIMESTER There are a few anatomic structures that may be seen in the second trimester that can be visualized more easily in the third trimester. The fetal adrenal glands can be located on the superior poles of the kidneys. It is usually possible to distinguish the adrenal cortex and adrenal medulla. The lens of the eye and fetal scalp hair are more easily distinguished in the early third trimester. Helpful adjuncts to estimation of gestational age also appear in the third trimester. Ossification of the distal femoral epiphysis occurs predictably around 32 weeks. The proximal tibial epiphysis ossifies around 35 weeks. These two ossification centers of the long bones are seen as prominent echoes distinct from the ends of the long bones. Because fetal growth rate varies widely in the third trimester, estimation of fetal weight alone may not accurately predict the gestational age. Finding the additional ossification centers of the long bones may provide more guidance for the patient who has not entered prenatal care until late in the pregnancy. A worrisome source of third-trimester bleeding is placental abruption, which is mainly a diagnosis made on clinical grounds. In some cases ultrasonography provides information about gestational age and fetal life, allowing for more informed management decisions. In extreme cases, ultrasonography may reveal a portion of an abruption ( Fig. 8.26) or a free-floating placenta and intrauterine fetal death.

FIG. 8.26. Placental abruption. The edge of the placenta (P) has been elevated by the bleeding. A portion of the amnion and chorion attached to the edge appears to be floating in the uterine cavity.

Transvaginal scanning with appropriate caution in cases of undiagnosed vaginal bleeding in pregnancy can be very useful. The transducer should be inserted slowly while the sonographer watches the path of the transducer on the monitor. The ability to guide the tip of the transducer relative to maternal anatomy avoids the pitfalls of blind digital or speculum examination. Placental tissue overlying the internal os is seen clearly ( Fig. 8.27), so the uncertainty of terms such as partial placenta previa or low-lying placenta previa can be avoided. The percentage of placenta overlying the internal cervical os can be reported to provide more information. The largest reportable percentage is 50%, indicating a central placenta previa ( Fig. 8.28).

FIG. 8.27. Transvaginal ultrasonogram of a placenta previa. The placenta (P) covers the internal os. The tip of the transvaginal probe is in the anterior fornix. See color figure 8.27.

FIG. 8.28. Transabdominal ultrasonogram of a central placenta previa. The intersection of arrows indicates the approximate location of the cervix, which cannot always be visualized using transabdominal sonography. The arrows are drawn along the bladder-uterine interface and the axis of the vagina.

TECHNIQUES OF FETAL TESTING The amount of amniotic fluid surrounding the fetus has been shown to be a very sensitive indicator of fetal status. Poor placental blood flow in cases of maternal hypertension or diabetes results in a diminished substrate for fetal kidney filtration and a subsequent decrease in the amount of fetal urine. A low AFI is one of the early indicators of such compromise, although a wide variation in the normal range is possible. Estimation of the fluid index requires measurement of the deepest vertical pocket of amniotic fluid not occupied by fetal parts or umbilical cord. This measurement is performed in each of four quadrants of the uterus. Measurements should be performed with the patient supine and the transducer held perpendicular to the floor. The sum of the four measurements expressed in centimeters equals the AFI. Two approaches have been suggested to define oligohydramnios and polyhydramnios using the AFI. One method is to use fixed values of less than or equal to 5 centimeters for oligohydramnios and greater than or equal to 18 centimeters for polyhydramnios. The second approach is to use gestational-specific percentiles below or equal to the fifth percentile for oligohydramnios and above or equal to the 95th percentile for polyhydramnios. Some authors have also proposed defining adequate amniotic fluid volume as the presence of a pocket of fluid measuring at least 2 centimeters horizontally and vertically. The method selected should depend on the experience and data from each sonographer's institution and further correlation with outcomes. Manning and Platt described the biophysical profile (BPP) as a method for noninvasive monitoring of fetal status. There are five parts to the BPP. Four of these parameters are ultrasonographic variables that are evaluated during a 30-minute observation period: fetal breathing, fetal movement, fetal tone, and amniotic fluid volume. Fetal breathing can be observed by watching the fetal diaphragm or chest wall for characteristic breathing motion. The fetus must breathe continuously for 30 seconds during the 30-minute observation period to receive a normal score. Fetal movement of the limbs or body must occur at least 3 times to be considered adequate. Normal fetal tone is demonstrated by one active extension and flexion of the limbs or trunk. Opening and closing of the fetal hand also qualifies for a normal tone score. Slow extension of the extremity or trunk with incomplete return to flexion is not considered normal. The fourth part of the BPP is measurement of the amniotic fluid volume. Originally the normal fluid volume was described as the presence of at least one pocket of amniotic fluid measuring 1 centimeter in vertical dimension. Manning and colleagues later modified this definition to 2 centimeters measured vertically. Many centers use a normal AFI as a substitute measurement. Each ultrasonographic variable is scored zero (abnormal) or two points (normal). The fifth component of the BPP is a nonstress test, which receives a score of two points when reactive and zero when nonreactive. The best possible BPP score is ten. A detailed discussion of the use of the BPP in antepartum care can be found in Chapter 9. Doppler velocimetry of both maternal and fetal vessels is based on the principle that sound waves returning from a moving reflector will be affected by the direction and speed of that reflector relative to the transmitting or receiving source. A reflector traveling toward the receiver will cause the return signal to be higher in frequency. The opposite is true for a reflector moving away from the receiver. Return echoes from the receding reflector will be lower in frequency than the transmitted pulse. The degree of change in the returning signal can be used to calculate the velocity (speed and direction) of the reflector. By applying this property to the red blood cell mass moving within maternal or fetal vasculature, characteristic profiles were obtained for mother and fetus. Doppler interrogation of the maternal uterine artery may reveal a characteristic notch during diastole, which is a sensitive predictor for preeclampsia during the pregnancy. The umbilical artery Doppler waveforms can be used to evaluate the resistance of the placental circuit. The systolic-to-diastolic (S/D) ratio is defined as the ratio of peak velocity in systole to the velocity nadir in diastole. This value is typically between 1.5 and 2.5 in the third trimester. As placental flow resistance increases with vascular compromise, the S/D ratio also increases. Increased resistance to the point of stopping or reversing diastolic blood flow is worrisome for fetal outcome and should prompt further evaluation ( Fig. 8.29).

FIG. 8.29. Absent diastolic flow with or without reverse flow (R) is an indicator of poor fetal health. S, systole. (From Diana M. Strickland, RDMS, RDCS; with permission.)

INTERVENTIONAL ULTRASONOGRAPHY Ultrasonography is now a standard adjunct for amniocentesis, chorionic villus sampling, percutaneous umbilical blood sampling (PUBS), and other invasive procedures. Amniocentesis for genetic studies generally is performed after 13 weeks gestation. The tip of the amniocentesis needle is visualized during insertion to reduce the number of attempts. Post-procedure determination of fetal heart rate usually is performed to assure that the procedure was well tolerated. Chorionic villus sampling can be performed blindly, but many providers find that the procedure is easier and less risky when transabdominal imaging provides visualization of the uterus and guidance of the catheter. PUBS depends heavily on ultrasonographic imaging to locate the placental umbilical cord insertion for needle placement, for monitoring the fetus during sampling, and for post-procedure monitoring to watch for hemorrhage. Ultrasonographic guidance may be employed during interventional procedures for fetal benefit. Intrauterine fetal blood transfusion can be performed in severe cases of isoimmunization when the fetus is too premature to deliver. Replacement of fetal thyroid hormone in cases of maternal thyroid dysfunction can also be performed. The fetus can then be treated and followed with ultrasonography for resolution of the goiter, indicated by return of the thyroid gland to normal size. Selected obstructions of the fetal urinary tract can be bypassed with percutaneous catheters that are removed after birth. This procedure may also be performed in cases in which fluid-filled areas are found to be compressing thoracic structures ( Fig. 8.30). In cases of twin-to-twin transfusion syndrome, therapeutic drainage of the polyhydramnic twin may permit the pregnancy to progress to a reasonable chance of viability.

FIG. 8.30. Insertion of shunt in unilateral hydrothorax. The anechoic, fluid-filled area is drained by the needle during insertion of the shunt. The baby was delivered at term with no pulmonary complications.

FUTURE DIRECTIONS New developments in the field of diagnostic ultrasonographic imaging include power Doppler imaging, harmonic imaging, three-dimensional (3D) ultrasonography, real-time 3D ultrasonography (often called four-dimensional [4D] imaging), compound imaging, B-flow imaging, and panoramic ultrasonographic imaging. Power Doppler imaging uses the same principle as Doppler velocimetry. However, instead of targeting a single vessel, power Doppler imaging is used over a region of tissue. When the Doppler shift is detected in the return echoes, that region of tissue is assigned flow data that appear on the screen as colors different than those of the surrounding image. Direction of flow is not indicated. The resulting image contains a standard ultrasonographic image of the target tissue, with a color overlay representing vascular flow. This type of imaging may become useful in placental flow studies, analysis of the umbilical cord, or detection of ectopic pregnancy. Use of power Doppler should be monitored carefully, because the acoustic output increases dramatically with its use. Harmonic imaging utilizes the harmonics of the transmitted pulse to receive double the transmitted frequency. This improves the image by removing clutter and artifact and improving resolution. This feature is very useful for the patient whose body habitus makes imaging difficult. Three-dimensional ultrasonography relies on a special transducer that can sample tissue in multiple sections simultaneously and on computer processors that can reconstruct the data as a virtual block of tissue. This block can then be viewed at any angle and in any section. Proponents of 3D ultrasonography advocate its use in first-trimester anatomic surveys, monitoring of cervical cerclage, and evaluation of fetal surface defects such as cleft lip or myelocele. This technology requires good fetal position and a moderate amount of amniotic fluid in front of the targeted tissue in order to create adequate images. Special transducers, for both transvaginal and transabdominal imaging, are required to insonate the tissue of interest. The extra graphics processors and memory required to generate 3D images are the rate-limiting steps to producing real-time 3D ultrasonography. Also known as 4D ultrasonography, real-time 3D ultrasonography is currently in development. This technology requires the same transducers as 3D ultrasonography but uses faster processors to generate 3D images in actual real time. The same limitations apply to fetal position and amniotic fluid. It might be possible eventually to assess the functional status of fetal organs with 4D ultrasonography. The results from current research will provide the evidence for using real-time ultrasonography in different applications. Compound imaging utilizes electronic beam focusing to emit ultrasound energy at angles to the axis of the transducer. A single target will be insonated from numerous different angles by the array of emitters across the transducer face. Shallow reflectors that would otherwise block tissue penetration are bypassed by the angled signals from and to the transducer. Reflectors from deeper in the tissue are imaged more precisely and a more detailed grayscale image is produced. The increased resolution of modern transducers and data processing capabilities of systems have led to the development of B-flow imaging. Rather than using the Doppler effect to calculate velocities of blood flow, the B-mode function actually collects the tiny echoes emitted from blood cells. The resulting image is a real-time anatomic image combined with grayscale flow imaging. Color is not required for the display. An advantage of B-flow is the decreased exposure of the patient to ultrasound energy compared with Doppler interrogation. Panoramic imaging relies on the ultrasound system to link a series of images into a large single image, much as a vacationer would align snapshots into a panoramic landscape photo. Although this mode is not real time, it has the advantage of displaying images much larger than the transducer's field of view.

CONCLUSION In the changing spectrum of obstetric practice, ultrasonography has established a niche that is unlikely to be challenged significantly. It is a safe imaging modality when used properly. Numerous applications exist, from anatomic surveys to guidance during interventional procedures. The skills of the provider and the quality of the equipment have a direct impact on the sensitivity of the examination in detecting anomalies. Although the usefulness of screening every pregnant woman with an ultrasonographic examination is still in question, there is no doubt that under the guidelines of accepted medical indications, ultrasonography improves prenatal care.

SUMMARY POINTS Every ultrasonographic examination should be as complete as possible within the scope of the provider and the equipment. There is no evidence to suggest that medical ultrasonography carries a risk to the mother or fetus; nevertheless, such risks might be identified in the future. An ultrasonographic examination should be performed only when medically indicated. The sensitivity of ultrasonography in detecting fetal anomalies is variable and relies on the experience of the provider and the quality of equipment. Specialists with additional training in ultrasonography, such as a maternal-fetal medicine fellowship, should perform ultrasonographic examinations for abnormal findings or for genetic indications. RECOMMENDED READINGS Introduction American College of Obstetricians and Gynecologists. First Trimester Screening for Fetal Anomalies with Nuchal Translucency. Committee Opinion, 1999. American College of Obstetricians and Gynecologists. Ultrasound in Pregnancy. Technical Bulletin 187, 1993.

General Principles American College of Obstetricians and Gynecologists. New ultrasound output display standard. Committee Opinion No. 180, Nov 1996. Copel JA, Pilu G, Green J, et al. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. Am J Obstet Gynecol 1987;157:648–655. Ewigman BG, Crane JP, Frigoletto FD, et al. Effect of prenatal ultrasound screening on perinatal outcome, RADIUS study group. N Engl J Med 1993;329:821–827. Fleischer A, et al., eds. Sonography in obstetrics and gynecology: principles and practice, sixth ed. New York: McGraw-Hill, 2001. Kremkau, Frederick W. Diagnostic ultrasound: principles and instruments, sixth ed. Philadelphia: WB Saunders, 2002. Manning FA, Platt LD, Sipos L. Antepartum fetal evaluation. Development of a fetal biophysical profile score. Am J Obstet Gynecol 1980;136:787. Nelson, Lewis H. Ultrasonography of the placenta-a review. Laurel, MD: American Institute of Ultrasonography in Medicine, 1994.

Documentation and the Basic Ultrasonographic Examination AIUM Guidelines for performance of the antepartum obstetrical examination, 1994. From AIUM.org. Accessed December 6, 2002.

Techniques of Fetal Testing Moore TR, Cayle JE. The amniotic fluid index in normal human pregnancy. Am J Obstet Gynecol 1990;162:1168–1173. Phelan JP, et al. Amniotic fluid volume assessment with the four-quadrant technique at 36-42 weeks gestation. J Reprod Med 1987;32:540–542.

Chapter 9 Fetal Monitoring Danforth’s Obstetrics and Gynecology

Chapter 9 Catherine Y. Spong

Fetal Monitoring

WHY PERFORM FETAL MONITORING? WHO SHOULD BE MONITORED? WHAT CAN WE MONITOR? Fetal Heart Rate Uterine Activity Fetal State (Tone/Breathing/Movements) Amniotic Fluid Volume HOW DO WE MONITOR? Equipment INTERPRETATION OF THE FETAL MONITOR FETAL HEART RATE INTERPRETATION Baseline Fetal Heart Rate Fetal Heart Rate Variability Periodic Patterns OTHER METHODS AND INTRAPARTUM EVALUATIONS Fetal Scalp Blood Sampling Percutaneous Umbilical Blood Sampling Fetal Scalp Stimulation and Fetal Vibroacoustic Stimulation Fetal Pulse Oximetry (Fetal Oxygenation) SUGGESTED READINGS Asphyxia and Cerebral Palsy Fetal Heart Rate Monitoring Fetal Scalp Stimulation/Vibroacoustic Stimulation Prolonged Decelerations in Fetal Heart Rate Baseline Contraction Stress Test/Oxytocin Challenge Test Modified Biophysical Profile Fetal Scalp Blood Sampling Doppler Ultrasound Fetal Pulse Oximetry Fetal Cardiotocography Plus Pulse Rate-Interval Analysis ST Waveform Analysis

In the beginning of the 19th century, reports on the presence of fetal heart tones were published, and nearly 150 years later, continuous fetal heart rate (FHR) monitoring became a reality. By 1998, electronic fetal monitoring was used in 84% of all U.S. births, regardless of whether the primary caregiver was a physician or a midwife. With the advent of these technologies, fetal monitoring is implemented in nearly all pregnancies, either in the antepartum or intrapartum period. The challenge of fetal surveillance is to identify those fetuses whose physiological defense mechanisms are compromised, in order to be able to act before decompensation has occurred. The goal is to prevent fetal and neonatal morbidity, and especially mortality.

WHY PERFORM FETAL MONITORING? Since its inception, the primary objective of FHR monitoring has been to identify the fetus in distress so that measures might be taken in time to avert permanent fetal damage or death. However, a clear consensus regarding the definition of “fetal distress” has not been established. It has been described as “a condition in which fetal physiology is so altered as to make death or permanent injury a probability within a relatively short period of time,” and is usually considered to denote disruption of normal fetal oxygenation, ranging from mild hypoxia to profound fetal asphyxia. The term hypoxia refers to the reduction of tissue oxygen supply below physiologic levels. Asphyxia, derived from the Greek word meaning “a stopping of the pulse,” implies a combination of hypoxia and metabolic acidosis. Historically, the clinical diagnosis of birth asphyxia has been based on findings such as meconium-stained amniotic fluid, abnormal FHR patterns, low Apgar scores, abnormal blood gases, and neonatal neurologic abnormalities. When present together, these findings are highly suggestive of a recent asphyxial insult. Isolated abnormalities, however, correlate poorly with birth-related asphyxia and subsequent neurologic impairment. In 2002, the American College of Obstetricians and Gynecologists Task Force on Neonatal Encephalopathy and Cerebral Palsy stated the criteria to define an acute intrapartum event sufficient to cause cerebral palsy. Essential criteria (must meet all four): 1. 2. 3. 4.

Evidence of a metabolic acidosis in fetal umbilical cord arterial blood obtained at delivery (pH Hastings Cent Rep 1999;29:S1–S22. Schechtman KB, Gray DL, Baty JD, et al. Decision-making for termination of pregnancies with fetal anomalies: analysis of 53,000 pregnancies. Obstet Gynecol 2002;99:216–222.

Surgical Abortion American College of Obstetrics and Gynecologists. Statement on intact dilatation and extraction. ACOG Statement of Policy. January 12, 1997. Berkowitz RL, Stone JL, Eddleman KA. One hundred consecutive cases of selective termination of an abnormal fetus in a multifetal gestation. Obstet Gynecol 1997;90:606–610. Edwards J, Carson SA. New technologies permit safe abortion at less than six weeks gestation and provide timely direction of ectopic gestation. Am J Obstet Gynecol 1997;176:1101–1106. Hakim-Elahi E, Tovell HM, Burnhill MS. Complications of first trimester abortion: a report of 170,000 cases. Obstet Gynecol 1990;76:129. Hern WM, Zen C, Ferguson KA, et al. Outpatient abortion for fetal anomaly and fetal death from 15–34 weeks' gestation: techniques and clinical management. Obstet Gynecol 1993;81:301–306. Howie FL, Henshaw RC, Naji SA, et al. Medical abortion or vacuum aspiration? Two-year follow up of patient preference trial. Br J Obstet Gynaecol 1997;104:829–833. Munsick RA, Fineberg NS. Cervical dilation from multiple laminaria tents used for abortion. Obstet Gynecol 1996;87:726–729. Owen J, Hauth JC, Winkler CL, et al. Midtrimester pregnancy termination: a randomized trial of prostaglandin E

2 versus

concentrated oxytocin. Am J Obstet Gynecol 1992;167:1112–1116.

Paul M, Lichtenberg ES, Borgatta L, et al. A clinician's guide to medical and surgical abortion. New York: Churchill Livingstone, 1999.

Sawaya GF, Grady D, Kerlikowske K, et al. Antibiotics at the time of induced abortions: the case for universal prophylaxis based on a meta-analysis. Obstet Gynecol 1996;87:884–890. Shulz KF, Grimes DA, Christensen DD. Vasopressin reduces blood loss from second-trimester dilatation and evacuation abortion. Lancet 1985;2:353–356. Singh K, Fong YF, Prassad RN, et al. Evacuation interval after vaginal misoprostol for preabortion cervical priming: a randomized trial. Obstet Gynecol 1999;94:431–434. Singh K, Fong YF, Prassad RN, et al. Vaginal misoprostol for pre-abortion cervical priming: is there an optimal evacuation time interval? Br J Obstet Gynaecol 1999;3:266–269.

Medical Abortion Baird DT. Mode of action of medical methods of abortion. J Am Med Wom Assoc 2000;55[3 Suppl]:121–126. Blanchard K, Clarke S, Winikoff B, et al. Misoprostol for women's health: a review. Obstet Gynecol 2002;99:316–332. Carbonell I, Esteve JL, Varela L, et al. 25 mg or 50 mg of oral methotrexate followed by vaginal misoprostol 7 days after for early abortion. Gynecol Obstet Invest 1999;47:182–187. Creinin MD, Edwards J. Early abortion: surgical and medical options, current problems in obstetrics. Gynecol Fertil 1997;20:1–32. Creinin MD, Vittinghoff E, Keder L, et al. Methotrexate and misoprostol for early abortion: a multicenter trial. 1. Safely and efficacy. Contraception 1996;53:321–327. Creinin MD, Vittinghoff E. Methotrexate and misoprostol versus misoprostol alone for early abortion: a randomized controlled trial. JAMA 1994;272:1190–1195. Creinin MD. Medical abortion with methotrexate 75 mg intramuscularly and vaginal misoprostol. Contraception 1997;56:367–371. Creinin MD, Vittinghoff E, Schaff E, et al. Medical abortion with oral methotrexate and vaginal misoprostol. Obstet Gynecol 1997;90:611–616. Danielson KG, Marions L, Rodriguez A, et al. Comparison between oral and vaginal administration of misoprostol on uterine contractility. Obstet Gynecol 1999;93:275–280. Goldberg AB, Greenberg MB, Darney PD. Misoprostol and pregnancy. N Engl J Med 2001;344:38–47. Grimes DA. Medical abortion in early pregnancy: a review of the evidence. Obstet Gynecol 1997;89:790–796. Ho PC, Ngai SW, Liu KL, et al. Vaginal misoprostol compared with oral misoprostol in termination of second-trimester pregnancy. Obstet Gynecol 1997;90:735–738. National Abortion Federation. Protocol recommendations for use of methotrexate and misoprostol in early abortion. Approved August 30, 1996. Revised December 18, 2000. Newhall EP, Winikoff B. Abortion with mifepristone and misoprostol: regimens, efficacy, acceptability and future directions. Am J Obstet Gynecol 2000;183[2 Suppl]:s44–s53. Pymar HC, Creinin MD, Schwartz JL. Mifepristone followed on the same day by vaginal misoprostol for early abortion. Contraception 2001;64:87–92. Schaff, EA, Fielding SL, Westhoff C. Randomized trial of oral versus vaginal misoprostol at one day after mifepristone for early medical abortion. Contraception 2001;64:81–85. Trupin, SR, Moreno C. Medscape Women's Health eJournal. Available at: http://www.medscape.com/viewarticle/415129. Accessed [add date]. Trupin SR. Abortion. EMedicine Journal. Available at: http://www.emedicine.com/med/topic5.htm. Accessed December 27, 2002. Walker K. Schaff E. Fielding S, Fuller L. Monitoring serum chorionic gonadotropin levels after mifepristone abortion. Contraception 2001;64:271–273.

Complications Adler NE, David HP, Masor BN, et al. Psychological responses after abortion. Science 1990;248:41–44. Brind J, Chinchilli VM. Induced abortion and the risk of breast cancer. N Engl J Med 1997;336:1834 [Letter; comment on, N Engl J Med 1997;336:81–85]; discussion 1835. Dagg PK. The psychological sequelae of therapeutic abortion—denied and completed. Am J Psychiatry 1991;148:578–585. Darney PD, Atkinson E, Hirobayashi K. Uterine perforation during second trimester abortion by cervical dilation and instrumental extraction: a review of 15 cases. Obstet Gynecol 1990;75:441–444. Ferris LE, McMain-Klein M, Colodny N, et al. Factors associated with immediate abortion complications. Can Med Assoc J 1996;154:1677–1685. Grimes DA, Cates W Jr. Complications from legally induced abortion: a review. Obstet Gynecol Surv 1979;34:177–191. Henriques CU, Wilken-Jensen C, Throrense P, et al. A randomized controlled trial of prophylaxis of post-abortal infection: ceftriaxone verses placebo. Br J Obstet Gynaecol 1994;201:610–614. Iles S, Gath D. Psychiatric outcome of termination of pregnancy for fetal abnormality. Psychol Med 1993;23:407–413. Lawson HW, Frye A, Atrash HK, et al. Abortion mortality, United States 1972–1987. Am J Obstet Gynecol 1994;171:1365–1372. Mitchison S. Suicides after pregnancy: study did not show association between induced abortion and suicide. BMJ 1997;314:902 [Letter; comment on, BMJ 1996;313:1431–1434]; discussion 902–903. Stotland NL. The myth of the abortion trauma syndrome. JAMA 1992;268:2078–2079. Woodward G. Intraoperative blood loss in midtrimester dilatation and extraction. Obstet Gynecol 1983;62:69–72.

Psychological Effects Adler, NE, David HP, Major BN, et al. Psychological factors in abortion: a review. Am Psychologist 1992;47:1194–1204. Adler NE. Abortion: a social-psychological perspective. J Soc Issues 1979;35:100–119. Adler N. Abortion and the null hypothesis. Arch Psychiatry 2000;57:785–786. Cozzarelli C, Sumer N. Mental models of attachment and coping with abortion. J Pers Soc Psychol 1998;74:453–467. Major B, Cozzarelli C, Cooper LM, et al. Psychological responses of women after first-trimester abortion. Arch Gen Psychiatry 2000;57:777–784. Major B, Richards C. Personal resilience, cognitive appraisals, and coping: an integrative model of adjustment to abortion. J Pers Soc Psychol 1998;74:735–752. Morgan C, Evans M, Peters JR. Suicides after pregnancy. BMJ 1997;314:902–903. Smith E. A follow-up study of women who request abortion. Am J Orthopsychiatry 1973;43:574–585.

Chapter 32 Pelvic Infections and Sexually Transmitted Diseases Danforth’s Obstetrics and Gynecology

Chapter 32 David A. Eschenbach

Pelvic Infections and Sexually Transmitted Diseases

VULVA Herpes Human Papillomavirus Vestibulitis Furunculosis Bartholinitis Chancroid Granuloma Inguinale Lymphogranuloma Venereum ACUTE URETHRAL SYNDROME VAGINITIS Examination Candidiasis Trichomoniasis Bacterial Vaginosis Toxic Shock Syndrome SYPHILIS CERVICITIS ENDOMETRITIS GONORRHEA Course of the Disease Symptoms and Signs Diagnosis Drug Therapy of Uncomplicated Lower Genital Tract Gonorrhea CHLAMYDIAL INFECTION GENITAL MYCOPLASMAS ANAEROBIC BACTERIA SALPINGITIS Epidemiology Bacteriology Pathogenesis Diagnosis History Physical Examination Laboratory Tests Endometrial Biopsy Ultrasound and Computed Tomography Laparoscopy Examination of the Male Partner Treatment OOPHORITIS GENITAL TUBERCULOSIS Pathogenesis Clinical Forms Diagnosis Treatment SUMMARY POINTS SUGGESTED READINGS

A large number of microbes are present in the female reproductive tract, and pelvic infections are common. Many pelvic infections are sexually transmitted ( Table 32.1). This chapter provides data on usual presentations and updated treatment of pelvic infections.

TABLE 32.1. Sexually transmitted infections

The impact of pelvic infections on women ranges from minor annoyance to serious illness and, rarely, even death. The cost of treating pelvic infections is enormous from both direct medical costs and indirect costs, including time lost from work. Using pelvic inflammatory disease (PID) as an example, previous estimates were that, by 2000, one of every four women who reached reproductive age in the 1970s had an episode of PID. Of women with PID, 25% will be hospitalized, 25% will have major surgery, and 20% will have tubal sterility. Upper genital tract sites (endometrium, fallopian tubes, ovaries) formerly considered sterile are subject to ascending microbial traffic and occasionally infection from lower genital tract microbes. Some microbes preferentially infect certain sites and give rise to characteristic symptoms while other microbes cause few symptoms until major pathologic changes occur or until congenital neonatal infection or male-partner infection ensues. Clinicians should have special knowledge of the infections caused by Neisseria gonorrhoeae, Chlamydia trachomatis, group B streptococci, Treponema pallidum, anaerobic bacteria, bacteria associated with bacterial vaginosis, and Mycobacterium tuberculosis; these infections either are common or potentially produce severe sequelae. It is now appreciated that most viral infections of the genital tract are asymptomatic. Several viruses are common and produce severe disease in both adults and neonates, including herpesvirus, cytomegalovirus, hepatitis B virus, human papillomavirus, and human immunodeficiency virus (HIV).

VULVA Herpes Type-specific serologic assays indicate that one third of women 20 to 45 years of age have been exposed to herpes simplex virus type 2 (HSV-2). Between 60% and

85% of women with HSV-2 antibodies never have a recognized genital infection. Despite the frequency of asymptomatic infection, HSV infection is a common cause of vulvar ulcers. Genital ulcers also are caused by syphilis and chancroid. Ulcers from HSV usually occur 3 to 7 days after exposure. Symptomatic primary (first) genital infections typically consist of multiple vesicles that rapidly produce ulcerations of the vulva that can be exceedingly painful. The cervix and vagina may also be involved, producing a gray, necrotic cervix and profuse leukorrhea. External dysuria is common, and bilateral inguinal lymphadenopathy is usual. Vulvar lesions may last for 3 or more weeks before complete healing. Constitutional symptoms of fever, malaise, headache (i.e., aseptic meningitis), and urinary retention (i.e., myelitis) may persist for a week. After primary infection, latent HSV usually localizes in the sacral ganglion and perhaps the dermis. Periodic asymptomatic viral shedding occurs, particularly in the first 6 months after primary infection. HSV is isolated on 1% of the days with no symptoms or physical evidence of infection. Most patients develop a secondary (recurrent) infection from latent virus weeks to months after the primary infection. Secondary lesions are less painful and more localized and last for a shorter time (3 to 7 days) than the lesions of primary infection. Systemic manifestations are unusual with secondary infection. From 75% to 85% of genital herpes infections are caused by HSV-2, with the remainder caused by HSV-1, the primary cause of oral herpes. The two types of herpes infections are clinically indistinguishable except that genital recurrence is unusual from HSV-1. Vesicles and ulcers contain many highly infectious virus particles, and viral shedding occurs until the lesions disappear. Thus, direct contact with either genital or oral HSV lesions leads to a high rate of infection. Transmission usually occurs by direct contact with ulcerative lesions. Transmission is greatest during a primary infection, intermediate during a secondary infection, and probably least with asymptomatic shedding. The diagnosis of herpes can be made clinically if typical, painful, shallow multiple vulvar ulcers are present. However, many HSV lesions are atypical. Laboratory confirmation of atypical lesions and lesions that appear during pregnancy is best attained by virus isolation (which can usually be achieved within 48 hours) or by polymerase chain reaction (PCR) identification. Other direct HSV identification methods, including Pap smear, fluorescein tagging, and immunoperoxidase staining, are insensitive. Accordingly, a negative direct Pap smear does not exclude HSV infection. Complement fixing and neutralizing antibodies appear within 1 week of the onset of infection; failure of an experienced laboratory to identify antibodies within 3 weeks is evidence against HSV infection. However, even high antibody levels do not protect against recurrent HSV infection, although antibody passively transferred to the fetus offers considerable protection against neonatal infection. The rising incidence of herpes infection and the potentially serious fetal infection caused by HSV make this an important infection in pregnancy. New guidelines for herpes are discussed in Chapter 19. Oral acyclovir (Zovirax), 400 mg three times daily, famciclovir (Famvir), 250 mg three times daily, or valacyclovir (Valtrex), 1 g twice daily all for 7 to 10 days, shortens the ulcerative phase. Antiviral therapy does not eradicate HSV or prevent recurrence. Patients with episodic recurrent herpes should be provided a supply of drugs to take for 5 days beginning with prodromal symptoms or within a day of the lesion appearance. Patients with six or more yearly recurrences may benefit from suppressive therapy of acyclovir, 400 mg twice daily, famciclovir 250 mg twice daily, valacyclovir, 500 mg daily, or valacyclovir, 1 g daily for up to 1 year. The expense of this drug limits routine or prolonged use. Local therapy of genital herpes is limited to pain relief. Local treatment neither penetrates into virus-containing cells nor influences epithelial damage from HSV. Corticosteroids and antimicrobial ointments offer no benefit as they prevent drying and, thus, delay healing. Wet-to-dry therapy is often helpful (i.e., 10-minute sitz bath three or four times daily followed by drying with hair dryer). Human Papillomavirus Genital human papillomaviruses (HPV) are DNA viruses that are distinct from papovaviruses that cause the common wart. HPV thrives in the moist genital area and usually is sexually transmitted. HPV infection is common and typically subclinical. HPV DNA is found in the genitalia in 30% to 45% of women by PCR DNA amplification. The vulva is positive for HPV DNA in more than 40%, and the cervix in over 30%. Only 1% of the women have visible warts, and only 9% have a history of genital warts. The average incubation period for visible warts is 3 months. Genital warts most commonly occur on the labia and posterior fourchette ( Fig. 32.1). They originally appear as individual lesions, although large confluent growths can occur if neglected. Vaginal and cervical warts are even more common than labial warts, although most of these are flat lesions visible only by colposcopy. Over 30 HPV types infect the genital tract. Visible genital warts are usually caused by HPV types 6 and 11; 3% of college women had these types. The flat-wart variant is caused by HPV types 16, 18, 31, 33, and 35 (found in 22% of college women tested) and is visible only by colposcopy. A biopsy of flat or atypical-appearing cervical warts is required to exclude cervical neoplasia. Biopsies of warts also should be performed for pigmented, unresponsive, or fixed lesions, or in immunocompromised patients. HPV types 16, 18, 31, 33, and 35 are associated with high-grade cervical dysplasia and cervical cancer where the HPV DNA is integrated into the cancer cell. Women with flat warts should have frequent Pap smears. At present, routine typing of HPV is not recommended to aid PAP smear interpretation or predict cervical dysplasia.

FIG. 32.1. Condylomata acuminata of the vulva.

Vulvar warts must be differentiated from the less verrucous, flatter growths of syphilitic condyloma latum ( Fig. 32.2) and from carcinoma in situ of the vulva; dark field examination or punch biopsies may be required to differentiate these lesions. Small to medium-sized verrucous lesions can usually be treated with patient-applied podofilox (Condylox), imiquimod (Aldara), or by providers (cryotherapy, podophyllin, or trichloroacetic acid). Intralesional interferon and laser surgery represent alternative regimens. Small amounts of podophyllin (0.25 mL) should be used to avoid severe burns. Podophyllin, imiquimod, and podofilox are contraindicated during pregnancy. Large amounts of podophyllin have produced coma in adults and fetal death in pregnancy. A biopsy should be done on atypical lesions before therapy is initiated because podophyllin causes bizarre histologic changes that persist for months. Cryotherapy, trichloroacetic acid, or laser ablation can be used on vaginal warts during pregnancy. Recurrence rates of 50% probably relate to the failure of these methods to kill the virus in adjacent untreated areas. Severe burns have occurred from the use of 5-fluorouracil (5-FU) to treat warts; as a result its use is not recommended. Large warts may not respond to surgical or laser removal alone but also may require regional interferon therapy. Examination of sexual partners is unnecessary because most are already infected.

FIG. 32.2. Condylomata lata of the vulva and perineum. (From Curtis AH, Huffman JW. A textbook of gynecology, sixth ed. Philadelphia: WB Saunders, 1950, with permission.)

Vestibulitis Patients with vestibulitis characteristically have pain with vaginal penetration (i.e., intercourse or tampon insertion) and, in extreme cases, have difficulty sitting or wearing tight clothing. This condition is frequently treated as vaginitis because acidic vaginal discharge increases local irritation. Patients typically have an erythematous area, most commonly at the 4-o'clock and 8-o'clock positions just outside the hymenal ring. There is no clear evidence that HPV or bacteria cause the inflammation, but an accelerated inflammatory response to Candida is suspected in many patients. Treatment is often not effective, but regimens include topical corticosteroids (i.e., without an alcohol base), local corticosteroid injection, oral tricyclic antidepressants, pelvic floor muscle physical therapy, and, in severe cases, skinning vulvectomy. Furunculosis Hair follicles or areas of hidradenitis in the vulva may become infected by staphylococci or other bacteria, giving rise to pustules. This condition must be distinguished from herpetic and syphilitic lesions. The diagnosis can be made by culture or by the finding of Gram-positive cocci in Gram stains of pus. If only a few small lesions are present, treatment with hot, wet compresses or hexachlorophene scrubbing helps. If a larger area is involved, administration of antistaphylococcal antibiotics is required until infection subsides, which may take weeks. Daily low-dose suppressive antibiotic therapy (e.g., erythromycin, 250 mg) can reduce frequent recurrences. Bartholinitis Two stages of Bartholin gland infection occur. The first is an acute infection of the duct and gland, usually caused by either N. gonorrhoeae or C. trachomatis. If infection causes obstruction of the duct, an abscess stage can result. Anaerobic bacteria can be isolated from most abscesses. Rarely, synergistic vulvar gangrene results from bartholinitis. Cultures and a Gram stain of material expressed from the duct may identify gonococci. Cervical gonococcal and chlamydial cultures should be obtained, and treated, if present. Patients with an abscess usually require abscess marsupialization or incision with placement of a catheter in the abscess cavity for 3 to 6 weeks to establish a new duct. Simple incision and drainage should be avoided, since it does not address the drainage of mucus from a functioning gland. Recurrent infection from vaginal flora and mucus cyst formation are common sequelae of bartholinitis. Chancroid The soft chancre of chancroid is a painful ulcer with a ragged, undermined edge and a raised border. In contrast, the syphilitic chancre is painless and indurated. “Kissing ulcers” on opposing surfaces of the vulva occur. Tender, unilateral adenopathy is common, and node suppuration occurs in about 50% of patients with lymphadenopathy. The incubation period of this sexually transmitted disease (STD) is 2 to 5 days. The infection is caused by Haemophilus ducreyi, a Gram-negative bacterium that forms a school-of-fish pattern on Gram-stain preparation. The organism is fastidious, and it is best identified by culture of material from aspirated lymph nodes or from the chancre using special selective media or PCR tests. The differential diagnosis includes syphilis, genital herpes, and lymphogranuloma venereum. Preferred treatment is azithromycin (Zithromax), 1 g orally, or ceftriaxone sodium (Rocephin), 250 mg intramuscularly, in single doses, or ciprofloxacin, 500 mg twice daily for 3 days or erythromycin, 500 mg three times daily for 7 days. Sexual partners should be examined and treated. Granuloma Inguinale Granuloma inguinale is rare in temperate climates and is usually considered an STD, although gastrointestinal transmission can occur. The initial papular lesion typically ulcerates and develops into a soft, painless, progressive granuloma that may be covered by a thin, gray membrane. The granuloma may spread over the course of many months to involve the anus and rectum ( Fig. 32.3). Lymph nodes are moderately enlarged and painless, but they do not suppurate. The infection can become chronic, and long-standing disease may cause genital scarring and depigmentation, as well as lymphatic fibrosis with consequent genital edema.

FIG. 32.3. Vulval granuloma inguinale of relatively recent origin. Some lesions are separate, others confluent. The margin of lesion is raised and scrolled; the base is granular and covered imperfectly by thin, gray slough. (From Demis DJ, Crounse RG, Dobson RL, et al, eds. Clinical dermatology, Vol 3. Hagerstown, MD: Harper & Row, 1972, with permission.)

Infection is caused by a gram-negative bacillus, Calymmatobacterium granulomatis, which is difficult to culture because it is an intracellular parasite. The identification is usually made from scraped material or a biopsy specimen obtained from the periphery of the lesion. Bipolar-staining bacteria are best identified within mononuclear cells (i.e., Donovan bodies) by Wright or Giemsa staining. Therapy of choice is a 3-week course of doxycycline, 100 mg, or trimethoprim/sulfamethoxazole (double strength) or ciprofloxacin, twice daily. Erythromycin and azithromycin offer alternative therapies.

Lymphogranuloma Venereum The incubation period for lymphogranuloma venereum (LGV) is 2 to 5 days. Thereafter, a transient, primary, painless genital or anorectal ulcer develops. Multiple, large, confluent inguinal nodes develop 2 to 3 weeks later and eventually suppurate. Acute infection may cause generalized systemic symptoms. If untreated, the infection enters a tertiary phase that can lead to extensive lymphatic obstruction. This development, together with continued infection, causes fistulae or strictures of the anal, urethral, or genital area. Women with LGV are particularly susceptible to rectal stricture. Edema and elephantiasis of the external genitalia and lower extremities are other serious sequelae. The infection is caused by the sexually transmitted organism C. trachomatis, an intracellular bacterium. Only L 1-3 Chlamydia serovars, which produce accelerated in vitro tissue destruction, typically cause LGV. The diagnosis can be made by culturing chlamydiae from genital lesions or lymph nodes. The most specific and sensitive serologic test is the microimmunofluorescent antibody test, in which the specific L immunotypes are identified. The results of complement fixation (CF) tests are positive in 95% of patients with LGV, but the CF test lacks specificity; test results are often falsely positive in patients who do not have LGV but have previously been exposed to Chlamydia. High (=1:64 CF) titers offer some specificity. LGV responds to 3-week regimens of doxycycline or erythromycin in the usual doses. Large lymph nodes should be aspirated to avoid chronic drainage. Surgical excision of scarred areas may be necessary.

ACUTE URETHRAL SYNDROME Acute cystitis is present in approximately 50% of women with symptoms of dysuria and urinary frequency. Cystitis is defined by pyuria and midstream urine cultures that contain more than 10 5 organisms per milliliter of coliform or staphylococcal organisms. It is now apparent that about one-half of the remaining symptomatic women also have cystitis, but with less than 10 5 coliforms or Staphylococcus saprophyticus organisms per milliliter of urine obtained by suprapubic aspiration or urethral catheterization. Virtually all of these women have pyuria of eight or more leukocytes per high-power field of urine. The pyuria that occurs among another 25% of women with recent onset of internal dysuria and urinary frequency and negative urine cultures is termed acute urethral syndrome. These patients usually have C. trachomatis. The remaining 25% of patients with these symptoms have no pyuria, bacteriuria, or chlamydial infection and a variety of diseases are involved, including candidal or herpetic vulvitis, or no infection origin. Treatment of acute urethritis consists of therapy for the infectious agent, whether it is coliform or S. saprophyticus cystitis, or C. trachomatis urethritis.

VAGINITIS Vaginitis is the most common reason for a gynecologic visit. Symptoms of vaginitis include increased vaginal discharge, vulvar irritation and pruritus, external dysuria, a foul discharge odor, and a yellow discharge color. However, symptoms are very poor indicators of the specific cause of vaginitis. Women with infectious vaginitis have either an STD (i.e., trichomonads) or a quantitative increase in normal flora (i.e., Candida, Gardnerella vaginalis, anaerobes). At least four types of infectious vaginitis are found: candidal, trichomonal, bacterial vaginosis, and, in children, gonococcal. Every effort should be made to establish the diagnosis of one of these specific infections and to avoid the diagnosis of a nonspecific vaginitis. A specific diagnosis is mandatory to select effective therapy. Treatment of nonspecific vaginitis inevitably fails. Other conditions that may cause excessive vaginal discharge include cervicitis, normal cervical mucus from cervical ectopy, vaginal foreign bodies (most commonly, retained tampons), and allergic reactions to douching or vaginal contraceptive agents. Atrophic vaginitis among postmenopausal women can produce burning and dyspareunia, but an infectious cause is not established. A small amount of vaginal discharge may be normal, particularly midcycle, when large amounts of cervical mucus production produce a clear vaginal discharge. A normal vaginal discharge should not have a foul odor or produce irritation or pruritus. Examination External genitalia may be normal or edematous, erythematous, excoriated, or fissured. Local vulvar disease, especially vestibulitis, must be excluded from a secondary effect of vaginitis. On speculum examination, the vaginal mucosa may be erythematous. Discharge characteristics that are important to observe are viscosity, floccular appearance, color, and odor. Vaginal pH status must be determined. A pH less than 4.5 indicates Candida or a normal vaginal discharge. A potassium hydroxide (KOH) odor test and a microscopic examination consisting of a normal saline and 10% KOH wet mount should be done. A drop of each solution is mixed with discharge. Before placing a cover glass over the two separate drops, the KOH portion is tested for the presence of a fishy amine odor. Microscopic examination of the KOH portion is made for hyphae under the 100× objective, and examination of the saline portion is made for trichomonads and clue cells under the 400× objective. Multiple causes of vaginitis are frequent. Vaginal cultures are not particularly helpful except when used selectively to identify Candida. Microscopy is specific, but only 80% sensitive in identifying various types of vaginitis. When infectious vaginitis is suspected in patients in whom a specific diagnosis cannot be established, a repeat examination should be performed 2 weeks later. Candidiasis The most prominent symptom of candidiasis is vulvar and vaginal pruritus. Increased vaginal discharge is infrequent. Vulvar signs of edema, geographic erythema, and fissures may occur. Classically, the vaginal walls are red and contain adherent, white, curdy plaques. However, most women with candidiasis have atypical symptoms, little discharge, and no erythema. Candida albicans causes about 90% of vaginal yeast infections. Noncandidal species cause the remaining infections. These saprophytic fungi are isolated from the vagina in 15% to 25% of asymptomatic women. Thus, the mere presence of vaginal Candida does not always identify an infection, but large numbers of organisms lead to symptomatic vaginitis. However, severe symptoms develop in some women with only a few organisms but with an accelerated immune response to Candida. Candidiasis occurs because changes in host resistance or immune response allow organism detection or cause inflammation. The most widely accepted risk factors for candidiasis include pregnancy, diabetes, and use of immunosuppressive drugs and broad-spectrum antibiotics. Frequent vaginal intercourse and vaginal douching also are risk factors. Because cellular, not humoral, immunity is required to resist candidal infections, pregnant women and patients receiving immunosuppressive drugs that decrease cellular immunity are predisposed to candidiasis. Candidal overgrowth is also favored by high urine glucose levels that can occur in diabetes or pregnancy. Broad-spectrum antibiotics cause suppression of the normal vaginal and gastrointestinal bacterial flora, allowing fungal overgrowth. The role of oral contraceptives in candidal infection remains controversial. Most women have uncomplicated infection defined as a sporadic, mild, C. albicans infection in a host with normal immunity. About 5% of women have complicated infection defined as recurrent, severe, or non- albicans infection. Many of these occur in immunosuppressed women (pregnant, diabetic, or debilitated or immunosuppressed). Candidiasis is best diagnosed in KOH wet-mounts. Vaginal plaques, vaginal discharge, or vulvar scrapings from the edge of the erythematous border are mixed with 10% KOH ( Fig. 32.4). The mycelial form is usually found only during an infection; mycelia can be identified by KOH wet mount in 80% of cases. The pH of vaginal discharge is normal (i.e., 4.7 or less). Fungi can readily be isolated on various media. However, Candida is part of normal vaginal flora, and a positive culture does not necessarily indicate infection. Cultures for Candida should be limited to KOH wet-mount–negative patients with symptoms or signs of candidiasis. In fact, 50% of women with candidiasis have a negative wet mount, but a positive Candida culture.

FIG. 32.4. Candida albicans growing as hyphae and pseudohyphae within infected tissue (×320). (From Monif GRG. Infectious diseases in obstetrics and gynecology. Hagerstown, MD: Harper & Row, 1974, with permission.)

Local vaginal therapy is used because most antifungal preparations are not absorbed from the intestinal tract. For uncomplicated infection, various intravaginal azole agents used for 3 to 5 days are equally effective for women with primary and infrequent candidal vaginitis. These agents include miconazole, clotrimazole, butoconazole (Femstat), tioconazole (Vagistat-1), and terconazole (Terazol). Azole drugs are not absorbed to any degree from the vagina, and the same local regimens can be used safely in pregnancy. The insertion of boric acid powder in capsules into the vagina is also effective. A one-time dose of fluconazole (Diflucan), 150 mg orally, or itraconazole (Sporanox), 400 mg initially then 200 mg for 2 days, is also effective for those with uncomplicated infection. Patients with complicated infection from severe symptoms need 7 to 14 days of intravaginal azole therapy or fluconazole 150 mg with the dose repeated in 4 days. Oral nystatin administration to decrease gastrointestinal colonization does not markedly improve therapeutic cure rates or diminish recurrence rates. About 15% of male sexual contacts of women with candidiasis have symptomatic balanitis; symptomatic males should be identified and treated to prevent recurrent female infection. The number of non- albicans candidal infections appears to have increased. These infections respond poorly to azole therapy, including fluconazole. Boric acid or nystatin therapy works best. Even gentian violet 1% aqueous solution has a limited place in such patients. Patients with frequently recurrent candidiasis represent the most difficult problem in treatment. Extended 2- to 3-week vaginal therapy, male therapy, and reduction of sugar intake are usually ineffective. A glucose tolerance test and HIV testing should be performed in recurrent or resistant cases to exclude unrecognized diabetes and HIV infection. In addition, some women with candidiasis have other concurrent vaginal infections; a repeat physical and wet-mount examination may clarify the problem. Complicated infection from frequently recurring candidiasis (four or more times a year) should be treated first with standard anticandidal therapy for 2 weeks, followed by suppressive therapy with an intravaginal azole or boric acid twice weekly or an intravaginal azole daily for 5 days once a month, oral fluconazole, 150 mg weekly, or oral itraconazole, 400 mg monthly. Suppressive treatment should continue for at least 1 year; recurrent candidiasis is usually reduced to no or one infection yearly, but recurs in 30% to 40% of patients upon cessation of suppression. Trichomoniasis Characteristic symptoms of trichomoniasis include a profuse, yellow, malodorous, often uncomfortable vaginal discharge with vulvar irritation. Trichomonas vaginalis is a common sexually transmitted organism, present in 3% to 15% of asymptomatic women and in up to 20% of women who attend clinics for STDs. T. vaginalis is most likely identified in symptomatic women with recent acquisition. However, about 50% of all women with trichomoniasis are asymptomatic. Most male contacts of women with trichomoniasis asymptomatically carry the organism in the urethra and prostate. A classic profuse, frothy, yellow vaginal discharge is present in only about one-third of women. The vulva may be edematous and inflamed by the discharge. Sometimes, subepithelial redness of the cervix (i.e., strawberry cervix) is seen with the naked eye; smaller red areas are more commonly identified colposcopically. The discharge in women with symptomatic trichomoniasis often has a pH greater than 4.5 and forms an amine odor with 10% KOH. Motile trichomonads are demonstrated in the saline wet-mount smear ( Fig. 32.5). Trichomonads are larger than white blood cells (WBCs) and have a jerky motility. The wet mount usually also contains many polymorphonuclear leukocytes. Although the wet mount can identify trichomonads with 80% sensitivity among symptomatic women, less than 50% of all women with trichomoniasis by culture have positive wet mounts. Trichomonads also can be seen occasionally on a Pap smear.

FIG. 32.5. Characteristic configuration of a trichomonad seen in wet smear at high-power magnification. (From Monif GRG. Infectious diseases in obstetrics and gynecology. Hagerstown, MD: Harper & Row, 1974, with permission.)

T. vaginalis is an anaerobic protozoan. A culture of this organism is easy to perform but not readily available, and culture should be limited to cases where the diagnosis is suspected, but cannot be confirmed by wet mount. Screening cultures in asymptomatic women are not recommended, except for certain high-risk populations. Women with trichomoniasis should also be cultured for N. gonorrhoeae because of a close association between the microbes. T. vaginalis resides not only in the vagina but also in the urethra, bladder, and Skene glands, so systemic, rather than local, therapy is needed. Metronidazole is effective in treating trichomoniasis; the preferred regimen is 2 g in one dose because of complete patient compliance and high effectiveness. Extended 7-day metronidazole therapy, 500 mg twice daily, does not increase the 95% cure rate of a single dose. Simultaneous treatment of the male sexual partner is recommended. Recurrent trichomoniasis is usually attributable to either a lack of compliance or reexposure to an untreated sexual partner. A single recurrence should be treated with the recommended regimens. However, increasing in vitro and in vivo resistance of T. vaginalis to metronidazole exists, and repeated treatment failure should be treated with metronidazole, 2 g daily for 3 to 5 days. The Centers for Disease Control and Prevention (CDC) has consultation available for patients who fail this regimen (see Sexually Transmitted Disease Treatment Guidelines, 2002). Metronidazole therapy is controversial because of its tumor-causing potential in humans. In animals, large doses (equivalent to 350 to 1,000 human doses) cause tumors. The drug also causes mutation of salmonellae associated with carcinogenic potential. No increased tumor rates were found in a small series of women evaluated for up to 10 years after metronidazole therapy for trichomoniasis. However, these data are only slightly reassuring that the drug does not cause cancer because longer, larger studies are needed to exclude this possibility. A meta-analysis and other studies show no consistent evidence of teratogenesis from use of metronidazole in pregnancy. The CDC Sexually Transmitted Disease Treatment Guidelines, 2002 no longer advise avoiding metronidazole in pregnancy. Treatment of asymptomatic T. vaginalis has not reduced prematurity and should be limited to symptomatic women. Persistent discharge after adequate treatment for trichomoniasis should lead to repeat examination for Trichomonas, candidiasis, and gonorrhea. Bacterial Vaginosis Bacterial vaginosis describes the vaginal condition resulting from overgrowth of both anaerobic bacteria and G. vaginalis. Both anaerobes and G. vaginalis are normal inhabitants of the vagina, but overgrowth of the normal Lactobacillus-dominant flora by these bacteria results in a thin, homogeneous, fishy-smelling, gray vaginal discharge that adheres to the vaginal walls and often is present at the introitus. In contrast to findings in other causes of vaginitis, the vaginal epithelium appears normal, and WBCs are usually not present. The fishy amine odor produced by anaerobes is accentuated when 10% KOH is added to the discharge.

The diagnosis of bacterial vaginosis is based on the presence of three of the following four characteristics of the discharge: pH greater than 4.5, a homogeneous thin appearance, a fishy amine odor with the addition of 10% KOH, and clue cells. Clue cells are vaginal epithelial cells with many organisms attached. The cell border of clue cells is so obscured by adherent bacteria that it is not identifiable. In bacterial vaginosis, 20% to 50% of the epithelial cells are clue cells. Polymorphonuclear leukocytes and lactobacilli are notably absent. Gram stains used for diagnosis rely on a reduction in Lactobacillus morphotypes and an increase in small Gram-negative rods and Gram-positive cocci. Commercially available card tests for high concentrations of G. vaginalis (Affirm), pH and trimethylamine (FemExam), and proline aminopeptidase (PIP Activity Test Card) are available. Cultures are not helpful because anaerobes and G. vaginalis can be recovered from women without bacterial vaginosis. In fact, up to 40% of asymptomatic women without vaginitis carry G. vaginalis. The distinguishing feature of bacterial vaginosis is the 10- to 1,000-fold increased concentration of anaerobic bacteria and G. vaginalis. Factors leading to the overgrowth of anaerobes and G. vaginalis are poorly established but include the absence of Lactobacillus and a new sexual partner. Sexual transmission of the infection is considered a risk factor, but is unproven, especially since the treatment of sexual contacts does not prevent recurrence. Treatment is not advocated for most asymptomatic women with bacterial vaginosis because it can spontaneously disappear. However, the 10- to 1,000-fold increased concentration of potentially virulent bacteria in the vagina is related to upper genital tract infection after surgery. A significant, increased relative risk (RR) of postoperative infection has been reported in patients with bacterial vaginosis following cesarean section (RR = 6), hysterectomy (RR = 3 to 4), and induced abortion (RR = 3). Bacterial vaginitis has also been associated with spontaneous PID and postpartum endometritis after vaginal delivery. Treatment of bacterial vaginosis is particularly beneficial for those undergoing elective surgery. Bacterial vaginosis in pregnancy has also been related to premature delivery (RR = 2 to 4), amniotic fluid infection (RR = 2 to 3), and chorioamnionitis (RR = 2 to 3); treatment of women with prior preterm delivery has reduced the incidence of preterm delivery, but treatment of low-risk asymptomatic women had no effect on preterm delivery. Metronidazole, 500 mg orally twice daily for 7 days, metronidazole gel, intravaginally once daily for 5 days, and clindamycin cream, intravaginally at bedtime for 7 days are the recommended regimens for bacterial vaginosis. Alternative regimens include 2 g oral metronidazole in a single dose, 300 mg oral clindamycin twice daily for 7 days, and clindamycin ovules intravaginally at bedtime for 3 days. Metronidazole and clindamycin are particularly effective against the anaerobes. Fluoroquinolones, tetracycline, sulfonamides, and erythromycin are ineffective. Treatment of the male sexual contact with metronidazole does not prevent recurrent bacterial vaginosis and is not recommended. Toxic Shock Syndrome Toxic shock syndrome is an acute illness caused by toxin-producing Staphylococcus aureus. About 6% of women carry S. aureus in the vagina, but only 2% of women have S. aureus capable of producing the toxic shock toxin. The syndrome is highly associated with menstruation and probably with tampon use, but it has also occurred from S. aureus infection of the breast and endometrium after delivery and from abdominal surgical wounds. Characteristic features include a high fever (>102°F [38.9°C]), a diffuse rash, hypotension, skin desquamation (usually 1–2 weeks later), and a wide variety of systemic effects, including gastrointestinal (vomiting, diarrhea), muscular (myalgia), mucous membrane (hyperemia), renal (elevated blood urea nitrogen or creatinine level), hepatic (enzyme abnormalities), hematologic (thrombocytopenia), and neurologic (disorientation, coma). Vaginal or specific-site cultures recover S. aureus. Blood, throat, and cerebrospinal fluid cultures, together with serologic tests for Rocky Mountain spotted fever, leptospirosis, and measles, are usually indicated to exclude diseases with similar clinical presentations. A vaginal tampon, if present, should be removed. Patients should be hospitalized and, when indicated, given large fluid volumes for blood pressure maintenance. Beta-lactamase–resistant antibiotics are recommended, and if other causes of bacterial sepsis such as meningococcemia cannot be excluded, additional antibiotics are necessary. Other life-supporting measures such as intubation, vasopressor administration, and dialysis are often necessary. The case:fatality ratio has been reduced from 15% to 3% with supportive therapy. Antibiotics are of no proven benefit in the acute stage, but they do reduce recurrence rates from 30% to 5%. Although the effectiveness is uncertain, it is prudent for all women to avoid the prolonged and overnight use of tampons. It is recommended that postpartum women not use tampons for 6 to 8 weeks after delivery. Women with toxic shock syndrome should be warned of recurrent episodes and advised against resuming tampon use.

SYPHILIS Physicians must constantly be aware of possible syphilitic infection, particularly in populations with high rates of HIV infection. Most women with syphilis are asymptomatic and have only serologic evidence of infection. T. pallidum rapidly enters lymphatics after exposure, but a primary chancre lesion usually takes about 3 weeks to develop. The classic chancre ulcer is painless and firm with sharply defined, raised edges; however, most syphilitic ulcers are atypical. Any suspicious genital ulcer should be studied by dark field examination. Serous material expressed from the ulcer base is mixed with saline solution, and because T. pallidum is an anaerobe, this mixture must be immediately placed under a cover slip with the edges occluded by petroleum jelly. Identification of typical spirochetes by dark field microscopy establishes a diagnosis of primary syphilis. Dark field examination of ulcer material from possible syphilitic ulcers should be done on 3 consecutive days. VDRL test or rapid plasma reagin (RPR) and fluorescent treponemal antibody (FTA) serology should be performed for a patient with a suspicious lesion. If the serologic results are nonreactive and spirochetes cannot be demonstrated by dark field examination, serologic tests should be repeated in 1 month. Secondary syphilis appears 6 or more weeks later and is characterized by a symmetric, macular, papular, or papulosquamous rash and generalized, nontender lymphadenopathy. Condylomata lata (see Fig. 32.2) are highly infectious, hypertrophied, wartlike lesions of secondary syphilis that usually occur in moist areas such as the vulva or perineum; they must be distinguished from other vulvar lesions. Superficial, painless mucosal erosions of the mouth or vagina, called mucous patches, develop in one-third of patients. Systemic symptoms of fever, weight loss, and malaise may occur. Serologic tests are positive in the secondary stage. Untreated patients will enter a latent phase of syphilis during which clinical and physical manifestations are absent. Diagnosis in the latent phase is established by serologic tests. Intermittent spirochetal bloodstream invasion may occur in the early latent phase (first 4 years). In pregnancy, the risk of congenital fetal infection in the primary and secondary phases of syphilis is 80% to 95%; the risk during the early latent phase is 70%. During the late latent phase, immunity develops, which reduces blood invasion, and the risk of congenital syphilis decreases to 10%. Congenital syphilis is reported to be on the rise, affecting 1 in 10,000 live-born infants. Fetal or perinatal death occurs in 40% of those with congenital syphilis. About one-third of adult patients with untreated late syphilis manifest central nervous system or cardiovascular symptoms of tertiary syphilis. VDRL and RPR tests detect a nontreponemal, nonspecific reagin antibody. The tests can be titrated, and the titer either falls or disappears after therapy for early or secondary syphilis. Thus, the VDRL test can be used to judge the activity of either a first episode or reacquired infection in a patient with documented syphilis. Treated patients with latent syphilis may retain high, stable VDRL titers. Acute bacterial or viral infections can give rise to acute false-positive serologic reactions that last for up to 6 months. Several conditions, such as aging, addiction to drugs, autoimmune disease, and pregnancy, can lead to chronic, nonspecific, false-positive VDRL reactions. False-positive VDRL titers usually are 1:8 or less. By contrast, the FTA test involves a specific antitreponemal antibody, and false-positive FTA reactions are rare. Patients with a positive VDRL reaction must have a confirmatory FTA test to exclude a false-positive VDRL reaction. Patients with a false-positive VDRL reaction will have a negative FTA reaction. In patients with syphilis, the FTA test remains positive indefinitely, and because the test is not titrated, repeat FTA testing should not be done in a known positive patient. The treatment schedules for syphilis currently recommended by the U.S. Public Health Service's CDC are as follows: Early syphilis: Early syphilis is defined as primary, secondary, or latent syphilis of less than 1 year's duration. The drug of choice is penicillin G benzathine, 2.4 million U intramuscularly in a single dose. Alternative choices for penicillin-allergic patients include 2-week regimens of doxycycline, 100 mg twice daily, or tetracycline, 500 mg four times daily. Ceftriaxone, 1 g daily intramuscularly for 8 to 10 days, may be used if close follow-up can be ensured. Syphilis of longer than 1 year's, or of unknown, duration: Penicillin G benzathine, 2.4 million U intramuscularly each week for 3 successive weeks (7.2 million U total), is the drug of choice in this situation. Alternative choices for the penicillin-allergic include doxycycline and tetracycline for 4 weeks. Intravenous aqueous penicillin G or penicillin G procaine is recommended for neurosyphilis. Erythromycin is not recommended for neurosyphilis. Spinal tap to exclude asymptomatic neurosyphilis is recommended for those with neurologic or ophthalmologic signs, other evidence of active disease (aortitis, gummas), HIV infection, treatment failure, and infection for more than 1 year with a titer of 1:32 or greater. Neurosyphilis and syphilis in HIV-positive patients should be treated by infectious disease specialists. Syphilis in pregnancy: Parenteral penicillin is the only documented efficacious treatment in pregnancy. Treatment with penicillin is the same as for the corresponding stage of syphilis among nonpregnant women. For pregnant patients who are allergic to penicillin, tetracycline is not used because of toxicity and erythromycin is not used because of high failure rates to cure the fetus. Penicillin is so superior to other antibiotics for treating syphilis in pregnancy that pregnant, penicillin-allergic patients should be desensitized. The Jarisch-Herxheimer reaction commonly occurs in early syphilis, and pregnant women should be hospitalized in anticipation of this possibility for monitoring of themselves and the fetus. The reaction is ascribed to the sudden massive destruction of spirochetes by antibiotics; it is marked by fever, myalgia, tachycardia, and occasionally hypotension. The reaction usually begins within 24 hours and subsides spontaneously in the next 24 hours. All patients need to be followed with quantitative serologic tests to monitor treatment results and offered HIV testing. All sexual partners need to be contacted and tested for syphilis.

CERVICITIS Acute cervicitis is defined as the presence of yellow cervical mucopus or an increased number of WBCs in cervical mucus. Symptoms are usually limited to a purulent vaginal discharge. Physical findings include mucopus in the endocervical canal or bleeding after swabbing of the cervix. Organisms that infect the cervical columnar epithelium, C. trachomatis or N. gonorrhoeae, can be isolated separately or in combination from about one-half of women with purulent cervicitis. Unknown microbes cause the other cases. The diagnosis can also be established by the finding of more than 10 WBCs per 1,000× microscopic field. Infectious ulcers of the cervix caused by herpesvirus, syphilis, and chancroid must be distinguished from erosion and the other conditions described in Chapter 33. Depending on the nature of the lesion, Gram stain, Pap smear, culture, dark field examination, colposcopy, and, in some cases, biopsy may be required. If N. gonorrhoeae is found by culture or DNA technique, treatment should be the same as for gonorrhea, including treatment for coexisting C. trachomatis. If N. gonorrhoeae is not found, azithromycin or doxycycline regimens used for C. trachomatis are recommended. Ofloxacin or levofloxacin can also be used.

ENDOMETRITIS Lymphocytes and neutrophils normally appear in the endometrium in the second half of the menstrual cycle; their presence does not necessarily constitute endometritis. However, plasma cells are not normally in the endometrium, as they represent an immune response, usually to a bacterial antigen. Endometritis produces nonspecific symptoms, and it should not be diagnosed unless plasma cells, neutrophils within glands, or a specific causative infection is found. Endometritis may occur in the following situations: puerperal endometritis (see Chapter 19) chlamydial or gonococcal endometritis, often occurring among patients with salpingitis endometritis after endometrial instrumentation or surgery tuberculous endometritis purulent endometritis, occurring in pyometra caused by a cervical stricture or after radium insertion endometritis occurring characteristically in the presence of an intrauterine device (IUD). The fact that chronic endometritis is associated with the use of IUDs is well documented; it results from organisms attached to the IUD surface rather than from the foreign body per se. Transfundal endometrial cultures of hysterectomy specimens from women who had used tailed IUDs for more than a few weeks uniformly recovered bacteria, while cultures from women who did not use an IUD were negative. Bacteria that can be recovered are usually of low pathogenicity, but more virulent intrauterine bacteria occasionally cause malodorous discharge and salpingitis. In addition, an anaerobe, Actinomyces israelii, has been found in Pap smears from about 5% of women using IUDs, but not in those who do not use IUDs. This organism appears to colonize the IUD, and when it is found on a Pap smear, asymptomatic patients should be warned of abnormal discharge or abdominal pain, representing infection. If symptomatic infection occurs, the IUD should be removed, and patients treated with ampicillin for 2 to 4 weeks. Asymptomatic patients should not have the IUD removed because of A. israelii. Chronic plasma cell endometritis in nonpregnant women who do not use an IUD is often related to endometrial infection with C. trachomatis and, to a lesser extent, N. gonorrhoeae. Plasma cell endometritis occurs in up to 50% of women with acute cervicitis and over 80% of women with acute salpingitis. Symptoms are abnormal uterine bleeding and mild uterine tenderness. Untreated endometritis can progress to clinically evident salpingitis.

GONORRHEA Gonorrhea is caused by the Gram-negative diplococcus N. gonorrhoeae which attach only to columnar or transitional cells by pili and are rapidly brought intracellularly by pinocytosis. They attract leukocytes, commonly giving rise to purulent discharge. Gonorrhea is usually sexually transmitted, although organisms can be acquired by neonates passing through an infected cervix, causing gonorrheal ophthalmia. Course of the Disease N. gonorrhoeae in the lower genital tract infects the urethra, Bartholin glands, and endocervix. The anus and rectum can also be infected either from cervical infection or during anal coitus. Urinary frequency, dysuria, and a purulent vaginal discharge are the first symptoms to appear 2 to 5 days after exposure. At least 50% of women with N. gonorrhoeae infection have no symptoms. Many women do not seek medical attention if these symptoms are mild. The discharge occasionally is locally irritating and causes vulvar edema and soreness. Pharyngitis may result from gonorrheal pharyngeal infection. In 2% of infected women, disseminated gonococcal infection occurs, causing fever, septicemia, dermatitis, arthritis, endocarditis, or meningitis, in various combinations. Untreated gonorrhea is associated with premature delivery and premature rupture of membranes. In 10% to 17% of women with untreated gonorrhea, the organisms ascend to produce upper genital tract infection or acute PID ( Fig. 32.6). Acute PID is the most common serious sequela of gonorrhea. The mechanical and antibacterial properties of cervical mucus probably provide a barrier against upward extension, but during menstruation the mucus barrier is lost, and gonococci can disseminate to the uterus and fallopian tubes. A transient endometritis occurs as the organisms pass through the uterine cavity and reach the fallopian tubes, where they produce an acute and usually bilateral inflammatory reaction of tubal mucosa. The tubes characteristically become swollen and reddened when the muscularis and serosa become inflamed. If exudate drips from the fimbriated ends of the tubes, pelvic peritonitis occurs that ultimately can cause peritoneal adhesions. The swollen and congested fimbriae may adhere and produce tubal occlusion.

FIG. 32.6. Mode of transmission of gonococcal pelvic infection. Portal of entry is external genitalia. The organism enters cervix, following mucous membrane, passes up through the uterine cavity, and attacks the fallopian tube. Pelvic peritonitis results from escape of pus from tubal fimbria. (From Wharton LR. Gynecology and female urology. Philadelphia: WB Saunders, 1943, with permission.)

The process can take any of the following courses. With prompt, appropriate antibacterial therapy, the infection may subside with little permanent damage to the reproductive tract. The fimbriae may occlude, producing permanent tubal infertility. The swollen and congested fimbriae may adhere to one another or to the ovary, trapping the exudate in the tube and giving rise to pyosalpinx or, if the ovary becomes infected, a tuboovarian abscess. The mucosal folds may adhere to one another, forming glandlike spaces that are filled at first with exudate and later, as the process becomes chronic, with watery secretion in follicular salpingitis ( Fig. 32.7). If the infection subsides after agglutination of the fimbriae and closure of the distal tube, watery secretion accumulates and distends the tube, forming a hydrosalpinx ( Fig. 32.8).

FIG. 32.7. Follicular salpingitis, an end stage of gonorrheal salpingitis. Mucosal folds are adherent, giving rise to innumerable round or irregular cystlike cavities lined by cuboidal epithelium. (From Kelly HA. Operative gynecology, Vol 2. New York: Appleton, 1898: plate XI; drawing by Max Brodel, with permission.)

FIG. 32.8. Hydrosalpinx. (From Curtis AH, Huffman JW. A textbook of gynecology, sixth ed. Philadelphia: WB Saunders, 1950, with permission.)

Symptoms and Signs Except for the discharge, which can be milked from the urethra or is present in the vagina or cervix, there are few signs of acute gonococcal infection in women. Bilateral, mild to severe lower abdominal pain may occur with acute salpingitis. Pelvic peritonitis can cause pain on movement of the cervix, direct and rebound abdominal tenderness, muscle guarding that prevents abdominal palpation, and tender adnexa to various degrees on bimanual examination. However, a sizable proportion of women with salpingitis experience either mild or no symptoms. In subacute salpingitis, infection continues with signs and symptoms that are even less overt than those of the acute stage. In the end stage of salpingitis, the uterus and the adnexa can be fixed by pelvic peritoneal adhesions. The adnexa may be either adherent to the posterior aspect of the uterus or prolapsed in the cul-de-sac, which may pull the uterus into a retroverted position. Notable features are dyspareunia, sterility, and chronic, aching pelvic pain that increases before menstruation. Diagnosis The diagnosis of gonorrhea depends on a culture of N. gonorrhoeae or identification of N. gonorrhoeae by DNA tests. The finding of intracellular Gram-negative diplococci in the Gram stain of cervical or urethral exudate points to gonorrheal infection, but Gram stains for gonorrhea are insensitive. Gonococcal testing should be performed on women with positive Gram stains, symptoms or signs suggestive of gonorrhea (e.g., cervicitis, undiagnosed vaginitis, dysuria), other STDs, bartholinitis or skenitis, acute lower abdominal pain suggestive of acute salpingitis, or suspected disseminated gonococcal infection, as well as on women with male sexual contacts with gonorrhea. Sites to be tested in order of importance are: cervix, anal canal, pharynx, and urethra. Vaginal discharge should be wiped away from the cervix before mucus is obtained for testing. Genital samples must be cultured on Thayer-Martin or similar media containing antimicrobial agents that inhibit growth of the normal bacterial and fungal flora. Patients with gonorrhea should also receive a test for chlamydial infection because 10% to 30% of women with N. gonorrhoeae also have C. trachomatis. Because gonorrhea is an STD, it usually is present in the male partner. The fact that more than 40% of the male contacts of women with gonorrhea are asymptomatic carriers who otherwise do not seek treatment underscores the importance to identify and treat male sexual contacts. Drug Therapy of Uncomplicated Lower Genital Tract Gonorrhea The CDC-recommended treatment schedules for gonorrhea reflect increased resistance to penicillin and tetracycline and in Asia, the Pacific rim, and the West Coast to quinolones, frequent coexistence of chlamydial infection with gonorrhea, potentially serious complications, and frequently no testing for C. trachomatis. Penicillin is no longer recommended to treat gonorrhea because ß-lactamase production by extrachromosomal plasmids destroys ß-lactam antibiotic activity in a large number of gonococcal strains. Resistance to quinolones will likely increase. The recommended drug therapy regimens are given in Table 32.2. A loading dose of cephalosporin or quinolones is used to inhibit N. gonorrhoeae. Patients allergic to penicillin should receive spectinomycin (Trobicin) or other cephalosporins or quinolones. If C. trachomatis is not excluded, azithromycin or doxycycline is indicated. Quinolones and doxycycline should not be given to pregnant women. Special antibiotic regimens are recommended for patients with complicated gonococcal infections. These regimens are published in the widely circulated CDC Sexually Transmitted Disease Treatment Guidelines 2002. Prophylactic regimens to prevent ophthalmia neonatorum include silver nitrate, erythromycin, and tetracycline applications.

TABLE 32.2. Treatment regimens for gonorrhea and chlamydia

The first-choice regimens are so effective that test-of-cure cultures are not recommended. A rescreening culture in 1 to 2 months is reasonable. Patients treated with alternative regimens should have a test-of-cure culture in 4 to 7 days after therapy. All male sexual contacts need referral for evaluation and treatment. Patients with gonorrhea should have serologic tests for syphilis. Those with incubating syphilis (i.e., seronegative, without clinical signs of syphilis) are likely to be cured by all the regimens mentioned, except for quinolones and spectinomycin. Patients treated with these regimens need a follow-up serologic test for syphilis.

CHLAMYDIAL INFECTION C. trachomatis is a sexually transmitted bacterium that is often associated with gonorrhea. Chlamydiae infect the same tissues and produce the same symptoms and

diseases as gonorrhea. Chlamydial infection causes urethritis, bartholinitis, cervicitis, endometritis, salpingitis, Fitz-Hugh and Curtis syndrome (i.e., perihepatitis), and LGV. Treatment of C. trachomatis in pregnancy reduces premature delivery. Neonates born of mothers with chlamydial cervical infection have up to a 40% risk of chlamydial conjunctivitis and a 20% risk of chlamydial pneumonia. As with gonococcal infection, male sexual contacts have both symptomatic and asymptomatic urethritis. C. trachomatis is an obligate intracellular bacterium that attaches to columnar or transitional epithelial cells, is engulfed by pinocytosis, and remains within a phagosome membrane that protects it from host defense mechanisms. The bacteria replicate until they replace most of the cell and, ultimately, cause the cell to rupture. Infective particles are released into the extracellular space, and the process is repeated. Replication time is a relatively slow 24 to 48 hours, explaining the characteristically long latent period between the time of exposure and the onset of symptoms, which ranges from weeks to months. C. trachomatis is three to five times more common than N. gonorrhoeae in developed countries because it is not routinely sought in asymptomatic patients. Chlamydial infection is most often asymptomatic and frequently not identified until overt infection occurs. Infection is particularly common among teenagers. C. trachomatis is associated with serious sequelae, including tubal infertility and ectopic pregnancy and it appears able to produce permanent tissue damage more readily than N. gonorrhoeae. Permanent tissue damage appears largely related to the immune response to infection. Reinfection, chronic infection, and infection in the presence of antibody to chlamydial heat shock protein are particularly associated with tissue damage. Chlamydial infection should be suspected with acute urethritis, mucopurulent cervicitis, and salpingitis. The rate of chlamydial salpingitis approximates that of gonorrhea. Chlamydial infections can be diagnosed by culture, a direct monoclonal antibody slide test, an enzyme-linked immunosorbent assay, or DNA techniques. New DNA methods using ligase chain reaction or PCR offer both sensitivity and specificity not achieved with older tests. Samples should be taken from the cervix or from urine with DNA tests. Azithromycin and doxycycline are the most effective drugs to treat chlamydial infection (see Table 32.2). Erythromycin, ofloxacin (Floxin), and levofloxacin (Levaquin) are alternatives. In pregnancy, erythromycin and amoxicillin are preferred, but reports document the effectiveness and safety of azithromycin.

GENITAL MYCOPLASMAS Genital mycoplasmas have often been thought a microbe in search of a disease because they are ubiquitous and not highly virulent. Mycoplasma hominis is recovered from the vagina in 15% to 70% of women, and Ureaplasma urealyticum is recovered from 40% to 95% of women. A third isolate, Mycoplasma genitalium, is now associated with endometritis. These organisms are phylogenetically positioned between bacteria and viruses. The most convincing role for mycoplasmas in human female infections is as a pathogen in postpartum fever. Mycoplasmas are recovered from the blood of 10% to 15% of women with postpartum fever, and antibodies to M. hominis are demonstrated in 50% of such women. Their role in salpingitis is less clear. Mycoplasmas are recovered from the tubes of 5% to 15% of women with salpingitis, but in primate model studies, M. hominis produces an adnexitis and not salpingitis. Maternal lower genital tract U. urealyticum is not associated with low birth weight, and treatment of U. urealyticum does not reduce preterm births. However, organisms are found in intraamniotic infection and cause cerebrospinal fluid and lung infection in premature neonates. Recovery of U. urealyticum from fetal tissue of mid-trimester spontaneous abortuses also suggests a relationship with abortion. The role of U. urealyticum in fertility is unsettled, but in some reports, mycoplasmas were not related to infertility or to higher pregnancy rates in infertile women. Both M. genitalium and M. hominis are sensitive to tetracycline. Erythromycin inhibits U. urealyticum in vitro but not M. hominis. However, neither antibiotic very effectively eradicates mycoplasmas from the vagina.

ANAEROBIC BACTERIA Anaerobic bacteria are highly associated with pelvic infections. Multiple anaerobic species, usually together with one or more aerobic bacteria, typically combine to form a polymicrobial infection. Intraabdominal abscess and postoperative, postpartum, and bacterial vaginosis infections are the most important examples of anaerobic infection. Anaerobic bacteria are part of the normal vaginal flora. Although many mechanisms by which anaerobic bacteria become pathogenic are unknown, two mechanisms known to cause anaerobic infection include (a) reduction of the redox potential that occurs with tissue trauma from surgery and (b) antibiotic selection that preferentially inhibits aerobic bacteria. Clinicians can virtually assume the presence of anaerobes in infections with a foul-smelling odor as only anaerobes produce odorous metabolic products. Anaerobes are virtually always isolated from an abscess. Anaerobic infections can also produce gas and cause thromboembolism. The anaerobic bacteria most commonly found in genital infections include anaerobic Gram-positive cocci ( Porphyromonas and Peptostreptococcus species), gram-negative rods ( Prevotella [ P. melaninogenicus, P. bivia], Bacteroides [ B. fragilis], and Fusobacterium species), and Gram-positive rods ( Clostridium species). Cultures should be obtained before antimicrobial therapy is begun. Because anaerobes are part of the normal flora, deep tissue cultures are required that are not contaminated by surface bacteria. Because 48 or more hours are required for anaerobe isolation, antibiotic selection is usually based on clinical signs. Anaerobic infection should be particularly suspected with abscess formation, a foul odor, gas formation, tissue necrosis, sterile cultures from obviously infected sites, and thromboembolism. Antibiotic sensitivity testing is only a rough guide to antibiotic susceptibility, but in vitro and in vivo experience has shown that clindamycin, metronidazole, imipenem/cilastatin sodium (Primaxin), second- and third-generation cephalosporins (cefoxitin sodium [Mefoxin], cefotaxime sodium [Claforan]), and extended-spectrum penicillins (ticarcillin disodium/clavulanate potassium [Timentin], amoxicillin/clavulanate potassium) are effective to treat anaerobic infections.

SALPINGITIS Acute primary salpingitis results when pathogenic bacteria in the cervix invade the fallopian tubes. N. gonorrhoeae, C. trachomatis, normal flora aerobic and anaerobic bacteria, and, perhaps, M. genitalium are known causes of tubal infections. Virtually all primary salpingitis occurs among sexually active, menstruating, nonpregnant women. Gonococcal and chlamydial infections account for 50% to 60% of cases. Tuberculous, parasitic, or fungal salpingitis is rare in industrialized countries. Salpingitis usually occurs without instrumentation or trauma to the genital tract; however, approximately 15% of cases occur after instrumentation (e.g., IUD insertion, dilation and curettage, abortion, hysterosalpingography). Perisalpingitis secondary to acute appendicitis or other intraabdominal infections accounts for less than 1% of cases. Acute salpingitis is a common event that annually develops in up to 1% of women between 15 and 39 years of age. Young, sexually active women between 15 and 24 years of age have the highest rate of infection. This infection has tremendous cost consequences. At least $1 billion is required to treat the 800,000 women with acute salpingitis in the United States annually, and $40 billion is spent to diagnose and treat tubal infertility. Epidemiology Most women are infected with sexually transmitted organisms. The rate of salpingitis is increased in women with multiple sexual partners. The high rate of salpingitis in young women is due to their increased rates of gonorrhea and chlamydial infection. Routine screening for C. trachomatis and N. gonorrhoeae has reduced salpingitis in Europe and the United States. Sexually active women at increased risk for STDs, especially women younger than 25 years and with multiple sexual partners, should be annually screened for chlamydial infection and gonorrhea. Such screening will prevent salpingitis and subsequent tubal infertility and ectopic pregnancy more effectively than any other measure. Previous salpingitis also predisposes women to subsequent salpingitis, probably because mucosa damaged from prior infection is more susceptible to infection than normal tissue. Patients with previous uncomplicated gonorrhea have a high rate of subsequent salpingitis, in part because of increased rates of subsequent gonorrheal infection. Prior chlamydial infection predisposes patients to salpingitis if a second chlamydial infection occurs. Repeated chlamydial infections produce a hyperimmune response that increases the chance and severity of tissue damage. The presence of an IUD is an independent risk factor for salpingitis. IUD users have a two-fold to four-fold increased rate of both salpingitis and tubal infertility, compared to non–IUD users. The highest rate of salpingitis in IUD users occurs within a few weeks of insertion as a result of the introduction of cervical bacteria into the endometrial cavity along with the IUD. Most infections in IUD users, however, occur long after insertion, probably because bacteria wick along the IUD tail from the vagina to the uterus and adhere to the IUD surface. IUDs also appear to enhance anaerobic bacterial growth, and their use is associated with Actinomyces and bacterial vaginosis infection. Because IUD use is also associated with tubal infertility, an IUD should not be inserted in women who desire future pregnancy. By contrast, barrier or oral contraceptive methods appear to protect against salpingitis. The protective effect of oral contraceptives on salpingitis appears to exist only for patients with chlamydial infection, possibly due to a down-regulation of the hyperimmune response caused by the organisms.

The role of male contacts with untreated gonococcal or chlamydial urethritis is often ignored by gynecologists. Only 25% of male contacts of women with gonococcal salpingitis are treated by the time the female partner develops symptomatic salpingitis. Over 50% of the male contacts with gonococcal urethritis are asymptomatic, and N. gonorrhoeae is isolated from 40% of these asymptomatic males. Men with nongonococcal urethritis represent reservoirs of chlamydial salpingitis. To reduce the rate of new and recurrent salpingitis, all male contacts of women with any type of salpingitis should be examined and cultured. If infectious organisms are found, males should be appropriately treated. Bacteriology Neisseria gonorrhoeae In most studies in the United States, N. gonorrhoeae is recovered from 40% to 50% of women with acute salpingitis. However, gonococcal prevalence varies greatly: N. gonorrhoeae is isolated from less than 20% of salpingitis cases in Sweden and from 80% of cases in certain urban populations in the United States. In women with both cervical gonorrhea and salpingitis, N. gonorrhoeae is the most frequent intraabdominal isolate, but the sole isolate in only 30% of these cases. The remainder have either no organisms or other organisms isolated alone or together with N. gonorrhoeae in the abdomen. Chlamydial infection frequently coexists with gonorrhea; in some studies, more than 50% of women with gonorrhea also had C. trachomatis in the cervix. Positive tubal gonococcal cultures are usually obtained during the early stages of infection, but not in the later stages of infection. Chlamydia trachomatis C. trachomatis is as important as the gonococcus in causing acute salpingitis. From 30% to 60% of women with salpingitis have C. trachomatis, and in most of these women the organisms can be isolated from the fallopian tube. Application of new DNA tests for C. trachomatis identifies even more chlamydial infections. Chlamydial salpingitis is under-diagnosed because many patients have mild symptoms and are not identified. It is now evident that women with mild symptoms and signs have the same amount of severe tubal damage as those with severe symptoms. Chlamydial salpingitis often produces mild symptoms and signs but may produce more severe tubal damage compared to gonococcal salpingitis. Nonsexually Transmitted Aerobic and Anaerobic Bacteria Nonsexually transmitted aerobic and anaerobic bacteria are normally present in cervical and vaginal flora and particularly include organisms associated with bacterial vaginosis. These organisms can be a direct cause of salpingitis, but they also cause secondary infection in combination with N. gonorrhoeae and C. trachomatis, IUD use, or instrumentation. Polymicrobial infection with these agents is common in salpingitis. In such cases, many different Gram-positive and Gram-negative aerobic and anaerobic organisms are isolated, particularly Porphyromonas, Prevotella, and Bacteroides spp, including B. fragilis. Anaerobic organisms are especially common in serious infections, and they are virtually always found in abscesses. Mycoplasmas Genital mycoplasmas have been recovered from the tubes or cul-de-sac in 2% to 20% of patients with salpingitis. In addition, more than 20% of patients with salpingitis have changes of mycoplasmal antibody titer suggestive of invasive infection. These organisms lack the virulence of N. gonorrhoeae and C. trachomatis, and they appear to cause an adnexitis rather than salpingitis. M. genitalium identified by DNA technology is related to endometritis. Pathogenesis Salpingitis occurs from vaginal and cervical bacteria ascending into the endometrium and fallopian tubes. The ascent of bacteria probably increases during menses, as evidenced by the onset of pain within 7 days of starting menses in one-half to two-thirds of patients with gonococcal salpingitis. Virulent gonococci also proliferate at menstruation, and less virulent gonococci are present at other times of the cycle. Other risk factors for salpingitis exist. Virulent bacteria in the cervix are more likely to cause salpingitis than nonvirulent bacteria, and C. trachomatis and N. gonorrhoeae are two virulent organisms capable of causing salpingitis. Virulence is occasionally evidenced by bacteria in the normal flora. Specific bactericidal antibodies to N. gonorrhoeae appear to reduce salpingitis. Clinically recognized salpingitis develops in only 10% to 17% of women with cervical gonorrhea, and most of these women probably develop tubal infection during the first one or two menstrual periods between gonococcal acquisition and the development of specific bactericidal antibodies. C. trachomatis may not depend on menses for ascent into the endometrium. Endometritis caused by C. trachomatis appears to be a common chronic intermediate infection that exists through several menstrual cycles. Tubal damage from chlamydial infection appears to occur from the immune response to infection. As mentioned, repetitive chlamydial infections elicit a hyperimmune response that accelerates tissue damage. The presence of chlamydial heat shock protein also contributes to an acceleration of tissue damage. The usual route of infection with either organism is the contiguous spread along mucosa from the cervix to the endometrial cavity and fallopian tubes (see Fig. 32.6). Lymphatic or hematogenous dissemination of organisms from the uterus to the adnexa is uncommon in nonpregnant women. It also is possible that organisms may be transported along the fallopian tubes by cilia or even carried by their attachment to spermatozoa or to other organisms. Fitz-Hugh and Curtis Syndrome Perihepatitis consisting of liver capsule inflammation without liver parenchymal damage is referred to as Fitz-Hugh and Curtis syndrome ( Fig. 32.9). Swelling of the liver capsule produces pain with inspiration, usually in the right upper quadrant. A purulent or fibrinous exudate appears on the capsular surface. Violin-string adhesions between the liver capsule and the anterior abdominal wall are a late manifestation of capsular inflammation.

FIG. 32.9. Liver adhesions in Fitz-Hugh and Curtis syndrome. Perihepatitis was formerly believed to be caused solely by N. gonorrhoeae, but C. trachomatis also causes this syndrome. Chlamydial heat shock protein is associated with perihepatitis, indicating that the syndrome is another manifestation of a hyperimmune response to chlamydiae. Some organisms travel transperitoneally from the fallopian tubes to reach the liver surface, but organisms may also reach the liver by lymphatic and hematogenous routes. The syndrome occurs virtually exclusively in women, although two men with this syndrome have been reported. Salpingitis is invariably the source, but the syndrome occasionally follows appendicitis and other causes of peritonitis. The Fitz-Hugh and Curtis syndrome is frequently misinterpreted as cholecystitis, viral pneumonia, or pyelonephritis. Liver enzyme levels may be mildly elevated. In women with salpingitis, 5% to 10% developed symptoms, but another 5% of women have asymptomatic perihepatitis. This latter group may have the violin-string adhesions recognized as an incidental finding at a subsequent surgery. Many women with Fitz-Hugh and Curtis syndrome note the onset of lower abdominal pain before or with the upper abdominal pain, but some develop such severe upper abdominal pain that they fail to complain of lower abdominal pain. Given the frequency of salpingitis and the infrequency of acute cholecystitis in women 15 to 30 years of age, Fitz-Hugh and Curtis syndrome is a more likely cause of upper quadrant pleuritic pain than cholecystitis and should be suspected in any woman with pleuritic upper quadrant pain and physical signs of salpingitis. Laparoscopy is useful to diagnose unclear cases. Diagnosis The largest unsolved problem with salpingitis is the lack of sensitive and specific diagnostic criteria. For an estimated one-half of women, salpingitis does not cause sufficiently typical symptoms to be diagnosed. In Table 32.3 women are presented who, despite not having a recognized episode of salpingitis, had severe enough tissue damage to cause infertility from tubal obstruction. Thus, patients with mild abdominal pain and other mild manifestations are often not identified. An emphasis must be placed on increasing the sensitivity for the diagnosis. The problem is the very broad spectrum of clinical severity among patients with salpingitis. Although severe manifestations are usually recognized as salpingitis, they occur in only 30% of patients. Insistence on rigid criteria, such as fever, severe tenderness, leukocytosis, and an elevated erythrocyte sedimentation rate (ESR), leads to a failure of diagnosis in nonovert cases.

TABLE 32.3. Proportions of patients with tubal occlusion from salpingitis with no history of salpingitis

On the other hand, a clinical diagnosis of salpingitis that relies on the history, physical examination, and nonspecific laboratory tests also has a large false-positive error rate. Several studies demonstrate that a clinical diagnosis of salpingitis can be confirmed by laparoscopy in about 65% of patients ( Table 32.4); about 20% of patients had no disease observed, and another 15% had other pelvic conditions, most commonly ovarian cyst, ectopic pregnancy, appendicitis, or endometriosis.

TABLE 32.4. Laparoscopic observations in patients with a clinical diagnosis of pelvic inflammatory disease

History Important history and physical findings in patients with presumed PID are listed in Table 32.5. However, none of these findings distinguish women with salpingitis from those with other causes of pelvic pain. Lower abdominal pain is the most consistent symptom in women with overt salpingitis, although it may be mild or even absent. Acute pain is present for less than 15 days in 85% of patients who present with PID. Most women with gonococcal salpingitis experience acute pain during menses; in chlamydial salpingitis, the onset of pain is often insidious and not associated with menses. The abdominal pain is usually continuous and most severe in both lower quadrants. It increases by movement, the Valsalva maneuver, and intercourse. Abnormal vaginal bleeding occurs in 15% to 35% of women with salpingitis. Symptoms of appendicitis and ectopic pregnancy overlap with those of PID. The risk of STD can be helpful in forming a tentative opinion. An increased risk of PID would be expected for women with multiple sexual partners, other STDs, symptomatic male partners, and prior gonorrhea or prior PID.

TABLE 32.5. Clinical findings in 176 women with suspected acute pelvic inflammatory disease (PID)

Physical Examination Most patients with salpingitis have lower abdominal, cervical, and bilateral adnexal tenderness (see Table 32.5). Cervical motion tenderness is a sensitive indicator of salpingitis. However, none of these findings is specific for PID and patients with other disease or with no apparent disease may have similar physical findings. Other associated findings lack the sensitivity to be useful. For example, although a temperature of 100.4°F (38°C) or higher is present more often in patients with than in those without salpingitis, only 45% of patients with laparoscopically confirmed salpingitis have a temperature greater than 100.4°F. Laboratory Tests Nonspecific tests as the peripheral WBC count and the ESR are helpful only if the results are abnormal, but they are often normal. Of patients with laparoscopically confirmed salpingitis, 50% have a normal WBC count and 25% have a normal ESR. C-reactive protein levels may be more useful. Laboratory signs such as yellow cervical mucopus and a cervical Gram stain with an increase of polymorphonuclear leukocytes of more than 30 per high-powered field appear to offer a more specific indication of salpingitis in patients with pelvic tenderness. It is mandatory to obtain a culture for gonorrhea and a test for chlamydial infection. Cervical culture for other organisms is not recommended. Endometrial Biopsy Endometrial biopsies are easy and safe to obtain. Histologic evidence of endometritis is based on finding plasma cells and polymorphonuclear leukocytes migrating through the epithelium. Histologic endometritis has a 90% sensitivity and specificity for diagnosing salpingitis, compared to laparoscopy. Ultrasound and Computed Tomography Abnormal vaginal ultrasound findings correlate with a diagnosis of salpingitis made by laparoscopy, but these findings remain too insensitive with mild tubal abnormalities and too nonspecific for a certain diagnosis. Ultrasound is useful to distinguish an abscess from an inflammatory mass within the adnexa, define a mass in obese or excessively tender patients, and follow the size of a mass with treatment. Computed tomography has successfully been used for the same purposes; it may be especially helpful if ultrasound is difficult to perform, as in peritonitis or a recent abdominal incision. Laparoscopy Laparoscopy provides the most accurate way to diagnose salpingitis. It should be used when the diagnosis is unclear, particularly in patients with severe peritonitis, to exclude a ruptured abscess and appendicitis. It is estimated that for every 100 times a clinical diagnosis of PID is made without visual confirmation, three patients with appendicitis are treated for PID (see Table 32.4), resulting in a critical delay in the correct diagnosis. Pain and tenderness from acute PID should abate 3 or 4 days after antibiotics begin. Patients without reduced tenderness on antibiotic therapy also benefit from laparoscopy. About 20 of 100 women with a clinical diagnosis of PID have no abnormality at laparoscopy. All cases where laparoscopy is performed should have cultures taken from the fimbriated ends of the tubes, regardless of the findings, to detect the small number of patients with endosalpingitis and normal-appearing tubes. Open laparoscopy is used to identify and percutaneously drain pelvic abscesses. The abscess is visualized, and a 14-French catheter is placed into the abscess, which is drained of pus and carefully rinsed with sterile bacteriostatic water. A closed drainage system is then connected to the catheter for 1 to 3 days until drainage ceases. About 90% of abscesses are successfully treated by percutaneous drainage. Examination of the Male Partner Examination of the male sexual partner helps establish the diagnosis of PID and treat STDs. At least 80% of male contacts of women with PID are not treated by the time PID occurs in the female partner. If there is no urethral discharge, a Gram stain and urethral material for N. gonorrhoeae and C. trachomatis identification should be obtained. Treatment Adequate treatment of salpingitis includes an assessment of severity, antibiotic therapy, additional general health measures, close patient follow-up, and treatment of the male sexual partner. Patients with the mildest manifestations can be treated as outpatients. Specific indications for hospitalization include severe manifestations of salpingitis (i.e., severe peritonitis, severe nausea, or fever higher than 100.4°F [38°C]), a suspected abscess, outpatient antibiotic failure, and an uncertain diagnosis

with severe symptoms. Patients should be examined within 2 to 3 days and again at 7 and 21 days later to verify a satisfactory response to treatment. If an IUD is in place, it should be removed 24 to 48 hours after therapy is started. Ideally, the antibiotic should be selected according to the organism isolated, but, in salpingitis, empiric therapy is used. The treatment regimens recommended by the CDC were designed to treat gonococcal and chlamydial salpingitis and anaerobic salpingitis. Inpatient regimens include either intravenous cefoxitin or cefotetan disodium (Cefotan) and oral doxycycline or intravenous clindamycin and gentamicin, followed by oral clindamycin or doxycycline for a total of 14 days. The clindamycin regimen is also effective for patients with chlamydial infection. Parenteral therapy can be stopped 24 hours after clinical improvement. Outpatient regimens include a loading dose of an antibiotic recommended for gonorrhea and treatment of Chlamydia with an option to treat for anaerobic infection. One treatment regimen consists of ofloxacin 400 mg twice daily or levofloxacin 500 mg once daily for 14 days with or without metronidazole. A second regimen includes ceftriaxone or cefoxitin with probenecid in a single dose plus doxycycline with or without or metronidazole, 500 mg twice daily for 14 days. It is also possible to use other antibiotic regimens with similar antimicrobial activity. Gonococcal salpingitis responds more rapidly to antibiotics than nongonococcal salpingitis. Recommended agents must be used in full doses because partially treated, subacute salpingitis may result from the use of lower doses. Hospitalized patients with peritonitis but no adnexal abscess usually respond rapidly to the regimens. An adnexal abscess, even if systemic manifestations are mild, should be treated with antibiotics that inhibit B. fragilis because of its frequency in pelvic abscesses. Clindamycin, metronidazole, cefoxitin, or imipenem/cilastatin should be used to treat a known or suspected pelvic abscess. Abdominal surgery is indicated for a ruptured abscess. Colpotomy drainage is usually preferable for an unruptured midline abscess filling in the cul-de-sac. If laparotomy is performed for a presumptive diagnosis of appendicitis but instead acute salpingitis is found, the procedure should be limited to taking a tubal culture and closure of the abdomen. If laparotomy is required for an unresolved abscess or adnexal mass, surgery should be limited to the most conservative procedures that will be effective. Unilateral abscesses respond to unilateral salpingo-oophorectomy if appropriate antibiotic regimens are used; routine hysterectomy and bilateral salpingo-oophorectomy are seldom needed to treat acute salpingitis in young women. As mentioned, percutaneous drainage of abscesses is usually successful. When chronic pain occurs, surgery should be deferred as long as possible to allow maximum healing. Analgesics and oral contraceptives to prevent ovulation may suffice until the swelling and fixation decrease. Surgery may be indicated for persistent pain that does not respond to conservative measures, for recurrent attacks of pelvic pain, or for a pelvic mass that does not resolve. Laparoscopy with lysis of adhesions usually suffices, but adnexectomy or, rarely, even hysterectomy is required. Surgery is also necessary for infertile patients.

OOPHORITIS Most cases of oophoritis are secondary to salpingitis. The ovary becomes infected when purulent material from the fallopian tube enters the ovary. If the tubal fimbriae are adherent to the ovary, the tube and ovary together may form a large retort-shaped tuboovarian abscess. Antibacterial therapy, as previously outlined, is immediately indicated, and surgery is mandatory if the mass is considered to be leaking or ruptured or if it fails to resolve. Ovarian abscess after gynecologic surgery is particularly resistant to antimicrobial treatment. Oophoritis may occur without accompanying salpingitis in mumps and septicemia. Oophoritis of this type usually results in lower abdominal pain that lasts for only a few days during an acute infectious illness. The ovarian infection usually subsides without incident, although abscesses may occur. If bimanual examination is not satisfactory, ultrasound scans may be used to detect abscesses.

GENITAL TUBERCULOSIS Female genital tuberculosis remains relatively uncommon in the United States. Less than 1% of salpingitis cases can be attributed to Mycoplasma tuberculosis. Pulmonary tuberculosis remains a problem in many impoverished areas, and its rate is rising in parallel with the number of female patients with HIV infection. Although the spread of tuberculosis from the primary pulmonary infection to the pelvis usually occurs early in tubercular infection, early detection of genital infection is seldom feasible. Genital tuberculosis develops in about 10% of patients with pulmonary tuberculosis. Pathogenesis Virtually all genital infections are secondary to a pulmonary infection, which usually spreads by the bloodstream from the lungs to the fallopian tubes within 1 year of the primary pulmonary infection ( Fig. 32.10 and Fig. 32.11). Direct extension occurs from the tube in several directions: to the pelvic peritoneum and ovary, to the endometrium, and to the cervix. Less common lymphatic extension to the genitalia can occur from abdominal sources or by direct extension from the intestinal tract. Genital tuberculosis is seldom caused by an ascending infection from a sexual partner with tuberculous epididymitis.

FIG. 32.10. Mode of transmission of tuberculous pelvic infection. Tubercle bacillus invades pelvic organs by way of bloodstream from distant focus in the lung or other organ. (From Wharton LR. Gynecology and female urology. Philadelphia: WB Saunders, 1943, with permission.)

FIG. 32.11. Tuberculosis of fallopian tube (×105). (From Curtis AH, Huffman JW. A textbook of gynecology, sixth ed. Philadelphia: WB Saunders, 1950, with permission.)

The initial tubal lesion may remain localized for a considerable time (in some cases years), or it may extend to the interior tubal mucosa. Endosalpingitis results either in an exudative phase where ulcer formation at the site of caseous degeneration produces a typical moth-eaten pattern hysterosalpingogram, or in an adhesive phase, where large tubercles are present within the tubes. Dense perisalpingian adhesions are characteristic. In contrast to bacterial salpingitis, tubal occlusion, particularly fimbrial closure, does not occur early in tuberculous disease, and the tubes may remain patent despite relatively marked destruction of the tubal wall. Tubal infection is present in virtually all women with genital tuberculosis. Endometrial infection is present in 50% to 80% of cases. Infected endometrium is shed monthly at menses and the endometrium becomes reinfected from tubal seeding. Myometrial infection occurs but only in advanced cases. Cervical infection occurs in

only 10% to 25% of patients, resulting in either a papillomatous or ulcerative lesion that grossly resembles cervical carcinoma. Clinical Forms Latent Genital Tuberculosis In the latent form, genital tuberculosis appears partially or completely arrested after the initial tubal infection, and patients have few or no pelvic complaints. Pelvic physical findings are normal. The diagnosis is usually made in the course of an investigation of infertility (by endometrial biopsy or dilation and curettage) or by chance at laparotomy. In the latent phase, a precarious balance exists between the disease and the host defense mechanisms. Active but latent infection has been documented 30 years after an initial infection. Tuberculous Salpingitis Tuberculous salpingitis is a more advanced infection that may develop immediately after the primary hematogenous tubal spread, or it may follow a prolonged latent phase. The tubes are grossly enlarged by the inflammatory reaction. Although the symptoms and findings can mimic those of acute bacterial salpingitis, clinical manifestations are usually indolent and prolonged. Additionally, tuberculous salpingitis does not respond to the antibiotic therapy used for acute bacterial salpingitis. Despite these differences from bacterial salpingitis, tuberculosis is often discovered only by histologic examination of an excised tube. Tuberculous Peritonitis In tuberculous peritonitis, widespread infection of all peritoneal surfaces produces ascites, adhesions, and innumerable small nodules (i.e., tubercles) throughout the abdomen. The serosal surfaces of the pelvic organs are typically involved, and the tubes are often patent. This form results from hematogenous or lymphatic spread. Diagnosis A history of pulmonary tuberculosis is often present in patients with genital tuberculosis, but simultaneous active pulmonary infection is uncommon. A normal chest film does not exclude genital tuberculosis because pulmonary lesions are found in only 30% to 50% of genital tuberculosis cases. The most common complaints are sterility and pelvic pain. Deteriorating health and menstrual abnormalities may also occur. Menorrhagia may be associated with abdominal pain, but amenorrhea or oligomenorrhea also occurs with tuberculous peritonitis. Most women with genital tuberculosis are in their 20s and 30s, but because of the frequent tendency for long latency periods, the disease may become active even after menopause. As mentioned, tubal tuberculosis may mimic acute bacterial salpingitis. However, when a pelvic problem does not conform to expected rules, the first consideration should be ectopic pregnancy and the second pelvic tuberculosis. Pelvic tuberculosis should be strongly considered when salpingitis occurs in a woman who is considered to be virginal. A first-strength, purified protein derivative tuberculin skin test is important because a negative result virtually rules out the possibility of tuberculosis. Diagnosis is best established by endometrial biopsy, which should be performed in the week preceding menstruation, when the endometrium is thickest and is most likely to contain tubercles. A portion of the specimen should be cultured, and the remainder submitted for histologic examination. Repeated cultures of the menstrual blood can also be obtained. If cultures of either biopsy specimens or menstrual blood are negative, curettage may yield results. Antimicrobial susceptibility testing should be performed to predict drug resistance. Endometrial biopsy, curettage, and culture of menstrual flow or endometrial tissue can provide an exact diagnosis of genital tuberculosis if the results are positive; if the results are negative, tuberculosis cannot be excluded. Diagnostic laparotomy is indicated if these measures fail to verify a diagnosis in a patient with a history and pelvic findings suggestive of genital tuberculosis. Diagnostic laparoscopy may be performed if there is minimal likelihood of tuberculous peritonitis, but caution must be used because of the possibility of perforating a loop of adherent bowel. Hysterosalpingography may reveal a characteristic tubal pattern, but it may also cause severe exacerbation of disease and should not be used if tuberculosis is possible. The incidental discovery of reproductive tract tuberculosis may be the first indication that the patient has tuberculosis. In such patients, it must be determined whether or not other sites (e.g., lungs, urinary tract, bone, and gastrointestinal tract) also are infected. Treatment A three-drug therapy consists of isoniazid, 300 mg daily, rifampin, 600 mg daily, and pyrazinamide, 1,500 mg daily, for 18 to 24 months. In areas of high resistance, streptomycin or ethambutol (Myambutol) is added as a fourth drug. Patients without adnexal masses should have an endometrial biopsy for culture and microscopic examination 6 and 12 months after starting therapy. Persistent organisms need to be susceptibility-tested to identify drug-resistant strains. Laparotomy is performed if adnexal masses persist for 4 months and rifampin or streptomycin should be given preoperatively. Bilateral salpingectomy and removal of other tuberculous foci may be performed in young women with minimal disease. A bilateral salpingo-oophorectomy and total hysterectomy are indicated in those with advanced disease or advanced age. Pregnancy after tubal tuberculosis is rare, even in women with minimal disease. If the tubes are damaged by genital tuberculosis, efforts to improve fertility by tubal operations are usually futile.

SUMMARY POINTS STDs are common, particularly in young sexually active women with multiple sexual partners. However, most STDs occur among the larger population of women not at high risk for STDs. STDs are often asymptomatic for both women and their sexual partners. Many classic symptoms associated with various STDs are common in women without infection and, conversely, they are not present in many women with STDs. Many signs present in women with STDs also are common and they are present in women with no identifiable infection. Specific laboratory tests are usually required to confirm infection. The best course of treatment of STDs incorporates a combination of antimicrobial therapy directed at the patient and her partner, education about reducing STD exposure, and, when appropriate, careful follow-up examination and testing. SUGGESTED READINGS 1.

Al-Tawfiq JA, Spinnola SM. Haemophilus ducreyi: clinical disease and pathogenesis. Curr Opin Infect Dis 2002;15:43–47.

2.

Bauer HM, Ting Y, Greer CE, et al. Genital human papillomavirus infection in female university students as determined by a PCR-based method. JAMA 1991;265:472.

3.

Boardman LA, Peipert JF, Brody JM, et al. Endovaginal sonography for the diagnosis of upper genital tract infection. Obstet Gynecol 1997;90:54.

4.

Centers for Disease Control and Prevention. Sexually transmitted disease treatment guidelines 2002. MMWR 2002;51:1–77.

5.

Eschenbach DA, Wolner-Hanssen P, Hawes SE, et al. Acute pelvic inflammatory disease: associations of clinical and laboratory findings with laparoscopic findings. Obstet Gynecol 1997;89:184.

6.

Hillier SL, Krohn MA, Rabe LK, et al. The normal flora, H2O2-producing lactobacilli and bacterial vaginosis in pregnant women. Clin Infect Dis 1993;16[Suppl 4]:S273.

7.

Kimani J, McClean IW, Biyayo JJ, et al. Risk factors for Chlamydia trachomatis pelvic inflammatory disease among sex workers in Nairobi, Kenya. J Infect Dis 1996;173:1437.

8.

Korn AP, Landers DV, Green JR. Pelvic inflammatory disease in human immunodeficiency-virus infected women. Obstet Gynecol 1993;82:765.

9. Larsson P-G, Platz-Christensen J-J, Thejls H, et al. Incidence of pelvic inflammatory disease after first-trimester legal abortion in women with bacterial vaginosis after treatment with metronidazole: a double-blind, randomized study. Am J Obstet Gynecol 1992;166:100. 10.

Martin DH, Mroczkowski TF, Dalu ZA, et al. A controlled trial of a single dose of azithromycin for the treatment of chlamydial urethritis and cervicitis. N Engl J Med 1992;327:921.

11.

McCormick JK, Yarwood JM, Schlievert PM. Toxic shock syndrome and bacterial superantigens; an update. Am Rev Microbiol 2001;55:77–104.

12.

Morcos R, Frost N, Hnat M, et al. Laparoscopic versus clinical diagnosis of acute pelvic inflammatory disease. J Reprod Med 1993;38:53.

13.

Quinn TC, Gaydos C, Shepherd M, et al. Epidemiologic and microbiologic correlates of Chlamydia trachomatis infection in sexual partnerships. JAMA 1996;276:1737.

14. Schacter J, Moncada J, Whidden R, et al. Noninvasive test for diagnosis of Chlamydia trachomatis infection: application of ligase chain reaction to first-catch urine specimens of women. J Infect Dis 1995;172:1411. 15.

Schaefer G. Female genital tuberculosis. Clin Obstet Gynecol 1976;19:223.

16.

Scholes D, Stergachis A, Hendrick P, et al. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection. N Engl J Med 1996;334:1399.

17. Shands KN, Schmid GP, Dan BB, et al. Toxic-shock syndrome in menstruating women: association with tampon use and Staphylococcus aureus and clinical features in 52 cases. N Engl J Med 1980;303:1436. 18.

Sobel JD, Faro S, Force RW, et al. Vulvovaginal candidiasis: epidemiologic diagnostic and therapeutic considerations. Am J Obstet Gynecol 1998;125:203–211.

19.

Soper DE, Bump RC, Hurt WG. Bacterial vaginosis and Trichomonas vaginalis are risk factors for cuff cellulitis after abdominal hysterectomy. Am J Obstet Gynecol 1991;163:1016.

20.

Taylor-Robinson D. Infections due to species of Mycoplasma and Ureaplasma: an update. Clin Infect Dis 1996;23:671.

21.

Tortlero-Luna G. Epidemiology of genital human papillomavirus. Hematol Oncol Clin N Am 1999;13:245–257.

22.

Walker CK, Kohn JG, Washington AE, et al. Pelvic inflammatory disease: meta-analysis of antimicrobial regimen efficacy. J Infect Dis 1993;168:969.

23.

Watts DH, Krohn M, Hillier SL, et al. Bacterial vaginosis as a risk factor for postcesarean endometritis. Obstet Gynecol 1990;75:52.

24. Westrøm L, Joesoef R, Reynolds G, et al. Pelvic inflammatory disease and infertility: a cohort study of 1844 women with laparoscopically verified disease and 657 control women with normal laparoscopic findings. Sex Transm Dis 1992;19:158. 25.

Whitley RJ, Roizman B. Herpes simplex virus infections. Lancet 2001;357:1513–1518.

Chapter 33 Benign Vulvovaginal Disorders Danforth’s Obstetrics and Gynecology

Chapter 33 Edward J. Wilkinson

Benign Vulvovaginal Disorders

BENIGN DISEASES OF THE VAGINA Vaginitis Lichenoid Vaginitis Congenital Vaginal Adenosis Acquired Vaginal Adenosis Vaginal Granulation Tissue Vaginal Prolapse of the Fallopian Tube VULVAR NONNEOPLASTIC EPITHELIAL DISORDERS Lichen Sclerosus Squamous Cell Hyperplasia Squamous Cell Hyperplasia and Granular Cell Tumor Lichen Planus Cicatricial Pemphigoid Psoriasis Plasma Cell Vulvitis (Vulvitis of Zoon) Vulvar Vestibulitis Pruritic Vulvar Squamous Papillomatosis Contact Dermatitis Atopic Dermatitis Fixed Drug Eruption (Dermatitis Medicamentosa) Behçet Syndrome Hidradenitis Suppurativa Miscellaneous Inflammatory Conditions ULCERS OF THE VULVA AND VAGINA Dark Lesions of the Vulva Compound Nevi Other Benign Pigmented Lesions Management of Pigmented Lesions Depigmentation Disorders BENIGN TUMORS OF THE VULVA Condyloma Acuminatum Leiomyoma of the Vulva Benign Vulvar Tumors of Sweat Gland Origin Papillary Hidradenoma Mammary-like Tissue: Arising from the Specialized Anogenital Glands Clear Cell Hidradenoma Syringoma Vascular Lesions Angiokeratoma Pyogenic Granuloma (Granuloma Pyogenicum) Bacillary Angiomatosis Kaposi Sarcoma VULVAR CYSTS Bartholin Cyst and Bartholin Abscess Bartholin Abscess Keratinous Cyst (Epithelial Inclusion Cyst) Mucous Cyst Columnar Cell Metaplasia Vulvar Endometriosis Apocrine Hydrocystoma/Cystadenoma of the Vulva Mesonephric-like Cyst (Wolffian-like Duct Cyst) Cyst of the Canal of Nuck (Mesothelial Cyst) SUGGESTED READINGS Benign Vaginal Conditions and Vaginitis Vulvar Non-Neoplastic Epithelial Disorders Including Vulvar Lichen Sclerosus and Squamous Hyperplasia (Hyperplastic Dystrophy) Ulcers and Infectious Diseases Vulvar Vestibulitis, Vulvar Squamous Papillomatosis, Nevi, and Cysts Additional Books and Book Chapters

This chapter addresses benign diseases of the vagina and vulva. Vaginal disorders including vaginal inflammation, vaginal adenosis, vaginal granulation tissue, and vaginal prolapse of the fallopian tube are discussed. Benign diseases of the vulva are presented in four major categories: nonneoplastic epithelial disorders, including vulvar lichen sclerosus and squamous cell hyperplasia; other inflammatory dermatoses, including vulvar vestibulitis; vulvar and vaginal ulcers, including infectious conditions that may cause ulceration; and a section on dark lesions of the vulva, including melanosis, lentigo simplex, nevi, and other benign pigmented lesions. The last section discusses the differential diagnosis of cysts of the vulva, including Bartholin cysts, and reviews common benign tumors of the vulva.

BENIGN DISEASES OF THE VAGINA Vaginitis Vaginitis may be of various causes and these have been classified in various ways; however, the classification outline of Sobel has been adapted for use here ( Table 33.1).

TABLE 33.1. Causes of vaginitis in adults

Lichenoid Vaginitis Lichenoid vaginitis is primarily a descriptive diagnosis. Clinically these patients are usually of reproductive age and have symptoms of dyspareunia or vulvovaginal discomfort and pain. On examination vaginal mucosal atrophic-like changes are usually seen associated with punctate red areas. Histopathologic findings include superficial chronic inflammation at the mucosal–submucosal interface (lichenoid in character), atrophic-like changes within the mucosa, and microscopic submucosal hemorrhagic areas. A number of these patients have had associated vulvar vestibulitis or a vulvar dermatosis. Treatment is directed toward the underlying cause of the inflammatory process, if known. If atrophy is present, vaginal estrogen cream may be effective. Congenital Vaginal Adenosis An association between a specific lesion of the lower genital tract and intrauterine exposure to diethylstilbestrol (DES) and related nonsteroidal synthetic estrogens has been established. Clear cell adenocarcinoma (carcinoma) of the vagina or cervix has been the most ominous but least frequently encountered sequel related to maternal administration of these drugs, with an estimated incidence of no more than 4 per 1,000 exposed young women. Benign anatomic anomalies, however, are exceedingly common and include vaginal adenosis and cervical vaginal ridges, including a cervical “hood” of vaginal tissue about the apparent ectocervix. Transverse vaginal septum and large cervical ectopian (eversion) involving the cervix and the upper vagina are also recognized abnormalities. The uterus may be small and the uterine cavity “T”-shaped. Vaginal adenosis is defined as the presence of benign glandular epithelium in the superficial vaginal wall usually of endocervical and, more rarely, endometrial type. It is frequently accompanied by squamous metaplasia that may be extensive and can, over time, completely replace the columnar epithelium, leading to a “remodeling” of the upper vagina and cervix which results in a relatively normal-appearing vagina and cervix. The endometrial type of adenosis is more commonly associated with clear cell adenocarcinoma. Acquired Vaginal Adenosis Acquired adenosis of the vagina (columnar cell metaplasia, mucinous metaplasia, and ciliated cell metaplasia) represents an acquired metaplastic change of the nonkeratinized squamous epithelium of the vulva vestibule to columnar epithelium of mucinous or tubal–epithelial type. This has also been described within the vulvar vestibule (see “ Vulvar Cysts”). Vaginal columnar cell metaplasia (acquired adenosis) presents as a somewhat red area in the upper vagina that may be associated with a mucoid vaginal discharge. This has been described associated with Stevens-Johnson syndrome as well as after vaginal laser or 5-fluorouracil (5-FU) therapy for vaginal intraepithelial neoplasia. Within the vagina, topical estrogen cream and acidification of the vagina can promote squamous metaplasia of the columnar epithelium, with resulting squamous mucosa. Surgical excision is not necessary in such cases. Vaginal Granulation Tissue Granulation tissue within the vagina may be found in the vaginal suture line following hysterectomy or other vaginal operative procedure. It can also occur following vaginal treatment that ulcerated or resulted in the loss of the vaginal mucosa, such as laser therapy. The patient may complain of postcoital bleeding or of an unusual mucoid-appearing discharge. On physical examination granulation tissue is relatively red to violet in color as compared to the normal adjacent squamous mucosa. The surface may be slightly elevated and have a granular or mucoid appearance as compared to the adjacent squamous epithelium. The tissue may be friable and bleed easily on contact or following biopsy. Histopathology demonstrates characteristic granulation tissue with small clustered capillary-like vessels and loose-appearing edematous stroma with acute and chronic inflammatory cells. Treatment usually consists of excisional biopsy or electrofulgeration. Monsel solution can control bleeding and be used if the granulation tissue recurs. Acidification of the vagina can promote vaginal mucosal healing. Vaginal Prolapse of the Fallopian Tube Prolapse of the fallopian tube through an apical vaginal suture line following hysterectomy may occur. The patient may experience pain with coitus and postcoital bleeding. On examination a red elevated mass is usually identified within the suture line, and the mass may be tender or painful on palpation or with pressure. In some cases the entire fimbriae may protrude from the suture line. If biopsy is performed the findings are of tubal epithelium usually with associated tubal muscle or typical fallopian tube plical architecture. Treatment is vaginal excision of the prolapsed portion of the tube.

VULVAR NONNEOPLASTIC EPITHELIAL DISORDERS As classified by the International Society for the Study of Vulvovaginal Disease (ISSVD), vulvar nonneoplastic epithelial disorders include lichen sclerosus, squamous cell hyperplasia (formerly hyperplastic dystrophy), and other dermatoses. This terminology replaced the ISSVD terminology, as reported in 1976, and does not use the term “vulvar dystrophy”. Lichen sclerosus and squamous cell hyperplasia usually present as white lesions of the vulva. They are both symptomatic, associated with pruritus or burning pain. Biopsies, usually taken from several sites bearing the white epithelium, are necessary to definitively distinguish these lesions. Diagrammatic representation of vulvar anatomy is included in Fig. 33.1.

FIG. 33.1. Diagrammatic representation of vulvar anatomy. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with

permission.)

Lichen Sclerosus Lichen sclerosus often presents as white, thin, parchment-like skin. Ecchymosis and superficial erosions or ulcerations may be seen, secondary to scratching ( Fig. 33.2).

FIG. 33.2. Lichen sclerosus of the vulva with excoriation. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Long-standing lichen sclerosus may result in marked introital stenosis with atrophy of the labia and agglutination of the labia minora and frenulum of the clitoris. Perianal and perineal involvement is common and the involved areas are typically symmetric in location. Lichen sclerosus may occur at any age, including childhood. The involved skin is often thin and wrinkled, and areas of ecchymosis and ulceration may be seen. In children, perianal lichen sclerosus may present with pain or bleeding on defication, and may be misinterpreted as sexual abuse. On histopathologic examination, lichen sclerosus is characterized by loss of rete ridges with marked thinning of the epithelium. Within the dermis loss of cellularity with subepithelial dermal fibrin deposition, edema, and chronic inflammation in the deeper dermis are characteristic features. The cause of lichen sclerosus is unknown; however there is a recognized familial occurrence. It is recognized that women with vulvar lichen sclerosus have some risk of squamous cell carcinoma within the vulvar areas involved. In prospective evaluations the frequency of squamous cell carcinoma arising within vulvar lichen sclerosus has been observed to be 3.5% or higher. The precursor epithelial changes are not well recognized, however epithelial hyperplasia is commonly associated. One known neoplastic change in the epithelium is referred to as “vulvar intraepithelial neoplasia; differentiated type.” This type of vulvar intraepithelial neoplasia (VIN) lesion is associated with keratinizing squamous cell carcinoma, but is not associated with human papillomavirus (HPV) as are the other VIN types and their associated carcinomas. Any suspicious lesion, such as an ulcer or markedly hyperkeratotic area, identified within an area of lichen sclerosus is best biopsied to exclude carcinoma. Therapy for lichen sclerosus of the vulva includes topical high potency corticosteroids (e.g., halobetasol 0.05% or clobetasol propionate 0.05%). Application is usually once per day and continued until a response occurs. Tapering of the topical medication to one or two times per week is considered for long-term therapy, but does require medical observation. Therapy is continued indefinitely, recognizing that lichen sclerosus is not cured, but is controlled by this therapy. Topical testosterone is generally only used with topical high-potency corticosteroids in women with severe atrophic changes who do not respond to the topical corticosteroid alone. Topical testosterone is not recommended in children or those who may have unacceptable reactions to testosterone because it is absorbed systemically. Topical progesterone can be considered as a supplement to topical corticosteroid in children or in individuals where the side effects of testosterone exposure are not acceptable. In addition to the topical corticosteroid, sitz baths and mild tranquilizers or antihistamines may be needed initially to relieve acute symptoms. Squamous Cell Hyperplasia Squamous cell hyperplasia (formerly hyperplastic dystrophy) typically presents as white thickened skin, often with accentuated skin markings ( Fig. 33.3). The skin is thickened with prominent wide and deep rete ridges (acanthosis), a nonspecific term used as a diagnosis only when HPV infection, fungal infection, and other known causative conditions have been excluded.

FIG. 33.3. Squamous cell hyperplasia of the vulva. Pruritic vulvar hypertrophied plaque of squamous cell hyperplasia with associated postinflammatory hyperpigmentation. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) See color figure 33.3.

Squamous cell hyperplasia (can include lichen simplex chronicus and neurodermatitis and previously referred to as hyperplastic dystrophy) is also characterized by white epithelium. The epithelium usually appears thickened with an irregular surface. Unlike lichen sclerosus, the process is typically not symmetric, and may be localized to a single site. Shrinkage and agglutination of the labia is not seen, nor are the focal areas of submucosal ecchymosis that may occur in lichen sclerosus secondary to scratching. Squamous Cell Hyperplasia and Granular Cell Tumor Squamous cell hyperplasia can be associated with granular cell tumor of the vulva. In some cases the hyperplasia is so marked as to resemble well-differentiated squamous cell carcinoma (pseudoepitheliomatous hyperplasia). Although this is a rare event, biopsy of the hyperplastic area, including some underlying dermis containing the tumor, is usually diagnostic. Granular cell tumors are usually benign tumors, although rare cases with recurrence and malignant behavior have been reported. The tumor is thought to be of nerve sheath origin. Treatment is local excision. The pathologic changes seen in squamous cell hyperplasia include prominent acanthosis, with elongation, widening and deepening of the rete ridges, and thickening of the epidermis. Hyperkeratosis may be present. A mild chronic inflammatory infiltrate may be present within the dermis. This diagnosis is one of exclusion and if a specific histopathologic diagnosis such as lichen simplex chronicus or lichen planus can be made, a specific diagnosis should be given. Chronic candidiasis or dermatophyte infection may result in similar epithelial changes. The identification of the causative organisms by clinical evaluation and culture of skin scrapings, or

silver fungus stains on histologic sections, permits a precise diagnosis and specific therapy. Squamous cell hyperplasia is responsive to topical corticosteroids (e.g., betamethasone valerate 0.1% ointment). Lesions usually respond within 2 weeks, and treatment can be tapered and discontinued after the lesion regresses. Long-term therapy is not necessary and may lead to problems related to chronic topical corticosteroid use such as epithelial atrophy. The avoidance or elimination of exogenous contact irritant factors is also beneficial. Lichen Planus Lichen planus (LP) presents most commonly in women over 40 years of age. Symptoms include pruritus and burning but some patients may be asymptomatic. The oral and vaginal mucosa also may also be involved by white, lacelike plaques (Wickham striae). Women with the vulvovaginal-gingival syndrome have an erosive form of LP and present with involvement of the oral, vaginal, and vulvar mucosa. Women with LP may present with vulvar pain, dyspareunia, burning, as well as postcoital bleeding. The bleeding is apparently related to vulvar vestibular or vaginal adhesions. Long-term changes include anal and vaginal stenosis. Clinical findings may involve both the vulvar vestibule as well as the vagina ( Fig. 33.4). The findings may initially be minimal with foci of reticulated papules. In later or more severe stages, LP may present as an erosive or desquamative process, with loss and agglutination of the labia minora, frenulum and prepuce, and shrinkage with stenosis of the vagina. Postinflammatory hypopigmentation of the involved vulvar skin may occur, resembling advanced lichen sclerosus. Vaginal and oral mucosal involvement help distinguish LP from lichen sclerosus. Histopathologic findings are related to the severity, age, and location of the process. Vaginal and vulvar vestibular mucous membrane findings may differ somewhat from cutaneous changes.

FIG. 33.4. Lichen planus of the vulva. The lichen planus involves the vagina and vestibule and was associated with severe dyspareunia for 8 months. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Microscopic features identified in the involved skin or mucosa include a bandlike, predominately lymphocytic, inflammatory infiltrate. Plasma cells are usually rare or absent in the skin, but may be found in mucosal lesions. Lichenoid inflammation involves the superficial dermis and overlying epidermis. Basal epithelial cells have liquefaction necrosis. Colloid bodies (Civatte bodies) are degenerated keratinocytes and valuable in the diagnosis of LP. Acanthosis, with wedge-shaped hypergranulosis and hyperkeratosis without parakeratosis, are typical findings. Immunofluorescent studies of a skin or mucosal biopsy submitted in Michelle fixative, or other appropriate fixative, or snap frozen, can be studied and characteristically will demonstrate fibrinogen deposits in the dermal–epidermal junction. Complement C3 and granular immunoglobulin M (IgM) and immunoglobulin G (IgG) deposits may be found. In the mucosa of the vestibule vagina, LP changes may include thinned epithelium, erosion, ulceration, and bullae. The most common differential diagnoses include lichen sclerosus, plasma cell vulvitis, cicatricial pemphigoid, localized scleroderma (morphea), and contact dermatitis. Clinical history, examination of the patient's skin and oral mucosa, and biopsies of specific lesions, including unruptured bullae, for light and immunofluorescent studies are needed to definitively distinguish these processes. The treatment of LP is best undertaken with a dermatologist as consultant. The most difficult issue is the adhesion formation that leads to scarring and stenosis of the vestibule and vagina. Topical corticosteroids are the usual initial approach, and may include high-potency, fluorinated corticosteroids to gain initial response and relief. Other treatment may include topical or oral cyclosporine, oral dapsone, and oral griseofulvin. Cicatricial Pemphigoid Cicatricial pemphigoid can involve the vulva and the vagina and usually presents clinically with subepithelial blisters, which clinically may resemble the blisters of pemphigus vulgaris. Eosinophils are prominent in the vesicles of cicatricial pemphigoid. Immunofluorescence studies are needed to distinguish this bullous lesion from others, including pemphigus and linear immunoglobulin A (IgA) disease. Cicatricial pemphigoid has linear IgG and complement C3 deposits, which are usually absent in LP. Clinical management is complex and consultation with a dermatologist is recommended. Psoriasis Psoriasis affects about 2% of the U.S. population. It is inherited as an autosomal dominant trait with incomplete penetrance. When it involves the vulva, it often involves the lateral labia majora and genital intertriginous areas. Sharp-bordered, silvery-crusted erythematous papules and plaques characterize psoriasis ( Fig. 33.5). Removing the loose silvery scale reveals multiple small punctate bleeding points. This finding is referred to as Auspitz sign. Presentation of psoriasis on the vulva may not demonstrate the typical silvery scales but rather red erythematous plaques, with the red plaques being the primary clinical finding. The lesions of psoriasis usually are somewhat symmetric. Without treatment the psoriasis typically has a long protracted course. The presentation of psoriatic lesions developing related to localized trauma is referred to as the Koebner phenomenon. The lesions present within 7 to 30 days following the localized trauma. Reiter syndrome may occur related to psoriasis although it not common in women.

FIG. 33.5. Psoriasis. There is marked erythema and the lesion has sharply demarcated borders. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Histopathologic features of psoriasis include hyperkeratosis and parakeratosis, with uniform acanthosis evidenced by a relatively uniform increase of length of the rete ridges. The rete, at their base, are wide and “club-shaped.” There is a decreased granular layer, and intraepithelial abscesses, composed predominately of

granulocytes, are within the epidermis and referred to as Munro abscesses. There is increased epithelial turnover reflected in increased numbers of mitotic figures in the basal and parabasal areas. The dermal papillae have prominent vessels and also may be club-shaped. The associated papillary dermis is edematous with little or no significant inflammatory cell infiltrate. The differential diagnosis includes squamous cell hyperplasia, seborrheic dermatitis, and chronic eczema. Although these all have acanthosis, unlike psoriasis, the rete ridges are not relatively uniform in length and Munro abscesses are not present. Seborrheic dermatitis usually has granulocytes about follicular ostia that can clinically resemble psoriasis, therefore biopsy is usually necessary for definitive diagnosis ( Fig. 33.6). The clinical presentation and distribution of the cutaneous lesions can be similar when comparing psoriasis and seborrheic dermatitis and their distinction can be difficult.

FIG. 33.6. Seborrhea. The involved vulvar skin is somewhat reddened. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) See color figure 33.6.

Plasma Cell Vulvitis (Vulvitis of Zoon) Women with plasma cell vulvitis typically have symptoms of pruritus and burning. It should be distinguished from other red lesions including those of known etiology ( Table 33.2). An erythematous macular lesion that is red to orange in color and may be hemorrhagic characterizes plasma cell vulvitis. The macular areas may be single or multiple, and may involve relatively large areas of the vulva.

TABLE 33.2. Inflammatory red lesions of the vulva with known etiology

Histopathologic features include a thinned epithelium with reduced rete ridge length and without a granular layer or keratinized surface. The horizontally orientated parabasal squamous cells have been referred to as “lozenge-shaped keratinocytes.” Superficial dermal inflammation is present involving the dermal–epithelial interface, and is lichenoid in character. The inflammatory cells present consist predominantly of plasma cells. Prominent blood vessels are found in the dermis, often associated with intradermal hemorrhagic areas and hemosiderin-laden macrophages. The differential diagnosis includes lichen planus and related dermatosis, as well as some infectious conditions, especially syphilis. The prominent plasma cell infiltrate may include perivascular plasma cells, however the plasma cells are not associated with spirochetes or other organisms on silver stain and serologic studies for syphilis are negative. The cause of plasma cell vulvitis is not known. Treatment is primarily topical corticosteroids but may include intralesional corticosteroids, as well as etretinate. Supportive care, to relieve acute symptoms, is also recommended. Vulvar Vestibulitis Vulvar vestibulitis is one of a number of conditions associated with vulvar vestibular pain. There remains controversy regarding the definition, diagnosis, and treatment of vulvar vestibulitis. For the purposes of defining vulvar vestibulitis it is defined in this chapter as originally described ( Friedrich 1987), specifically as a symptom complex characterized by severe external dyspareunia (dyspareunia on entry), inflammation of the vulvar vestibule, and exquisite tenderness of the vulvar vestibule in the areas of inflamed minor vestibular glands. It is recognized that asymptomatic women may have some minor redness or pressure tenderness within the vulvar vestibule and this should not be interpreted as vulvar vestibulitis. Histopathologic findings associated with vulvar vestibulitis are not specific. Within biopsies, or specimens obtained from partial vestibular excisions performed to relieve vestibulitis, a superficial inflammatory cell infiltrate is present consisting predominately of lymphocytes, but also usually also includes plasma cells, mast cells, and complement C3. The inflammation is within the superficial submucosa. If minor vestibular glands are present, the inflammation is also seen around the minor vestibular glands. The minor vestibular glands are simple glands lined by tall, mucus-secreting columnar epithelium. The inflammation usually does not involve the gland lumens. No specific etiology has been identified as a cause of vestibulitis, and studies for HPV have not demonstrated a causal association. Although no uniformly effective medical therapy has yet been defined, clinical management, including biofeedback, is sometimes effective in relieving symptoms. A list of some of the more common treatments follows. None of these treatments, however, have been proven to be uniformly effective in controlled trials. Vaginal and introital lubricants with or without the use of vaginal dilators. Vaginal and introital lubricants with or without the use of biofeedback. Interferon local injections (3 million U/mL), 1 million units per injection site with up to 12 injection sites distributed about the vestibule. Injections are with a fine 25- or 33-gauge needle and an insulin syringe. Three courses of injections are given per week for 4 weeks. Topical anesthetics (5% Xylocaine ointment or 2% lidocaine gel) applied to the tender sites p.r.n. Systemic antifungal therapy is advocated as initial therapy by some (e.g., Diflucan, or similar medications). Amitriptyline (Elavil) started at a low dose 10 to 25 mg orally, with increased doses up to 3 times per day in protracted or unresponsive cases. Local excision of the involved inflamed vestibular areas has been used; however, long-term follow-up studies show recurrence or persistence of symptoms in about one half of the cases. Pruritic Vulvar Squamous Papillomatosis Pruritic vulvar squamous papillomatosis is a clinical complex that includes vestibular papillae and pruritus. Vulvar pain or burning may also be present and dyspareunia is common. On examination of the vulva, multiple small clustered squamous papillae are present on the medial aspects of the labia majus ( Fig. 33.7). In some cases the papillae present following the onset of symptoms. The papillae are sometimes so small that magnification is necessary for identification (colposcopic magnification ×10).

FIG. 33.7. Vulvar vestibular papillomatosis. Clustered vulvar vestibular papillary projections of squamous papillomatosis. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Histopathologic examination demonstrates small (1–2 mm) squamous epithelial covered papillae supported on delicate vascular stalks. The papillae usually are multiple and occur in clusters on the vestibule. Although initially thought to be secondary to HPV infection, more recent work does not support this association. Contact Dermatitis Contact dermatitis may be an allergic or an irritant-related entity. Allergic dermatitis is a cell-mediated response related to sensitizing agents. Topical medications and rubber are example agents. Irritant agents can include urine, stool, soaps, deodorants, perfumes, or other agents related to such reactions within the skin. Irritant dermatitis is more common than allergic dermatitis in most settings. The cutaneous or mucosal areas involved are usually confined to the area of exposure and the dermatitis often persists for weeks to months following exposure. Histopathologic findings may be quite variable, and are related to whether it is an allergic or irritant-related process, the severity of the reaction, duration of the exposure, and length of time the inflammatory process has been present. Epithelial spongiosis with microvesiculation is usually present, and if the lesion is more severe or long-standing, erosion and ulceration may be seen. A superficial perivascular inflammatory infiltrate consisting predominantly of lymphocytes and histiocytes is often present. Eosinophils may be present, especially in allergic reactions. In long-standing contact dermatitis, epithelial thickening (acanthosis) with parakeratosis and hyperkeratosis can occur resulting in histopathologic features of squamous cell hyperplasia with or without features of lichen simplex chronicus. Atopic Dermatitis Atopic dermatitis of the vulva is usually associated with pruritus and burning. On examination the vulvar skin may appear dry and scaling. Scratching related to the pruritus can result in excoriation and thickening of the vulvar skin. Vulvar biopsy is generally not needed unless significant epithelial thickening or a questionable lesion is identified. The histopathologic findings of atopic dermatitis may include epithelial thickening with features of squamous cell hyperplasia. The findings may be nonspecific with epithelial spongiosis and associated superficial dermal inflammation. The inflammatory infiltrate predominately includes lymphocytes and macrophages, however, eosinophils and mast cells also may be present. The extent of the inflammatory cell infiltrate tends to correlate with the severity and duration of the dermatitis. Immunofluorescent studies are generally not necessary but if performed may identify immunoglobulin E (IgE) on epidermal Langerhans cells. Fixed Drug Eruption (Dermatitis Medicamentosa) Fixed drug eruptions typically are related hypersensitive reactions to systemic drugs. The eruptions typically can recur at the same skin sites upon reexposure to the same drug. Multiple reexposure can be associated with lesions in new sites as well in the original locations. After the drug is stopped, the lesions usually heal completely in the mucosa, but in the skin, postinflammatory hyperpigmentation is common. Cutaneous fixed drug eruptions are a complex group of drug hypersensitivity reactions. They are subclassified into four distinct types that can be summarized as follows. Type I usually presents as urticaria and is mediated through IgE antibodies. Type II presents as purpura or bullous lesions and is mediated through IgG and IgM antibodies. Type III may present as an urticarial or maculopapular reaction and may be associated with vasculitis and is related to immune complexes. Type IV may present as a contact dermatitis, exanthem, or a maculopapular eruption. The histopathological findings of fixed drug reaction may include epithelial interface vacuolar changes with associated necrotic keratinocytes. Within the dermis eosinophils may be identified. These findings may be otherwise nonspecific or only suggest drug eruption. Other testing, including patch testing for the suspect offending drugs, indirect and direct Coombs testing, as well as other laboratory testing, all maybe useful. In cases of suspect fixed drug eruption, a complete drug exposure history is essential. Behçet Syndrome Behçet syndrome is of unknown etiology and includes oral aphthous ulcers, genital ulcers, and ophthalmologic inflammation. In addition these patients may have colitis, acne, cutaneous nodules, thrombophlebitis, and, in severe cases, encephalopathy. Some cases lack the ocular changes ( Table 33.3, on vaginal and vulvar ulcers). When Behçet syndrome involves the vulva, deep ulcerations are usually seen that, in severe cases, result in fenestration of the labia, and occasionally labial gangrene ( Fig. 33.8). The disease is of unknown cause and may wax and wane, with vulvar ulcerations healing and then recurring. The vulvar ulcers are usually associated with oral ulcers

TABLE 33.3. Causes of vulvar and vaginal ulcers

FIG. 33.8. Behçet disease. Painful vulvar ulcer in a woman who also had oral ulcers of a similar appearance. Serologic and microbiologic studies for other potential pathogens were negative. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

On histologic examination the involved vulvar skin or mucosa is involved with a necrotizing arteritis. This is associated with a chronic inflammatory infiltrate that is typically perivascular and may involve the vessel wall with damage to the arterial media. The arteritis is associated with endothelial cell swelling and can result in arteriolar occlusion or venous thrombosis. Hidradenitis Suppurativa Hidradenitis suppurativa is an apocrine gland–related chronic inflammatory disorder. The initial presentation is of a painful, tender, deep-seated subcutaneous nodule or nodules in the vulva as well as axillae and other sites where apocrine glands occur. Subcutaneous spread of the inflammatory process results in confluent masses that, in some point in their course, involve the overlying skin and result in ulceration and draining sinuses. This process, over time, subsequently causes extensive scarring of the vulva or other involved sites. Fox-Fordyce disease may be associated in some cases. Histopathologic features in the early stages of hidradenitis suppurativa include perifolliculitis associated with acute and chronic dermal inflammation. In advanced cases severe inflammation results in the loss of the epithelial appendages with squamous epithelial ulceration with sinus tract formation. Management includes medical as well as surgical approaches. Medical treatment may include long-term retinoids (e.g., isotretinoin), anti-androgen therapy, or treatment with leuprolide. Surgical approaches may include laser ablation by unroofing of the sinus tracts, or excision of involved areas. Surgical approaches are usually appropriate in severe or refractory cases. Miscellaneous Inflammatory Conditions Inflammatory infectious conditions that may occur in the vulva, that present as red lesions, with their known causative agent, are listed in Table 33.2.

ULCERS OF THE VULVA AND VAGINA Table 33.3 lists the predominant causes of ulcers within the vulva and vagina. Dark Lesions of the Vulva Dark or pigmented lesions of the vulva are listed in Table 33.4.

TABLE 33.4. Dark lesions of the vulva Lentigo Simplex Lentigo simplex is one of the most common pigmented lesions biopsied on the vulva. Lentigo simplex clinically is a flat, non-elevated pigmented area on the skin that has normal surface skin markings ( Fig. 33.9). Lentigo simplex represents an area of skin with increased melanin with associated melanocytes. The epithelium may be slightly hyperplastic and hyperkeratotic. Lentigo simplex by definition is 5 mm or less in diameter. Larger flat pigmented lesions with essentially identical histopathologic features are referred to as melanosis vulvae. No treatment is needed although biopsy is performed to determine the basis of the pigmented lesion.

FIG. 33.9. Lentigo simplex. Several small flat pigmented lesions of lentigo simplex. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) See color figure 33.9. Melanosis Vulvae and Melanosis Vaginae Vulvar melanosis is a benign pigmented condition that is characterized by pigmentation of the vulvar skin without elevation or nodularity of the involved skin in an area that is, by definition, larger than 5 mm in diameter. Like lentigo simplex the melanosis is flat and the skin markings over the pigmented area are unchanged and similar to the adjacent skin. Histopathologic examination of vulvar melanosis and lentigo simplex demonstrate that they are essentially indistinguishable. Therefore the size of the lesion is important and influences the clinical diagnosis and the pathologic interpretation. Similar pigmentation with similar findings when found in the vagina is referred to as melanosis vaginae, and in some cases melanosis of both the vulva and the vagina may be found. Vulvar Nevomelanocytic Nevi The types of nevomelanocytic nevi identified on the vulva include junctional, compound, and intradermal. In most series compound and intradermal nevi are more common. Atypical vulvar nevi are a special category of nevus and these often have some features of compound nevi.

Junctional Nevi Junctional nevi have the nevus cells located within the epidermis as well as the dermal–epidermal junction. The nevus cells are immediately adjacent to the squamous keratinocytes. Junctional nevi are young and somewhat undifferentiated. Compound Nevi A compound nevus represents a continuum of maturation from a junctional to an intradermal nevus. In compound nevi, the nevus cells are within both the epidermis and the dermis. Nevus cells are seen within the dermal–epidermal junction, but also within the dermis, where dermal reticulum, collagen, and elastic fibers envelop the nevus cell nest. Compound nevi have elevated surfaces as compared to the surrounding skin. Intradermal Nevi Intradermal nevi do not have a dermal–epidermal component and the nevus cells are completely surrounded within the dermis by connective tissue elements. Over time an intradermal nevus may entirely regress, or evolve into an acrochordon (fibroepithelial polyp). Atypical Vulvar Nevi Of special interest are the atypical vulvar nevomelanocytic nevi that may clinically and histopathologically resemble malignant melanoma ( Fig. 33.10). Atypical nevi are found predominately in adolescent and premenopausal women. Pathologically they have features of compound nevi, having both a junctional and an intradermal component, but they also have cytoatypia, especially in the more superficial nevomelanocytes.

FIG. 33.10. Vulvar atypical nevus in an adolescent patient. The lesion is raised and has somewhat irregular borders. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) They can be distinguished histopathologically from malignant melanoma by three specific features. These include symmetry in horizontal and vertical dimensions, decrease in cell size with maturation of the nevus cells in the deeper levels, and lack of pagetoid single cell spread in the upper one-third of the epidermis. Congenital and Giant Nevomelanocytic Nevi Congenital nevi occur in approximately 10% of newborns and most of these nevi are less than 4 mm in diameter. Giant nevomelanocytic nevi, which by definition are 20 cm or larger in diameter (“garment type” nevus), are rare but are associated with increased risk of developing malignant melanoma in prepubertal individuals. The frequency of involvement of the vulva by these types of nevi is uncertain. Other Benign Pigmented Lesions Other benign pigmented lesions of the vulva include seborrheic keratosis, postinflammatory (reactive) hyperpigmentation, and dermatofibroma (benign fibrous histiocytoma). Seborrheic Keratosis Seborrheic keratosis often presents as a macular or papular pigmented lesion that usually feels waxy on palpation ( Fig. 33.11). The lesions are usually multiple and may be seen in other sites as well as the vulva in a given patient. In some cases the surface of the lesion can be peeled away to reveal punctate bleeding points at the base. In most cases, however, biopsy is necessary to establish the diagnosis. Biopsy is especially useful to distinguish seborrheic keratosis from vulvar intraepithelial neoplasia, which can also present as pigmented macular or papular lesions.

FIG. 33.11. Seborrheic keratosis. Typical appearance of a waxy-appearing, elevated, pigmented lesion. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) On histopathologic examination seborrheic keratosis has a thickened epithelium (acanthosis) with a hyperkeratotic surface. The deeper tips of the rete tend to have uniform length. No significant dermal compression or inflammatory infiltrate is present in most cases. Treatment of seborrheic keratosis is usually local superficial excision or cryotherapy. Reactive Hyperpigmentation Reactive hyperpigmentation is usually preceded by chronic inflammation (postinflammatory) or injury. Although the skin is pigmented in the involved areas, the surface of the skin maintains its normal character, without surface elevation. Biopsy is usually not necessary to establish the diagnosis although the usual evaluation for the cause of the inflammatory process is necessary. No therapy is needed. Dermatofibroma/Fibrous Histiocytoma Dermatofibroma is rare on the vulva, however, it may present as a slightly raised, nonpolypoid pale brown to red solitary subcutaneous mass. The term dermatofibroma is reserved for those lesions 1.5 cm in diameter or less. Masses larger than this, or that are polypoid in appearance, are referred to as fibrous histiocytoma. Biopsy is necessary to establish the diagnosis and to distinguish these benign histiocytic tumors from dermatofibrosarcoma protuberans or malignant fibrous histiocytoma. Management of Pigmented Lesions As a general rule, any elevated pigmented lesion of the vulva that is 5 mm or larger in greatest dimension, or otherwise clinically suspicious, should be excised. In these cases, excision, with submission of the specimen for histopathologic examination is both diagnostic and therapeutic. If malignant melanoma is suspected, and outpatient biopsy is considered reasonable, wide and deep excision is needed to properly diagnose and pathologically stage the lesion if it is a melanoma. Three centimeter (3 cm) margins of excision, both wide and deep, can be considered in such cases. Immediate consultation with a specialist familiar with the diagnosis and treatment of melanoma of the vulva may be the best alternative for management of patients with a lesion suspected of being melanoma, rather than attempted excision or biopsy in the outpatient setting. Depigmentation Disorders Depigmentation disorder include vitiligo, albinism, and postinflammatory hypopigmentation. Vitiligo is the loss of melanocytes from where they were once present, and is due to an inherited disorder. Albinism is a congenital inability of melanocytes to produce melanin. Postinflammatory hypopigmentation is the loss of melanocytes in the skin secondary to inflammation or injury ( Fig. 33.12).

FIG. 33.12. Postinflammatory hypopigmentation. The white area has lost melanocytes and melanin. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) See color figure 33.12.

BENIGN TUMORS OF THE VULVA Condyloma Acuminatum HPV infection is the viral cause of vulvar and vaginal condylomata acuminata. HPV is also associated with VIN, a lesion that is a known precursor of certain types of vulvar carcinoma. Condyloma acuminatum (genital wart) is a benign neoplasm that is usually sexually transmitted and can involve the vulva as well as the vagina, cervix, urethra, anus, and perianal skin. HPV-6 and HPV-11 are the usual HPV types associated with genital condyloma acuminatum; HPV-11 is observed in about one-fourth of cases. On clinical examination condylomata acuminata of the vulva are usually papillary, verrucous, or macular in character and can involve the skin as well as the vulvar vestibule, vagina, and cervix. The lesions are usually multiple. They may be asymptomatic but can be pruritic and may result in the patient coming to the clinic when she observes or palpates the lesion or lesions. Condylomata acuminata of the vulva and vagina may be associated with pregnancy, diabetes mellitus, perineal hygiene problems, immunosuppression, and other related problems. Approximately one-third to one-half of the women with vulvar condyloma acuminatum have associated cervical HPV-related lesions (cervical intraepithelial neoplasia/dysplasia). In children, genital HPV infection may be evidence of sexual abuse. Magnification of the vulva and vagina with a ring light, or colposcope, and the application of 3% to 5% acetic acid can greatly aid identification of smaller lesions because the HPV-related lesions that involve the mucosa are aceto-white, becoming white in color with the application of the acetic acid. Lesions detectable only by colposcopic examination are considered subclinical HPV infections. Histopathologic examination of a typical vulvar or vaginal condyloma acuminatum demonstrates epithelial thickening usually associated with acanthosis, koilocytosis, and parakeratosis. Lesions arising in the vulvar skin often have hyperkeratosis and a prominent granular layer. Superficial dermal or submucosal chronic inflammatory cells may be seen beneath the condyloma. Vulva condyloma acuminatum may persist for years however regression may occur and has been observed following pregnancy. Presentation following pelvic radiation therapy or with immunosuppression has also been observed. Differential Diagnosis Within the vulva there is an important distinction made between typical condyloma acuminatum and vulvar intraepithelial neoplasia of high grade (VIN 2 or VIN 3/moderate or severe dysplasia or carcinoma in situ). Many pathologists would prefer to refer to macular condyloma acuminatum (condyloma plana) of the vulva as VIN 1, as referenced in leading texts on vulvar pathology, but this is not a uniform practice. Unlike most VIN lesions, typical vulvar or vaginal condylomata are not usually associated with HPV-16 or other high risk HPV types as are VIN 2 or VIN 3 lesions. They are typically DNA-diploid whereas VIN 2 or VIN 3 lesions are usually DNA-aneuploid. Fibroepithelial Polyp Fibroepithelial polyps may clinically resemble large condylomata, however, the epithelium of a fibroepithelial polyp lacks significant proliferation, unlike the epithelium of typical condyloma acuminatum ( Fig. 33.13). This proliferation can be demonstrated immunohistochemically using Ki-67 (MIB-1) immunohistochemistry, where the proliferative cells of condyloma are seen throughout the epithelium, while fibroepithelial polyps only have proliferative cells in the basilar epithelial layers. Condylomata lata or secondary syphilis can suggest condylomata acuminata clinically; however, biopsy demonstrates inflammation and arteritis with many plasma cells. Spirochetes can be demonstrated by immunofluorescence on air-dried scraping of the surface of the lesion, or on Warthin-Starry silver stain of a tissue section. Syphilis serologic studies are appropriate in cases where there is question regarding the differential diagnosis.

FIG. 33.13. Fibroepithelial polyp (acrochordon) with a pedunculated appearance. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.) Squamous cell carcinoma can arise in vulvar condyloma acuminatum, and vulvar verrucous carcinoma may resemble a large condyloma. In addition, the so-called giant condyloma of Buschke and Löwenstein is now recognized to be a variant of verrucous carcinoma. The association of vulvar and vaginal condyloma acuminatum with invasive squamous cell carcinoma has been associated with lymphoproliferative disorders, such as Hodgkin disease, as well as chromosomal disorders with immunosuppression related to Fanconi anemia. Therapy There are many methods to treat vulvar condyloma acuminatum. These include clinician-applied topical concentrated trichloroacetic acid or bichloracetic acid. This treatment method may be effective for small vulvar condylomata. Larger lesions are more commonly treated with surgical excision, electrodesiccation with surface curettage, or electro-loop excision. Cryosurgery and laser ablation have also been used for larger lesions, however these two techniques are usually not the treatments of choice for larger lesions unless the clinician is experienced in their use on the vulva. Topical application by the clinician of a dilute podophyllin toxin preparation is also effective for small lesions, but contraindicated in pregnancy and should not be used in the vagina. These medications inhibit cell proliferation and can be neurotoxic if systemically absorbed through mucosal surfaces. Imiquimod is a topical, immune-enhancing medication that has also been successful in treating small and nonkeratinized condylomata. This medication can be applied by the clinician, or by a well-informed patient who is capable of doing the applications. Approximately 40% to 60% of children with laryngeal papillomatosis are born from mothers with a history of genital condyloma acuminatum. Laryngeal papillomas of infancy and childhood are thought to be acquired at delivery. Although the incidence of HPV infection of the larynx of the newborn from maternal genital papillomavirus infection is unknown, no correlation has been established between the volume of maternal wart tissue and occurrence of infantile laryngeal papillomatosis. Approximately one-half of patients with laryngeal papilloma have HPV-11. Leiomyoma of the Vulva Leiomyoma is a benign smooth muscle tumor and is the most common vulvar soft tissue tumor ( Fig. 33.14). In children, vulvar leiomyomata may be associated with

esophageal or gastric leiomyomatosis. Histopathologic examination demonstrates that the tumor is composed of smooth muscle cells that can have myxoid change or epithelioid features, both are especially noted in pregnancy. Leiomyoma, in contrast to leiomyosarcoma, has a low mitotic count, lacks significant cytologic atypia, and is not infiltrative or metastatic. Immunohistochemical studies can distinguish these tumors from rhabdomyosarcoma and rhabdomyoma. Therapy is local excision. Gonadotropin inhibition medical therapy may be of value to reduce tumor size before surgery.

FIG. 33.14. Leiomyoma. Large, firm, and somewhat mobile nontender tumor. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Benign Vulvar Tumors of Sweat Gland Origin There are several benign epithelial tumors of sweat gland origin that occur in the vulva. The two most important are papillary hidradenoma and syringoma. Other important benign epithelial tumors include mammary-like tumors and clear cell hidradenomas. Papillary Hidradenoma Papillary hidradenomas of the vulva may present as a nodule or a papillary mass. They may form a papillary-like mass that bleeds easily if the tumor mass everts from its subepithelial location. These tumors are usually located in the interlabial sulcus and occur after puberty ( Fig. 33.15). They are usually under 2 cm in diameter and are not infiltrative. The tumor is currently thought to arise from specialized mammary-like anogenital glands that are found predominately in the interlabial sulcus.

FIG. 33.15. Papillary hidradenoma, asymptomatic, located in the sulcus between the labia minus and majus. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

On histopathologic examination the deep relationship of the tumor with the underlying dermis reveals a circumscribed papillary tumor with a pushing, noninfiltrative boarder. The epithelial component of the tumor is composed of superficial secretory apocrine cells with a second deeper layer of myoepithelial cells below the secretory cells. Local excisional biopsy as a form of treatment is both diagnostic and therapeutic. Excision should be sufficiently deep to encompass the deep portion of the tumor margin. This will provide diagnostic pathologic material as well as be adequate therapy. Superficial biopsies can be somewhat misleading in some cases if the two cell layers of the papillary elements are not apparent and the relationship with the dermis is not seen. Mammary-like Tissue: Arising from the Specialized Anogenital Glands Mammary-like tumors were once considered to be from ectopic breast tissue in the vulva, however current evidence supports that the vulva is not in the human milk line and does not contain breast tissue. These tumors may be hidradenoma papilliferum, or resemble benign or malignant breast tumors. Metastatic breast adenocarcinoma is in the differential diagnosis in such cases. For the benign tumors the treatment is as for hidradenoma papilliferum. Clear Cell Hidradenoma Clear cell hidradenoma is rare on the vulva, however the histologic appearance, consisting of large regular cells with prominent clear cytoplasm, may present a diagnostic problem for the pathologist. As with the papillary hidradenoma, diagnostic excisional biopsy will be both therapeutic and will greatly aid in the pathologic diagnosis. These tumors, although usually solid, are well defined, with a pushing border at the tumor stromal interface without an infiltrative pattern. Syringoma Syringoma usually presents as multiple small, skin-colored papules often occurring bilaterally on the vulva. Pruritus may be a presenting symptom. Syringomas are considered to be benign adenomas of the ducts of the eccrine sweat glands. The histopathologic diagnosis can be made from a representative biopsy of one of the papules. The comma-shaped, dilated duct elements within the dermis usually contain an inner myoepithelial layer and luminal epithelial cell layer. Syringomas are benign tumors; however, malignant variants of skin appendage tumors that may contain syringoma-like elements (mixed tumors of the vulva) have been reported. Syringoma is usually not treated surgically unless there is a specific mass in question that needs excision for diagnostic purposes. Topical or medical treatment is preferred. Vascular Lesions Capillary hemangioma occurs in infants and children. It is a slightly elevated, well-defined, red-to-violet lesion with an irregular surface. Ulceration may occur. Capillary hemangiomas can be diagnosed clinically without biopsy. They typically regress over time without treatment. On microscopic examination the lesion consists of well-demarcated, clustered, small capillaries in a fibrous tissue. Cavernous hemangiomas also may occur in children but are rare in the vulva. They are larger, deeper, and more complex than capillary hemangiomas. They can be associated with pelvic hemangiomas. They can involve the clitoris or labia and resemble hypertrophy. Cavernous hemangiomas will usually regress over time and therefore do not require treatment unless they bleed or otherwise become symptomatic. Acquired hemangioma, also known as senile hemangioma, is common in the vulva in adults. The lesions are usually small, red-to-violet-to-purple papules, and are usually multiple. They do not require therapy but may resemble pigmented lesions and therefore may be biopsied. On microscopic examination acquired hemangiomas

are composed of clustered capillary-like vessels within the dermis, with fibrous tissue about the vessels that may be rich in collagen. Differential diagnosis of hemangiomas includes hemangiomatous-like changes that may be seen in the vulva after radiation therapy. In addition, angiokeratoma, pyogenic granuloma, bacterial angiomatosis, hemangiopericytoma, angiosarcoma, and Kaposi sarcoma are all vascular lesions that can be included in the differential diagnosis. Angiokeratoma Angiokeratoma is a benign vascular tumor that is a variant of hemangioma. They are relatively common on vulva and present as a dark red to purple to black, papular lesion that may have an appearance of a condyloma acuminatum ( Fig. 33.16). These vascular lesions are often multiple and usually asymptomatic although they may ulcerate and bleed. They can be associated with Fabry disease. Their elevated and dark appearance may result in them resembling a nevus and this prompts excisional biopsy.

FIG. 33.16. Angiokeratoma. Multiple angiokeratomas in a woman with Fabry disease. White blood cell a-galactosidase of 36 nm per hour per mg (normal = 50–80). (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

On microscopic examination the lesion has a warty appearance with acanthosis and usually some hyperkeratosis. A distinctive feature is that the basal layer of the overlying epithelium is nearly contiguous with the endothelial cells lining the underlying vascular channels of the angiokeratoma. This benign vascular lesion is confined to the superficial dermis and is not infiltrative, as seen with Kaposi sarcoma or angiosarcoma. Angiokeratoma is a benign lesion and does not require treatment unless bleeding or ulceration occurs, or excision is prompted because of its clinical appearance. Pyogenic Granuloma (Granuloma Pyogenicum) Pyogenic granuloma typically presents as an elevated, papular, red lesion that in the fully developed state is ulcerated. It is a variant of hemangioma and is analogous to the epulis tumor of pregnancy. Pyogenic granulomas on the vulva occur usually related to pregnancy but may arise in vulvar skin wound infection. The vascular tumor can have rapid growth and can become secondarily infected. Histologically, the findings resemble granulation tissue with inflammation. In fully developed cases at the ulcer edge a “collarette” of elevated dermis and epithelium surrounds the lesion. Bacillary Angiomatosis Bacillary angiomatosis clinically presents as a macular and papular red skin lesion or lesions that may resemble Kaposi sarcoma or hemangioma. These lesions are apparently caused by infection with the bacteria Bartonella henselae and B. quintana (formally, Rochalimaea henselae) and R. quintana. The bacterial organisms can be detected in tissue with appropriate silver stains. Kaposi Sarcoma Kaposi sarcoma is a vascular tumor associated with human immunodeficiency virus (HIV) and herpesvirus (herpesvirus VIII). Neoplastic-appearing, endothelial-appearing cells grow as spindle-shaped cells that form “slit-like” spaces that are filled with red blood cells.

VULVAR CYSTS Bartholin Cyst and Bartholin Abscess The Bartholin cyst is a result of distal obstruction of the Bartholin duct. Obstruction results in accumulation of secretion and cystic dilation of the duct ( Fig. 33.17). The lining of the Bartholin cyst is epithelium that can be squamous, transitional, low cuboidal mucinous epithelium, or otherwise not classifiable. The Bartholin cyst is filled with mucoid sialomucin that is sterile and stains with mucicarmine, periodic acid-Schiff (PAS) with and without diastase digestion, and Alcian blue at pH 2.5. The cyst epithelium is immunoreactive for carcinoembryonic antigen (CEA).

FIG. 33.17. Bartholin cyst. Bartholin duct cyst expanding the left labium majus. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Bartholin cysts are usually treated with marsupialization, using a Word catheter or equivalent to establish a new duct exit to the vestibular surface. In postmenopausal women, a persistent Bartholin mass after drainage or recurrent cysts may require surgical excision to resolve the process and exclude adenocarcinoma of the Bartholin gland. Bartholin Abscess

Bartholin abscess may be caused by Neisseria gonorrheae infection, or related to infection with Staphylococcus or anaerobic organisms. The presenting symptoms are pain and sometimes fever. The physical findings demonstrate tenderness, redness, and swelling in the Bartholin gland area. Microscopically, the Bartholin duct has acute inflammation within the duct as well as within the gland stroma about the duct. The abscess, when fully developed, contains purulent exudate. Treatment of choice is excision and drainage. Antibiotic therapy that the related organism is sensitive to is part of the treatment. In some cases the infection may subside without abscess formation. Persistent chronic inflammation may occur in some cases. Bartholin duct cyst may follow stenosis of the distal Bartholin duct secondary to chronic inflammation of the duct. Keratinous Cyst (Epithelial Inclusion Cyst) Keratinous cysts are relatively common on the vulva and can involve the majora, clitoris, perineal body, as well as other vulvar sites ( Fig. 33.18). These cysts usually range in size from 2 to 5 mm, but may be larger, and they are typically superficial and can be multiple. They may occur at any age, and can occur in newborns. Milia, a type of multiple keratinous cysts, can occur in the vulva. Keratinous cysts may occur in episiotomy scars and have been observed as a complication of female circumcision. Keratinous cysts contain keratinous material that is white to pale yellow in color and grumous or cheesy in character. Hair is not present in these cysts. Keratinous cysts are not considered premalignant, although carcinoma arising in keratinous cysts may occur.

FIG. 33.18. Keratinous cyst (epidermal inclusion cyst). The cyst is firm but compressible. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

On microscopic examination, the cyst lining is flattened, stratified squamous epithelium that is immunoreactive for high-molecular-weight keratin. These cysts may be associated with occluded sebaceous gland ducts that have undergone squamous metaplasia. In the adjacent tissue to the cyst wall, foreign body-type giant cells associated with keratinous material that has leaked into the dermis can be found. Treatment is not necessary if the cysts are asymptomatic. Surgical excision is considered if the diagnosis is in question or if the cysts are enlarging, symptomatic, or become infected. Mucous Cyst Mucous cyst occurs primarily in the vulvar vestibule ( Fig. 33.19). These cysts apparently arise from occlusion of minor vestibular glands. These cysts are typically simple cysts, superficial, and lined by mucus-secreting, cuboidal-to-columnar epithelium without associated myoepithelial cells. Metaplastic squamous epithelium may be present with the columnar epithelium lining the cyst. The epithelial cyst lining stains with both Alcian blue and Mayer mucicarmine. This distinguishes them from cysts of mesonephric origin that do not stain with these agents.

FIG. 33.19. Mucinous cyst of the vulvar vestibule. (From Wilkinson, Stone. Atlas of vulvar disease. Philadelphia: Lippincott, Williams and Wilkins, with permission.)

Columnar Cell Metaplasia Columnar cell metaplasia has been described within the vulvar vestibule epithelium, as well as within the vaginal mucosa. Columnar cell metaplasia presents as a solitary flattened red area of the labia minora, or as persistent redness of the upper vagina. In one reported case the process responded to topical estrogen cream. Columnar cell metaplasia of the nonkeratinized squamous epithelium of the vulva vestibule or vagina is a condition where columnar epithelium of mucinous or tubal–epithelial type replaces the squamous epithelium. This has been described after Stevens-Johnson syndrome as well as after laser and 5-FU therapy. Columnar cell metaplasia appears to be the origin of vulvar ciliated cysts. These cysts are lined with columnar epithelium resembling müllerian epithelium (tubal–endometrial epithelium), consisting of ciliated, secretory, and undifferentiated type cells. The cysts within the vestibule are considered acquired, since the müllerian system does not contribute to the development of the vestibule, which is of urogenital sinus origin. The cysts are distinguished from endometriosis by the absence of associated endometrial stroma or hemosiderin-laden macrophages. Therapy for these cysts in the vestibule is generally not needed and they can be followed by observation alone. If the cyst or cysts persist, enlarge, or become symptomatic, they can be excised. Vulvar Endometriosis Vulvar endometriosis represents ectopic endometrial epithelium. The origin may be from endometrium implanted following vulvar or vaginal tears, or at the time of delivery, or endometrium shed at menstruation. Vascular spread of endometrium to the vulva or vagina may also be a consideration. On clinical presentation, superficial vulvar or vaginal endometriosis usually presents as a mass that is blue-black to violet in color. In the vulva the mass is often in the posterior fourchette. Deeper endometriosis usually has no associated skin or mucous membrane discoloration. Cyclic pain and enlargement with the menstrual cycle

may be noted. Diagnosis may be made using fine-needle aspiration cytology or local excision. On histologic examination endometrial glands and stroma are usually identified, often associated with hemosiderin-laden macrophages. Apocrine Hydrocystoma/Cystadenoma of the Vulva Apocrine hydrocystoma is a cyst with an epithelial lining with apocrine differentiation. The lining cells are apocrine in character with pink cytoplasm and apical snoutlike secretion. Apocrine cystadenoma is a cyst lined with apocrine-differentiated cells with an epithelial lining that has micropapillary projections. These cysts typically arise in the apocrine gland–bearing areas of the vulva. Mesonephric-like Cyst (Wolffian-like Duct Cyst) Mesonephric-like cysts occur on the lateral walls of the vagina, and occasionally within the lateral vulvar vestibule. These cysts are typically thin-walled, translucent in appearance, and contain a clear fluid. The cuboidal to columnar epithelial lining is not ciliated. Using immunohistochemistry for smooth muscle, such as smooth muscle actin, smooth muscle can be seen deep and surrounding these cysts. Cyst of the Canal of Nuck (Mesothelial Cyst) Entrapped peritoneum (processus vaginalis) within the round ligament may result in a cyst of the canal of Nuck, causing labial enlargement. These cysts apparently arise from inclusions of the peritoneum at the inferior insertion of the round ligament into the labia majora. Cysts of the canal of Nuck typically occur in the superior labia majora or inguinal canal. Therefore they are analogous to spermatic cord hydrocele. These mesothelial-lined cysts can achieve substantial size and a concurrent inguinal hernia, which occurs in approximately one-third of cases, and sometimes obscures their diagnosis. These cysts may mimic inguinal hernia on clinical examination. SUGGESTED READINGS Benign Vaginal Conditions and Vaginitis 1.

Sobel JD. Vulvovaginitis in healthy women. Compr Ther 1999;25:335–346.

2.

Zaino RJ, Robboy S, Kurman RJ. Diseases of the vagina. In: Blaustein's pathology of the female genital tract, fifth ed. New York: Springer-Verlag, 2002:151–206.

Vulvar Non-Neoplastic Epithelial Disorders Including Vulvar Lichen Sclerosus and Squamous Hyperplasia (Hyperplastic Dystrophy) 3. Carlson JA, Ambros R, Malfetano J, et al. Vulvar lichen sclerosus and squamous cell carcinoma: a cohort, case control, and investigational study with historical perspective; implications for chronic inflammation and sclerosis in the development of neoplasia. Hum Pathol 1998;29:932–948. 4. Carlson JA, Lamb P, Malfetano J, et al. Clinicopathologic comparison of vulvar and extragenital lichen sclerosus: histologic variants, evolving lesions, and etiology of 141 cases. Mod Pathol 1998;11:844–854. 5.

International Society for the Study of Vulvar Disease. New nomenclature for vulvar disease. Obstet Gynecol 1976;47:122.

6.

Powell JJ, Wojnarowska F. Lichen sclerosus. Lancet 1999;353:1777–1783.

7.

Turner MLC. Vulvar manifestations of systemic diseases. Dermatol Clin 1992;10:445–458.

Ulcers and Infectious Diseases 8.

Elgart ML. Sexually transmitted diseases of the vulva. Dermatol Clin 1992;10:387–403.

9.

Moreland AA. Vulvar manifestations of sexually transmitted diseases. Semin Dermatol 1994;13:262–268.

10.

Sakane T, Takeno M, Suzuki N, et al. Behçet's disease. N Engl J Med 1999;341:1284–1291.

11.

Toussaint S, Kamino H. Noninfectious erythematous, papular and squamous diseases. In: Elder D, ed. Lever's histopathology of the skin, eighth ed. Philadelphia: Lippincott-Raven, 1997:51–184.

12.

Vettraino IM, Merritt DF. Crohn's disease of the vulva. Am J Dermatopathol 1995;17:410–413.

Vulvar Vestibulitis, Vulvar Squamous Papillomatosis, Nevi, and Cysts 13.

Friedrich EG Jr. Vulvar vestibulitis syndrome. J Reprod Med 1987;32:110–114.

14.

Friedrich EG Jr, Wilkinson EJ. Mucous cysts of the vulvar vestibule. Obstet Gynecol 1972;42:407.

15.

Growdon WA, Fu YS, Lebherz TB, et al. Pruritic vulvar squamous papillomatosis: evidence for human papillomavirus etiology. Obstet Gynecol 1985;66:564–568.

16.

Prayson RA, Stoler MH, Hart WR. Vulvar vestibulitis. A histopathologic study of 36 cases, including human papillomavirus in situ hybridization analysis. Am J Surg Pathol 1995;19:154–160.

17.

Ridley CM, Neill SM. Non-infective cutaneous conditions of vulva. In: Ridley CM, Neill SM, eds. The vulva, second ed. Oxford; Malden, MA: Blackwell Science, 1999:121–186.

18.

Rock B, Hood AF, Rock JA. Prospective study of vulvar nevi. J Am Acad Dermatol 1990;22:104–106.

19.

Sarma AV, Foxman B, Bayirli B, et al. Epidemiology of vulvar vestibulitis syndrome: an exploratory case-control study. Sex Transm Infect 1999;75:320–326.

20.

Schneider CA, Festa S, Spillert CR, et al. Hydrocele of the canal of Nuck. N J Med 1994;91:37–38.

21.

Sedlacek TV, Riva JM, Magen AB, et al. Vaginal and vulvar adenosis. An unsuspected side effect of CO

22.

van der Putte SCJ. Mammary-like glands of the vulva and their disorders. Int J Gynecol 1994.

2 laser

vaporization. J Reprod Med 1990;35: 995–1001.

23. Wolber RA, Talerman A, Wilkinson EJ, et al. Vulvar granular cell tumors with pseudocarcinomatous hyperplasia: a comparative analysis with well-differentiated squamous carcinoma. Int J Gynecol Pathol 1991;10: 59–66.

Additional Books and Book Chapters 24.

Black MM, McKay M, Braude P. Obstetric and gynecologic dermatology. London: Mosby-Wolfe, 1995.

25.

Fisher BK, Margesson LJ. Genital skin disorders. Diagnosis and treatment. In: Baxter S, ed. St. Louis: Mosby, 1998.

26.

Kaufman RH, Friedrich EG, Gardner HL. Benign diseases of the vulva and vagina, third ed. Chicago: Year Book Medical Publishers, 1989.

27.

Kurman RJ, Norris HJ, Wilkinson EJ. Tumors of the cervix, vagina, and vulva. In: Rosai J, ed. Atlas of tumor pathology, fascicle. Washington, DC: Armed Forces Institute of Pathology, 1992.

28.

Lynch PJ, Edwards L. Genital dermatology. New York: Churchill Livingstone, 1994.

29.

Wilkinson EJ, Stone IK. Atlas of vulvar disease. Baltimore: Williams & Wilkins, 1995.

30.

Wilkinson EJ, Xie D. Benign disease of the vulva. In: Kurman RJ, ed. Blaustein's pathology of the female genital tract, fifth ed. New York: Springer-Verlag, 2002.

31.

Wilkinson EJ. Neoplastic diseases of the vulva. In: Kurman RJ, ed. Blaustein's pathology of the female genital tract, fifth ed. New York: Springer-Verlag, 2002.

Chapter 34 Amenorrhea Danforth’s Obstetrics and Gynecology

Chapter 34 Marc A. Fritz

Amenorrhea

THE NORMAL MENSTRUAL CYCLE Differential Diagnosis of Amenorrhea Genital Tract Abnormalities Ovarian Disorders Pituitary Disorders Hypothalamic Disorders EVALUATION OF AMENORRHEA Medical History POLYCYSTIC OVARY SYNDROME Determination of Estrogen Status Hypothalamic Amenorrhea TREATMENT OF AMENORRHEA Genital Tract Abnormalities Ovarian Disorders Hypothalamic Disorders SUMMARY POINTS SUGGESTED READINGS Differential Diagnosis of Amenorrhea Genital Tract Problems Gonadal Dysgenesis Premature Ovarian Failure Pituitary Disorders (Hyperprolactinemia) Hypothalamic Amenorrhea Polycystic Ovarian Syndrome

Amenorrhea is the absence or cessation of menses and may result from a wide variety of pathologic conditions. It is a common symptom of an underlying abnormality in the reproductive system that may be anatomic (developmental or acquired), organic, or endocrinologic in nature. This chapter outlines the differential diagnosis of amenorrhea, discusses the indications and methods for evaluation, and describes options for treatment once a diagnosis and the patient's goals are clearly defined. However, the most common cause of amenorrhea in women of reproductive age is pregnancy and that should be ruled out before considering other etiologies.

THE NORMAL MENSTRUAL CYCLE Normal menstrual function involves a complex multilevel integration of endocrine signals, local autocrine and paracrine mechanisms, and target cell receptors, all operating at four distinct levels: the genital tract, the ovary, the pituitary gland, and the hypothalamus. First, normal menstrual function requires a normal genital outflow tract. The uterus must have a functional endometrium capable of response to estrogen and progesterone, and be continuous with the cervix, vagina, and introitus. Second, the ovaries must contain follicles responsive to pituitary follicle-stimulating hormone (FSH) and luteinizing hormone (LH) stimulation. Progressive follicular development and ovulation further require the normal operation of local intraovarian regulatory mechanisms that are sensitive to changes in the endocrine milieu. Third, pituitary gonadotrophs must have the capacity to synthesize and secrete gonadotropins in response to hypothalamic gonadotropin-releasing hormone (GnRH) stimulation. The relative amounts of FSH and LH released reflect changes in the pulsatile pattern of GnRH secretion and the feedback modulation of ovarian steroid and peptide hormones. Finally, specialized neurosecretory cells located in the medial basal hypothalamus (arcuate nucleus) must have functional communication with the pituitary gland and be able to synthesize and release GnRH in a pulsatile pattern that varies in response to stimuli from the environment and feedback signals from the periphery. Amenorrhea may result from congenital or acquired disease or dysfunction at the level of the genital tract, the ovary, the pituitary, or the hypothalamus. In fact, the single most common cause of amenorrhea—chronic hyperandrogenic anovulation (polycystic ovary syndrome; PCOS)—involves a number of interrelated pathophysiologic mechanisms that operate at the ovarian, pituitary, and hypothalamic levels, and does not fall neatly into any one specific category. The wide array of disorders that may be responsible suggests that amenorrhea may present a daunting diagnostic challenge but, in truth, evaluation is relatively straightforward, logical, and requires only tests and procedures with which all gynecologists should be quite familiar. With few exceptions, an accurate diagnosis can be confidently established in very little time and without great expense.

Differential Diagnosis of Amenorrhea Although the list of potential causes of amenorrhea is long, the majority of cases relate to one of five conditions: pregnancy, PCOS, hypothalamic amenorrhea, hyperprolactinemia, and ovarian failure ( Table 34.1). All of the remaining causes are relatively uncommon and only occasionally encountered in a lifetime of clinical practice.

TABLE 34.1. Causes of amenorrhea

Genital Tract Abnormalities The embryology of the female genital tract involves the medial migration and midline fusion of the paired müllerian (paramesonephric) ducts to form the uterus, cervix, and upper vagina, and the vertical fusion of that developing ductal system with the invaginating urogenital sinus to form the lower vagina and the introitus. Outflow tract abnormalities that result from failure of müllerian duct development include vaginal/müllerian agenesis and androgen insensitivity syndrome (AIS), where the uterus is altogether absent. Abnormalities caused by failure of vertical fusion include imperforate hymen, transverse vaginal septum, and cervical atresia. These conditions result in an accumulation of menstrual effluent above the level of obstruction (cryptomenorrhea). With two notable exceptions, all outflow tract abnormalities are developmental in origin and therefore cause primary amenorrhea. Asherman syndrome and cervical stenosis/obstruction are acquired conditions and therefore are causes of secondary, rather than primary, amenorrhea. Asherman syndrome results from intrauterine adhesions that obstruct or obliterate the uterine cavity as a consequence of inflammation (postpartum endometritis, retained products of conception) or trauma (curettage). Severe cervical stenosis with complete outflow

obstruction is a rare complication of cervical conization procedures or other surgical treatments for cervical intraepithelial neoplasia. Ovarian Disorders Ovarian failure occurs when few or no follicles remain that are capable of producing estradiol in response to pituitary gonadotropin stimulation. Follicular depletion may occur during embryonic life with no follicles remaining by infancy or early childhood, after puberty has begun but before menarche, or at some later time before menopause would normally be expected. Therefore, depending on when the available supply of ovarian follicles is functionally depleted, puberty may not occur, it may begin normally but stop before the first menses, or it may progress normally to and beyond menarche with secondary amenorrhea having onset at some later point in time. Gonadal dysgenesis is among the most common of all causes of primary amenorrhea (approximately 30%–40%) and results from an absence of ovarian follicles or accelerated follicular depletion during embryogenesis or the first few years of life. The gonads of affected individuals contain only stroma and appear as fibrous streaks. The most common form of gonadal dysgenesis is Turner syndrome, classically associated with a 45,X karyotype, but also with an assortment of other structural X chromosome abnormalities (deletions, ring, and iso-chromosomes). These X chromosomal anomalies may be present in all or only in some of the cells of the body (mosaicism), depending on the stage of embryonic development at the time that they arise. Although less common, individuals with gonadal dysgenesis also may have a normal 46,XX or a 46,XY karyotype (Swyer syndrome) in all cells or in one or more cell lines in mosaic individuals (e.g., 45,X/46,XX; 45,X/46,XY). Most, in the absence of ovarian follicles, have no significant secondary sexual development. A few may have transient normal ovarian function, the extent depending on when the limited supply of ovarian follicles is depleted. Approximately 15% begin but do not complete pubertal development and approximately 5% have sufficient follicles to complete puberty and begin spontaneous menstruation. Spontaneous pregnancies rarely occur and are associated with a relatively high risk for sex chromosome aneuploidy and spontaneous abortion. Premature ovarian failure (POF) results in secondary amenorrhea at some time after puberty has been completed. It is distinguished from gonadal dysgenesis on the basis of ovarian morphology and histology; instead of streak gonads, the ovaries in POF more closely resemble those of postmenopausal women. Approximately 1% to 5% of women will develop POF before the age of 40 years. The karyotype in individuals with POF is most often normal (46,XX), but also may reveal mosaicism (e.g., 45,X/46,XX). A specific cause for early follicular depletion in POF frequently cannot be determined and is presumed to result from inadequate germ cell migration during embryogenesis or accelerated atresia. POF is frequently associated with autoimmune disorders and in some cases (e.g., Addison disease) appears to result from an autoimmune lymphocytic oophoritis. Radiation and chemotherapy are two other important causes of ovarian failure. The effects of both are dependent upon dose and the age at time of treatment. Galactosemia is an autosomal recessive disorder of galactose metabolism caused by a deficiency of the enzyme galactose 1-phosphate uridyltransferase and another, albeit very rare, cause of POF. Affected women have fewer primordial follicles presumably due to the cumulative toxicity of galactose metabolites on germ cell migration and survival. Other, very rare, ovarian disorders that may cause amenorrhea include 17a-hydroxylase deficiency, aromatase deficiency, and the gonadotropin-resistant ovary syndrome. Unlike in ovarian failure, the ovaries of individuals with these disorders contain follicles and oocytes, but cannot produce estrogen. The enzyme 17 µ-hydroxlase mediates an early step in steroid hormone synthesis, without which progesterones cannot be converted to androgens and subsequently, estrogens. The enzyme aromatase mediates the conversion of androgenic precursors to estrogens; individuals with aromatase deficiency generally exhibit sexual ambiguity at birth, virilization at puberty, and multicystic ovaries. The gonadotropin-resistant ovary syndrome results from genetic mutations in the FSH or LH receptor or post-receptor signaling defects that prevent the ovaries from responding normally to gonadotropin stimulation; although present, ovarian follicles fail to develop beyond the early antral stage and therefore produce little estrogen. Pituitary Disorders Pituitary tumors may cause amenorrhea by directly compressing pituitary gonadotrophs or distorting the portal venous network that delivers hypothalamic GnRH stimulation, resulting in decreased FSH and LH secretion. They also may cause inadequate or excessive production of other pituitary hormones, via compression of pituitary tissue, interference with the portal venous circulation, or autonomous secretion (functional pituitary adenomas). Directly or indirectly, these endocrinopathies may disrupt normal ovarian function, thereby causing amenorrhea. Although metastatic lesions may occasionally be seen, malignant pituitary tumors are extremely rare. Virtually all pituitary tumors are benign adenomas that may be functional or nonfunctional; functional tumors may secrete prolactin, growth hormone (GH), thyroid-stimulating hormone (TSH), or adrenocorticotropic hormone (ACTH). Other uncommon pituitary disorders that may cause amenorrhea include the empty sella syndrome and Sheehan syndrome. The empty sella syndrome results from herniation of the subarachnoid space containing cerebrospinal fluid into the sella turcica with compression of the pituitary gland against the sellar floor, giving the sella an “empty” appearance when viewed by computed tomography (CT) or magnetic resonance imaging (MRI). Sheehan syndrome results from acute infarction and necrosis of the pituitary gland as a rare complication of shock due to obstetric hemorrhage. Depending on the extent of pituitary damage, clinical consequences may be limited to disorders of reproductive function (failed lactation, amenorrhea), or more general, with multisystem failure due to panhypopituitarism. Hypothalamic Disorders Absent or abnormal patterns of pulsatile hypothalamic GnRH secretion that fail to stimulate normal levels or patterns of pituitary gonadotropin secretion are the most common cause of amenorrhea. Both PCOS and hypothalamic amenorrhea that may result from emotional, nutritional, or physical stress are basically hypothalamic disorders, but their pathophysiology and clinical presentations differ considerably. Women with PCOS exhibit an increased frequency of pulsatile GnRH secretion that results in increased LH synthesis, hyperandrogenism, and impaired follicular maturation. In contrast, the inconsistent and generally low frequency pattern of pulsatile GnRH secretion in women with hypothalamic amenorrhea results in low levels of pituitary gonadotropin release that fail to stimulate or sustain progressive follicular development. Occasionally, a hypothalamic tumor (craniopharyngioma, meningioma, hamartoma, chordoma) may distort the tuberoinfundibular tract or portal venous network, thereby interfering with effective delivery of GnRH stimulation, and result in decreased pituitary FSH and LH secretion. Alternatively, interference with hypothalamic dopamine delivery to pituitary lactotrophs may cause hyperprolactinemia. In either case, hypothalamic tumors may result in a secondary hypogonadotropic hypogonadism and amenorrhea. In other rare instances, GnRH deficiency is congenital and associated with midline craniofacial defects or with anosmia due to a failure of olfactory axonal and GnRH neuronal migration during embryogenesis (Kallmann syndrome), resulting in primary amenorrhea and sexual infantilism.

EVALUATION OF AMENORRHEA A detailed medical history and physical examination are always important. In the patient with amenorrhea, elements of particular interest include growth and secondary sexual development (breast and pubic hair), menstrual history (if any), previous surgery or trauma to the pelvis or central nervous system (CNS), family history of hereditary disorders, evidence of physical, psychological or emotional stress, symptoms and signs of hirsutism or galactorrhea, as well as reproductive tract anatomy. Medical History The age at which menarche should be expected varies but, in general, the first menses should occur within 2 to 3 years after the initiation of pubertal development. In most young girls (approximately 80%), the first sign of puberty is an acceleration of growth, followed by breast budding (thelarche), and the appearance of pubic hair (adrenarche). In the remainder, adrenarche precedes thelarche by a brief interval, but the two events typically are closely linked. Consequently, menarche should be expected as early as age 10 (when puberty begins at age 8), and rarely later than age 16 (when puberty begins at age 13). On average, in the United States, the mean ages for thelarche, adrenarche, and menarche in black girls are 6 to 12 months earlier than in white girls. When secondary sexual development fails to begin by age 14, or begins but fails to progress at the normally expected pace, evaluation is indicated. Once menstrual cycles have been established, amenorrhea for an interval equivalent to three previous cycles, or 6 months, warrants evaluation. Questions relating to past medical history, general health, and lifestyle may identify a severe or chronic illness (diabetes, renal failure, inflammatory bowel disease), head trauma, or evidence of physical, psychological, or emotional stress. Weight loss or gain and the frequency and intensity of exercise may be revealing. Headaches, seizures, vomiting, behavioral changes, or visual symptoms may suggest a CNS disorder. Vaginal dryness or hot flushes are evidence of estrogen deficiency and suggest ovarian failure. Progressive hirsutism or virilization is evidence of hyperandrogenism that may result from PCOS, nonclassic (late-onset) congenital adrenal hyperplasia (CAH), or an androgen-producing tumor of the ovary or adrenal gland. Bilateral galactorrhea suggests hyperprolactinemia. Cyclic pelvic or lower abdominal pain or urinary complaints may be caused by developmental anomalies resulting in obstructed menstrual flow, including an imperforate hymen, transverse vaginal septum, or cervical atresia. A previous inguinal hernia repair or curettage suggests the possibility of a developmental anomaly or damage to the reproductive tract. The timing and duration of any treatment with progestational agents (oral contraceptive pills [OCP], depot-medroxyprogesterone acetate), GnRH agonists (leuprolide, goserelin, nafarelin), or other medications (phenothiazines, reserpine derivatives, amphetamines, opiates, benzodiazepines, antidepressants, dopamine antagonists) or drugs (opiates) may provide important diagnostic clues.

Physical Examination Body habitus often provides important clinical information. Height, weight, and body mass index (BMI) should be recorded. Short stature (less than 60 inches) is a hallmark of gonadal dysgenesis; sexual infantilism, webbing of the neck, low set ears and posterior hairline, widely spaced nipples, short fourth metacarpal, and a wide carrying angle of the arms (cubitus valgus) are among the classical stigmata of Turner syndrome. Low body weight is frequently associated with hypothalamic amenorrhea resulting from poor nutrition (eating disorders) or physical, psychological, or emotional stress. Obesity or an increased waist-to-hip ratio (>0.85) is often associated with insulin resistance and chronic anovulation. Examination of the skin may reveal a soft, moist texture as seen in hyperthyroidism; a rapid pulse and classic eye signs (exophthalmos, lid lag), a fine tremor, and hyperreflexia may provide further evidence to suggest a diagnosis of Graves disease. Conversely, dry, thick skin, a slow pulse, diminished reflexes, and thinning of the hair suggest hypothyroidism. Both hypothyroidism and hyperthyroidism may be associated with amenorrhea. Orange discoloration of the skin in the absence of scleral icterus may result from hypercarotinemia associated with excessive ingestion of low-calorie, carotene-containing fruits and vegetables in dieting women. Acanthosis nigricans, velvety hyperpigmented skin most commonly observed at the nape of the neck, in the axillae, and beneath the breasts, strongly suggests severe insulin resistance and the possibility of diabetes. Acne and hirsutism are indications of hyperandrogenism that may result from chronic anovulation (PCOS), nonclassic CAH, or ingestion of androgenic anabolic steroids. When accompanied by any sign of frank virilization (deepening of the voice, frontotemporal balding, decrease in breast size, increased muscle mass, clitoromegaly), the possibility of ovarian hyperthecosis or an ovarian or adrenal neoplasm must be considered. Breast development, as assessed by Tanner staging, is a reliable indicator of estrogen production or exposure to exogenous estrogens. Arrested breast development suggests a disruption of the hypothalamic–pituitary–ovarian (HPO) axis. When menarche has not followed adult breast development, a developmental anomaly of the reproductive tract should also be considered. The breast examination should include gentle compression, beginning at the base and moving toward the nipple. Microscopic examination of any expressed cloudy or white nipple secretions that demonstrate lipid droplets indicate true galactorrhea and suggest hyperprolactinemia. Abdominal examination may rarely reveal a mass as may result from hematometra or an ovarian neoplasm. Growth of hair from the pubic symphysis to the infraumbilical region suggests hyperandrogenism. Abdominal striae raise the possibility of Cushing syndrome, but much more often result from progressive obesity or previous pregnancy. As noted earlier, thelarche and adrenarche typically are closely linked events during puberty and, in general, breast development and growth of pubic hair progress in a symmetric manner. The Tanner stages of breast and pubic hair development should be consistent. Absent or scant growth of pubic hair is a classic sign of AIS when breast development is asymmetrically advanced. Attempts at office examination of the vagina in sexually infantile girls or those with a small hymeneal ring are often difficult and counterproductive, but whenever feasible, speculum examination should be performed. A patent vagina and visible cervix excludes müllerian/vaginal agenesis, AIS, and most obstructive causes of amenorrhea. In those with an absent or infantile vaginal orifice, rectal examination should be performed and may reveal a distended hematocolpos above the obstruction when the uterus is present and functional. Diagnostic Evaluation A careful history and physical examination will always narrow the range of diagnostic possibilities that must then be differentiated. The subsequent laboratory investigation or imaging should be focused on that differential diagnosis and, with few exceptions a diagnosis can be quickly and easily established. Sexual ambiguity and virilization should be evaluated as separate disorders, mindful that amenorrhea is an important element of their presentation. Abnormal Genital Tract Anatomy A history of primary amenorrhea accompanied by physical examination that reveals an absent or blind vagina indicates a developmental anomaly of the genital outflow tract. The list of diagnostic possibilities is short and includes an imperforate hymen, a transverse vaginal septum, cervical atresia, müllerian/vaginal agenesis, and androgen insensitivity syndrome ( Fig. 34.1). Because each of these disorders has unique features, they generally are not difficult to distinguish.

FIG. 34.1. Congenital absence of the vagina. The external genitalia are entirely normal in appearance without any evidence of ambiguity but there is simply no vagina present. The normal pubic hair indicates androgen responsiveness and eliminates the possibility of androgen insensitivity syndrome indicating that this is simply congenital absence of the müllerian duct system. Patients with an imperforate hymen or transverse vaginal septum/cervical atresia typically present at the expected time of menarche with complaint of cyclic perineal, pelvic, or abdominal pressure or pain and exhibit normal secondary sexual development. In those with an imperforate hymen, genital examination reveals no obvious vaginal orifice, but a thin, often bulging, blue perineal membrane and a fluctuant mass, resulting from the accumulation of mucus and blood in the vagina. In those with a transverse vaginal septum or cervical atresia, examination typically reveals a normal vaginal orifice, a short blind vagina of varying length, and a fluctuant pelvic mass well above to the level of obstruction (hematocolpos, hematometra, hematosalpinx). Differentiation of imperforate hymen and transverse vaginal septum/cervical atresia generally requires no laboratory investigation. Whereas a transabdominal or transperineal ultrasound examination will reveal the level and volume of sequestered menses, an abdominalpelvic MRI provides greater anatomic detail and helps to define the nature of an anomaly ( Fig. 34.2). In some instances, laparoscopy may be required to clearly identify the anatomy of a developmental anomaly.

FIG. 34.2. Congenital absence of the lower one third of the vagina. Magnetic resonance imaging demonstrates that a uterus is present ( A) as well as an upper vaginal pouch ( B). The upper vagina has formed a hematocolpos which began soon after menarche and presented as a pelvic mass. In contrast, patients with amenorrhea resulting from müllerian agenesis generally are otherwise entirely asymptomatic. They exhibit normal breast and pubic hair development, an absent vagina, and have no symptoms or signs of cryptomenorrhea because the uterus is altogether absent. The diagnosis is usually self-evident from physical examination alone. Further evaluation to exclude skeletal and urinary tract anomalies is indicated because approximately 12% to 15% of women with müllerian agenesis have skeletal abnormalities (vertebral anomalies are most common) and one third or more have urinary tract anomalies (ectopic kidney, renal agenesis, horseshoe kidney, abnormal collecting system). In girls who have not yet reached the age when menarche (and cryptomenorrhea) would be expected if a uterus was present, imaging must be interpreted cautiously because even abdominal/pelvic MRI can be misleading when the reproductive organs are immature. Remaining alert to the diagnostic possibilities, careful observation over time is preferable to invasive investigations that are otherwise unnecessary. Normal breast development, absent or sparse growth of pubic hair, and a short blind vagina clearly suggest AIS. Although the chromosomal sex is male (46,XY), the phenotype is female. The testes are undescended, often palpable in the inguinal canals (most commonly at the level of the external inguinal ring), and produce normal male levels of testosterone and müllerian inhibitory hormone (MIH). Whereas end-organ insensitivity to androgen action (due to abnormalities of the androgen receptor) prevents normal masculinization of the genitalia, MIH inhibits müllerian development in a normal fashion. Consequently, the external genitalia are those of a female (absent androgen action), the uterus is absent (normal MIH action), and the vagina is short and ends blindly (derived from the invaginating urogenital sinus, absent androgen action). The diagnosis may be suspected when other family members (e.g., aunt, sister) are affected with this X-linked disorder. Incomplete penetrance may result in growth of more pubic hair than might be expected and sometimes can be misleading. However, a serum testosterone concentration easily distinguishes androgen insensitivity syndrome (AIS; normal or modestly elevated above the range observed in normal males) from müllerian agenesis (normal range for a female). A karyotype firmly establishes the diagnosis. Cervical obstruction/stenosis and Asherman syndrome are abnormalities of genital tract anatomy, but physical examination of the genital tract most often is normal. When cervical stenosis causes symptoms, worsening dysmenorrhea or prolonged light staining or spotting after menses are the most common complaints; amenorrhea is a rare occurrence. In women with a history of previous conization or other cervical surgery or ablative therapy, uterine sounding may help to establish a diagnosis. Similarly, most women in whom previous infection or surgical trauma has resulted in intrauterine synechiae present with dysmenorrhea, hypomenorrhea, subfertility, or recurrent early pregnancy loss, rather than amenorrhea. In women whose history clearly suggests the possibility of intrauterine adhesions, ultrasound and hysterosalpingography can reveal their location and extent, but hysteroscopy is the definitive method for diagnosis. Normal Genital Tract Anatomy When physical examination reveals normal genital tract anatomy, further evaluation is required to determine the cause of amenorrhea. The possibility of pregnancy should always be considered and excluded. When breast growth is absent or inconsistent with age and associated with primary amenorrhea, the cause of delayed puberty should be determined. The vast majority of these cases have no pathology. In the remainder, evaluation may reveal thyroid disease, chronic illness (malabsorption, renal disease, eating disorders, inflammatory bowel disease), ovarian failure (e.g., gonadal dysgenesis), a pituitary disorder (tumor, empty sella syndrome, hyperprolactinemia), or a hypothalamic cause (Kallmann syndrome; physical, emotional, or psychological stress; tumor). Wrist x-rays for bone age and a GnRH stimulation test are important components of the evaluation for delayed puberty in children and adolescents, but the diagnostic possibilities and scope of evaluation in adolescents and adults with either primary or secondary amenorrhea are otherwise very much the same. Thyroid Function Tests Initial evaluation should include a measurement of serum TSH. The newest generation of ultrasensitive TSH assays in common use provides the means to detect both primary hypothyroidism (elevated TSH) and primary hyperthyroidism (low TSH). Either may result in chronic anovulation and amenorrhea. Any abnormal value should be confirmed and accompanied by measurement of serum thyroxine (tetraiodothyronine; T4) to better define the nature and extent of the

thyroid disorder. When TSH is elevated and the T4 concentration is normal, the diagnosis is subclinical hypothyroidism, best viewed as a compensated state wherein normal levels of T4 are maintained, but only under increased levels of pituitary stimulation. On rare occasions, both TSH and T4 levels may be low, suggesting hypothyroidism of pituitary origin that will require additional evaluation to include hypothalamic/pituitary imaging (MRI) and careful assessment to determine whether other pituitary functions also are affected. Prolactin A serum prolactin determination is another component of the initial evaluation of amenorrhea. Hyperprolactinemia may be associated with a variety of menstrual disturbances—oligomenorrhea and amenorrhea being the most common. In general, prolactin concentrations are higher in amenorrheic than in oligomenorrheic hyperprolactinemic women. Hyperprolactinemia may occasionally result in delayed puberty and primary amenorrhea, if it arises before menarche, and is among the most common causes of secondary amenorrhea. Hyperprolactinemia inhibits pulsatile hypothalamic GnRH secretion, resulting in depressed levels of pituitary FSH and LH secretion. The end result is anovulation or more profound hypogonadotropic hypogonadism, depending on the extent to which gonadotropin secretion is suppressed. One cannot rely on the symptom or finding of galactorrhea to identify individuals whose amenorrhea may result from hyperprolactinemia. Only approximately one third of hyperprolactinemic women will exhibit galactorrhea, probably because breast milk production requires several other hormones, including GH, T4, cortisol, insulin, and most important, estrogen and progesterone. Serum prolactin determinations therefore should be obtained in all amenorrheic women. Hyperprolactinemia has many causes. Among these are: prolactin-secreting pituitary adenomas other pituitary or hypothalamic tumors that may distort the portal circulation and thereby prevent effective delivery of hypothalamic dopamine (the putative prolactin inhibitory factor or hormone) a variety of drugs that lower dopamine levels or inhibit dopamine action (amphetamines, benzodiazepines, butyrophenones, metoclopramide, methyldopa, opiates, phenothiazines, reserpine, and tricyclic antidepressants) breast or chest wall surgery, cervical spine lesions, or herpes zoster (activation of the afferent sensory neural pathway that stimulates prolactin secretion, in a manner similar to suckling) hypothyroidism (increased hypothalamic thyrotropin-releasing hormone stimulates pituitary prolactin secretion directly) pharmacologic estrogens (OCP) other rare, nonpituitary sources (lung and renal tumors) or causes of decreased prolactin clearance (renal failure). All causes must be considered and systematically excluded; a careful history will eliminate many of the possibilities. When medications are the cause, prolactin concentrations are usually only moderately elevated and levels greater than 100 ng/mL generally are uncommon. Although medications may offer an obvious explanation, one cannot confidently assume they are the cause of hyperprolactinemia. If possible, a trial discontinuation or use of an alternative medication should be considered. When that is not possible and hypothyroidism has been excluded, further evaluation to exclude hypothalamic and pituitary tumors and other causes of hyperprolactinemia is appropriate. Imaging of the hypothalamic and pituitary regions to exclude mass lesions in hyperprolactinemic patients can be accomplished with head CT or MRI. MRI generally is regarded as the superior method because it is more accurate for identification of very small lesions or an empty sella and better defines tumor margins and relationships to surrounding structures. The indications for CT or MRI in the evaluation of hyperprolactinemia remain controversial. Those who advocate liberal use of imaging correctly emphasize that the likelihood of a pituitary tumor does not correlate with the prolactin concentration. Pituitary microadenomas (=10 mm) are very common (10%–30% prevalence in autopsy studies) and even macroadenomas (>10 mm) may be associated with only modest elevations (25–100 ng/mL) of prolactin because they are nonfunctional tumors or may have undergone necrosis. Moreover, imaging may reveal evidence of other hypothalamic disease that may be important (tumor, tuberculosis, sarcoidosis, aqueductal stenosis) and amenable to specific treatment. Those who prefer a more selective approach stress that imaging is costly and has a relatively low yield when performed routinely. They correctly emphasize that pituitary tumors rarely grow (even in pregnancy), are not a contraindication to hormone replacement or OCP, and that their natural course is unaffected by treatment with dopamine agonists (bromocriptine, pergolide, cabergoline). In this view, because diagnosis of a pituitary microadenoma generally has little or no impact on clinical management decisions, MRI should be limited to those with grossly elevated prolactin levels (>100 ng/mL) who are hypoestrogenic, in whom a macroadenoma of greater clinical significance is more likely, and to those with suspicious symptoms (visual disturbances, headaches) or findings (visual field defects, abnormal optic fundi). For asymptomatic patients with moderate hyperprolactinemia (20–100 ng/mL), some advocate a less costly coned down lateral view of the sella turcica and reserve MRI for those with an enlarged or abnormal sella (erosion of the sellar floor or clinoid processes) or hypothalamic calcifications that suggest a tumor (e.g., craniopharyngioma). The most prudent approach is to obtain an MRI whenever persistent hyperprolactinemia cannot be confidently attributed to medication or hypothyroidism. FSH Initial evaluation of amenorrheic women with normal genital tract anatomy should include measurement of serum FSH to distinguish ovarian failure (elevated FSH) from hypothalamic/pituitary disease or dysfunction that yields inadequate or ineffective patterns of gonadotropin secretion (low or normal FSH). An elevated FSH level generally is a reliable indicator of ovarian failure, but must be interpreted in the context of the clinical presentation. During the perimenopause, regardless whether it occurs prematurely or at the usual age, FSH levels may rise well before menses have ceased entirely. Those follicles that remain also are relatively insensitive to FSH, but many can and will respond to rising levels of stimulation when the requisite threshold FSH concentration is achieved. Once follicular growth begins and estrogen levels rise, FSH concentrations decline, albeit transiently, before rising again to the levels necessary to stimulate new follicular growth. FSH levels therefore are dynamic, often fluctuate widely during the perimenopause, and must be interpreted cautiously. The FSH level is high in the enigmatic gonadotropin-resistant ovary syndrome that may result from inactivating mutations in the FSH or LH receptor. However, specific efforts to diagnose these rare conditions (ovarian biopsy, genotyping) are academic and have no practical clinical value because the prognosis for future fertility is extremely poor and treatment options are no different from those for women with true ovarian failure. Serum FSH concentrations also are elevated in individuals with 17a-hydroxylase or aromatase deficiencies and galactosemia, but these conditions are extremely rare and do not enter into clinical consideration. With few exceptions, a high serum FSH level is an indication of ovarian failure. A history of previous radiation or chemotherapy may provide an obvious explanation. Doses of radiation under approximately 100 rads generally have no significant effect, but risk of ovarian damage rises progressively with higher doses. Individuals treated when young may have only transient amenorrhea with a return of menstrual cycles months or years later, but also are more likely to develop POF. Those treated as adults are at greater risk for immediate and irreversible ovarian failure. Alkylating agents (e.g., cyclophosphamide), used in the treatment of malignancies and other diseases (systemic lupus erythematosus), are extremely toxic to gonadal tissues. As with radiation, the dose required to induce ovarian failure is inversely related to age at the time of treatment. Other chemotherapeutic agents have the potential for ovarian damage, but their effects are less clear; risk increases with the number of agents involved in combination therapies. When ovarian failure occurs before age 30 and cannot be confidently explained, a karyotype should be obtained. An abnormal karyotype may be observed in up to one-half of all women with primary amenorrhea. Classic Turner syndrome (45,X), structural abnormalities of the X chromosome (deletion, ring, iso-chromosome), and mosaicism (e.g., 45,X/46,XX) are the most common abnormalities found. A karyotype also will detect the presence of a Y chromosome that might not otherwise be suspected. The phenotype in Swyer syndrome (46,XY gonadal dysgenesis) and some with Turner mosaicism (45,X/46,XY) is female because the dysgenetic (streak) gonads fail to produce both MIH and androgens. Consequently, the uterus, fallopian tubes, cervix, and vagina develop normally but the genitalia do not masculinize. However, a peripheral leukocyte karyotype alone cannot exclude the presence of occult Y chromosomal material. Further analysis with fluorescence in situ hybridization (FISH) using one or more probes that are specific for segments of the Y chromosome is required and should be performed in any individual with a 45,X karyotype or 45,X mosaic cell line. An occult Y chromosome must be identified because affected individuals are at significant risk (approximately 25%) for developing a unique type of germ cell tumor (gonadoblastoma) that may contain malignant elements (dysgerminoma, embryonal cell carcinoma, choriocarcinoma). Virtually all such tumors arise early in life. Over the age of 30, therefore, karyotype is unnecessary and ovarian failure can be confidently regarded as premature menopause. Approximately 25% of patients with gonadal dysgenesis have a normal karyotype (46,XX). As gonadal dysgenesis with a normal karyotype is associated with neurosensory deafness, audiometry should be considered. In women with secondary amenorrhea and unexplained POF, further evaluation to exclude autoimmune disease is appropriate as up to 40% may have autoimmune disorders. POF develops in 10% to 60% of women with Addison disease (adrenal insufficiency) and also is more common in women with diabetes mellitus (type 1), myasthenia gravis, and parathyroid disease than in healthy women. However, only those with Addison disease are likely to have a demonstrable autoimmune lymphocytic oophoritis. Ovarian biopsy is not indicated in clinical practice, but because POF may be a component of a polyglandular syndrome, general screening for autoimmune disorders is reasonable. Thyroid abnormalities are the most common and can be identified by measuring TSH, T4, and thyroid autoantibodies (antiperoxidase, antithyroglobulin). Screening may also include 24-hour urinary free cortisol, fasting blood glucose, serum calcium and phosphorus, and antinuclear antibody. More extensive or specific testing for autoimmune disorders is unnecessary in the absence of other clinical signs and symptoms of disease. Although autoimmune screening is logical and practical, no one or combination of serum markers can confirm a diagnosis of autoimmune ovarian failure. A low normal serum FSH concentration is an indication of hypothalamic or pituitary dysfunction and the most common result observed in clinical practice. PCOS and hypothalamic amenorrhea are the two main diagnostic possibilities and generally are easily distinguished by their clinical presentations.

POLYCYSTIC OVARY SYNDROME Although there is no universally accepted definition of PCOS, diagnosis generally is based on three criteria—ovulatory dysfunction, clinical evidence of hyperandrogenism (hirsutism, acne, androgenic alopecia) or hyperandrogenemia, and exclusion of other disorders (hyperprolactinemia, thyroid abnormalities, nonclassic CAH). Women with PCOS more commonly exhibit oligomenorrhea (75%) than amenorrhea (25%); irregular and infrequent menses typically begin soon after menarche, but may emerge later, often in association with progressive weight gain. Signs of androgen excess generally do not become evident until years later and progress gradually. Transvaginal ultrasound examination typically reveals ovaries that are modestly enlarged and contain numerous small follicles aligned in the periphery (“string of pearls”), but such findings are not useful for diagnosis because in up to one-third of normal women between the ages of 18 and 25 years, the ovaries have a similar polycystic appearance. Women with PCOS are frequently insulin-resistant (insulin sensitivity is reduced by 30%–40%), exhibit compensatory hyperinsulinemia (up to 80%), and are predisposed to glucose intolerance. Approximately 30% have demonstrable impaired glucose intolerance by glucose tolerance testing; fasting glucose is elevated in less than 10%. A fasting glucose/insulin ratio less than 4.5 is a fairly specific but somewhat insensitive diagnostic criterion for insulin resistance. Obesity is a common feature of women with PCOS (50%–75%) and exacerbates insulin resistance and the hyperinsulinemia that is associated with elevated androgen levels. Prolactin levels are mildly elevated in 10% to 25% of women with PCOS. Although the ratio of serum LH/FSH is frequently increased (>2.0), measurement of the serum LH level generally is not useful or necessary. The diagnosis of PCOS is not based on findings of ovarian or hormonal abnormalities, but on a history of chronic anovulation and clinical findings of androgen excess or obesity.

When hirsutism is severe or perimenarchial in onset, the possibility of nonclassic CAH also must be considered. Most commonly, nonclassic CAH results from a deficiency of the enzyme 21-hydroxylase which mediates an essential step in cortisol synthesis. Affected individuals cannot efficiently convert 17-hydroxyprogesterone (17-OHP) to 11-deoxycorticosterone (DOC; an intermediate step in cortisol synthesis). A follicular phase 17-OHP level greater than 2 ng/mL merits further evaluation with an ACTH stimulation test (serum 17-OHP before and 30–60 minutes after intravenous injection of 250 g ACTH) to confirm the diagnosis (post-stimulation serum 17-OHP concentration >1,000 ng/dL). Determination of Estrogen Status Evaluation of estrogen levels would seem logical for differentiating PCOS from hypothalamic amenorrhea and other hypoestrogenic disorders. Unfortunately, available methods cannot easily and reliably define the level of ovarian estrogen production. One cannot rely on symptoms and signs of estrogen deficiency to identify hypogonadal women. Genitourinary atrophy develops only gradually and is uncommonly observed in young women, even when estrogen levels are extremely low, and vasomotor symptoms typically are absent in women with hypothalamic dysfunction. Other methods for assessing the level of ovarian estrogen production include immunoassay of the serum estrogen concentration and “bioassays” based on clinical observation of the amount and character of cervical mucus (“estrogenic” mucus being clear, watery, and relatively abundant) or results of a “progestin challenge test” (presence or absence of withdrawal bleeding after administration of an exogenous progestin). Each of these methods may be useful, but each clearly also has pitfalls. A serum estradiol measurement is easy to perform and relatively inexpensive. One might reasonably expect low estrogen levels in women with hypothalamic amenorrhea and normal levels in women with PCOS. Unfortunately, estradiol concentrations fluctuate erratically and may be normal or low on any given day, and therefore can be misleading. Whereas observations of estrogenic cervical mucus clearly suggest a normal level of ovarian estrogen production, the absence of such findings cannot be confidently interpreted because many women exhibit such mucus only in the late follicular phase of the cycle when estrogen levels are relatively high, or not at all. The progestin challenge is based on the observation that progestin treatment (e.g., medroxyprogesterone acetate 10 mg daily for 5–7 days or progesterone in oil 100 mg i.m.) will induce menses only in those with normal circulating estrogen concentrations. For this purpose, a pure progestational agent must be used; endogenous estrogen status cannot be inferred from the response to an OCP that contains both estrogen and progestin. A positive test (bleeding after completion of progestin treatment) implies normal levels of estrogen production and a negative test (no withdrawal menses) suggests frank hypogonadism. However, withdrawal bleeding correlates poorly with estrogen status; both false-positive (withdrawal bleeding despite generally low levels of estrogen production) and false-negative (absent bleeding despite significant estrogen production) results are common. Up to 20% of women with oligomenorrhea or amenorrhea in whom substantial estrogen is present do not exhibit withdrawal bleeding. Conversely, up to 40% of women whose amenorrhea relates to stress, exercise, weight loss, or hyperprolactinemia, in whom estrogen levels are generally low, exhibit withdrawal bleeding. A false-positive progestin challenge also has been frequently observed in women with POF. Hypothalamic Amenorrhea In the absence of obesity or evidence of hyperandrogenism characteristic of PCOS, the most likely cause of amenorrhea in women with a normal or low serum FSH level is a functional disorder of the hypothalamus or higher CNS centers. Women with such “hypothalamic amenorrhea” generally present with secondary amenorrhea that is frequently accompanied by history of emotional stress, weight loss (dieting), poor nutrition (eating disorders, chronic illness), or regular strenuous exercise (endurance training). In contrast to women with PCOS, they typically have normal or low body weight and are poorly estrogenized. In the context of low levels of estrogen production, a low or normal serum FSH has the same implication—a dysfunctional HPO axis—because if that axis were intact and fully functional, the classic negative feedback relationship between estrogen and FSH would stimulate a compensatory increase in FSH secretion and result in an elevated serum FSH concentration. A low or normal serum FSH therefore clearly suggests hypothalamic or pituitary disease or dysfunction. Most women with hypothalamic amenorrhea have a normal FSH concentration; extremely low or undetectable FSH levels are seldom seen except in women with large pituitary tumors or anorexia nervosa. Amenorrhea associated with weight loss due to dieting is common; anorexia nervosa is fortunately much less common (15/100,000 women/year). In athletic women, the risk of amenorrhea is increased approximately three-fold over that in nonathletic women with the highest prevalence observed in endurance athletes (long-distance running). Chronic debilitating diseases (end-stage renal disease, malignancy, acquired immune deficiency syndrome, malabsorption) also may result in anovulation and amenorrhea. In such cases, hypothalamic amenorrhea represents a functional suppression of the reproductive system that may be viewed as a psychobiologic response to psychological, physical, or nutritional stress. One proposed unifying hypothesis emphasizes the concept of energy balance: When available energy is excessively diverted (exercise), or insufficient (dieting, malnutrition), reproduction is suspended in order to support essential metabolism for survival. The mechanism responsible may involve a stress-induced increase in hypothalamic corticotropin-releasing hormone (CRH) and endogenous opioid secretion that inhibits pituitary gonadotropin release directly, or inhibition of pulsatile hypothalamic GnRH secretion by increased dopamine or opioids. Hypothalamic–Pituitary Imaging The diagnosis of hypothalamic amenorrhea cannot be established until organic disease of the CNS, hypothalamus, or pituitary gland is excluded with a CNS MRI. Imaging is prudent, even when emotional stress, weight loss, poor nutrition, or regular strenuous exercise appear to offer an explanation for hypothalamic dysfunction and amenorrhea. Imaging may reveal a pituitary tumor, an empty sella, or evidence of a hypothalamic tumor, anomaly, or other disease. The vast majority of pituitary tumors are prolactinomas or nonfunctioning adenomas, but other varieties are rarely encountered. Additional evaluation can help to define the nature of a tumor and the extent to which other pituitary functions may be compromised. Grossly elevated prolactin levels clearly suggest a prolactinoma; more modest prolactin elevations may be observed when large tumors infarct or a nonfunctioning adenoma distorts sellar anatomy and disrupts normal dopamine delivery to pituitary lactotrophs. When the clinical presentation suggests Cushing syndrome, screening (24-hour urinary free cortisol, overnight dexamethasone suppression test) is indicated; additional evaluation (serum ACTH, adrenal CT or MRI) is required when results are abnormal. Physical findings that suggest acromegaly are an indication to measure serum insulin-like growth factor-1 (IGF-1; somatomedin-C); if elevated, an oral glucose tolerance test with GH levels should be performed (lack of suppression is diagnostic). Pituitary function testing (TSH, T4, prolactin, 24-hour urinary free cortisol, IGF-1) also is indicated when imaging reveals a macroadenoma or an empty sella to insure that the other trophic pituitary hormones are normal. Dynamic testing is unnecessary in otherwise asymptomatic women with a microadenoma or normal sella. The empty sella syndrome generally is a benign condition and not progressive. However, because of the possibility of an unrecognized coexisting tumor, periodic surveillance with a prolactin determination and MRI are indicated. Anatomic abnormalities of the hypothalamus are distinctly uncommon. Hypothalamic tumors (craniopharyngioma, hamartoma, meningioma) generally are rare and other mass lesions (tuberculosis, sarcoidosis) are even more so. Congenital hypogonadotropic hypogonadism associated with anosmia or hyposmia (Kallmann syndrome) is another rare inheritable disorder associated with a specific anatomic defect: Hypoplastic or absent olfactory sulci. The condition results from a failure of both olfactory axonal and GnRH neuronal migration during development and is genetic in origin (X-linked, autosomal dominant, or autosomal recessive). The most common form (X-linked) derives from mutations in a single gene ( KAL) on the short arm of the X chromosome that encodes a protein (anosmin-1) necessary for normal neuronal migration. The clinical presentation (primary amenorrhea, sexual infantilism, hypogonadotropic hypogonadism, and anosmia) and common association with other anatomic (cleft lip and palate) and neurologic (hearing loss, cerebellar ataxia, color blindness) abnormalities is not difficult to recognize. Very often, imaging fails to reveal any anatomic abnormality of the pituitary or hypothalamus and in otherwise asymptomatic individuals, the diagnosis is, by exclusion, hypothalamic dysfunction.

TREATMENT OF AMENORRHEA Treatment of amenorrhea is obviously focused on the specific etiology, when known, and always is tailored to the goals of the patient. Genital Tract Abnormalities In women with vaginal/müllerian agenesis, the primary goal of treatment—creation of a functional vagina—can be accomplished with a variety of methods when the time is appropriate. In most cases, progressive vaginal dilation will be successful in the motivated patient. The technique involves application of pressure to the point of moderate discomfort (approximately 20–30 min/d) using commercially available vaginal dilators, first in a posterior direction (to create a pouch), and then in the usual line of the vaginal axis (after about 2 weeks). After the desired depth is achieved, increasingly larger dilators will expand the vaginal diameter and create a functional vagina in approximately 3 to 6 months. Operative treatment of women with vaginal/müllerian agenesis generally should be reserved for those who refuse or poorly tolerate vaginal dilation. Traditionally, a neovagina has been created by dissection of the rectovaginal space and placement of a skin graft, held in place with a soft mold until the graft was established (McIndoe procedure). Subsequent regular intercourse or vaginal dilation must be maintained to avoid risk of fibrosis and loss of function. More recently, an operation that employs a transabdominal traction device has been described that can create a functional vagina within 7 to 9 days. Where available and applicable, the technique would appear to offer significant advantages over the traditional vaginoplasty procedure. In the rare woman with vaginal/cervical agenesis with a well-formed uterine body, it may be technically possible to create a neovagina continuous with the uterus with the goal of preserving fertility, but the associated long-term infectious morbidity has led most to conclude that the uterus should be removed in most circumstances. Reassurance and support are important elements of the management of vaginal/müllerian agenesis. Affected women should be counseled that although they are infertile, normal sexual function can be expected and that genetic offspring can be achieved by in vitro fertilization using oocytes retrieved from the normal ovaries and the sperm of the partner, with subsequent transfer of embryos to a gestational surrogate. A growing experience with these techniques has revealed no evidence to

indicate that congenital absence of the vagina and uterus is heritable in a dominant fashion. AIS presents both similar and different treatment challenges. A functional vagina again can be created by any of the techniques described previously for women with vaginal/müllerian agenesis, but surgical treatment much less often is required as the vagina generally is short but otherwise normal rather than altogether absent. Genetic offspring, of course, are not possible because women with AIS do not have oocytes. They have testes, most commonly located in the inguinal canals at the level of the external inguinal ring but sometimes found within the abdomen. Like women with gonadal dysgenesis whose karyotype contains a Y chromosome (46,XY, 45,X/46,XY), individuals with AIS are at increased risk for development of tumors in their undescended testes. Therefore, the testes should be removed. However, the risk of neoplasia is lower (approximately 5%–10% vs. 25%), tumors are rarely encountered before puberty, and the secondary sexual development that accompanies puberty in women with AIS from the aromatization of testosterone is more natural than can be accomplished pharmacologically. Consequently, removal of the testes and hormone replacement are best delayed until pubertal maturation is complete, generally by age 16 to 18 years. Initial treatment of women with an imperforate hymen, transverse vaginal septum, or cervical atresia centers on relief of symptoms related to accumulated menstrual fluid and debris. Surgical correction of imperforate hymen is straightforward, requiring only a cruciate incision in the hymen to the base of the hymeneal ring and excision of its central portion to allow drainage of sequestered menstrual fluid and subsequent normal menstruation. Surgical management of a transverse vaginal septum can be very challenging. In its simplest form, the procedure involves excision of the septum or dissection through the atretic segment to connect the margins of the lower and upper vaginal canals with split-thickness skin grafting occasionally required if the distance between the two precludes a primary junction without tension. In rare women with cervical atresia, as in those with vaginal agenesis who have a functional uterine body, the generally poor outcome and high postoperative morbidity associated with heroic efforts to preserve the uterus and fertility suggest that removal of the müllerian remnants is the safest management strategy. Uterine sounding and gentle cervical dilation often are all that is required to correct a symptomatic cervical obstruction/stenosis. Operative hysteroscopy is the preferred method for treatment of intrauterine synechiae that may be lysed by blunt dissection, scissors, electrodissection, or with a laser; any of these methods achieves results superior to blind curettage. When adhesions are severe or extensive, an intrauterine balloon catheter (left in place for approximately 7–10 days) may be useful to maintain separation between the walls of the uterine cavity. Treatment with a broad-spectrum antibiotic and an inhibitor of prostaglandin synthesis (nonsteroidal antiinflammatory drugs) helps to minimize risk of infection and uterine cramping. High-dose exogenous estrogen treatment (2.5 mg conjugated equine estrogens/day or its equivalent for approximately 4 weeks) also generally is recommended to encourage rapid endometrial reepithelialization and proliferation. Repeated procedures may be required to restore a normal uterine cavity. Approximately 50% to 75% of affected women may be expected to achieve a successful pregnancy, although risks of preterm labor, placenta accreta, placenta previa, and postpartum hemorrhage are increased. Ovarian Disorders Women with gonadal dysgenesis and sexual infantilism should be offered growth and sex hormone replacement therapy to promote growth to maximum potential, normal bone density development, and secondary sexual maturation. Treatment with exogenous recombinant human GH (50g/kg/d) stimulates an acceleration of growth that generally can be sustained for 6 years or more and results in an adult height as much as 10 cm over initial predicted height. In those with short stature, sex hormone replacement should be postponed until the bone age reaches 12 or greater to avoid premature epiphyseal closure and allow a longer interval of time for long bone growth. Sex hormone replacement therapy should follow the normal sequence of sex hormone production observed during adolescence and begin with low doses of estrogen alone (0.3 mg conjugated equine estrogens or 0.5 mg micronized estradiol). Dosage should be increased gradually and after approximately 6 to 12 months or first evidence of vaginal bleeding, cyclic treatment with a progestin should be added. To achieve maximum bone and breast development, higher doses (1.25–2.5 mg conjugated equine estrogens or 2–3 mg micronized estradiol) often are required. Once secondary sexual development is completed, longer-term sex hormone replacement should continue to avoid the otherwise inevitable emergence of estrogen deficiency symptoms and bone mineral depletion. Alternatively, a low-dose OCP may be offered, for convenience. Pregnancy may be achieved in women with gonadal dysgenesis through in vitro fertilization using oocytes provided by a genetically normal oocyte donor. However, pregnancy carries a unique risk for women with 45,X gonadal dysgenesis (Turner syndrome) as numerous cases of aortic dissection, aneurysm, and spontaneous rupture during pregnancy now have been reported. Preconceptional echocardiography may identify patients at risk for whom pregnancy is contraindicated, but catastrophic events may occur despite the absence of abnormal findings. Should they conceive spontaneously or by using donor oocytes, echocardiography should be performed at least once in each trimester, and the practitioner should remain alert to any evidence of progressive dilation of the ascending aorta. Women with 45,XY or 45,X/46,XY gonadal dysgenesis should have their gonads removed soon after diagnosis to avoid the substantial risk of gonadal tumors. Unlike those with AIS whose functional testes produce sex steroids that will promote normal secondary sexual maturation, women with gonadal dysgenesis have nonfunctional streak gonads. Consequently, there is no advantage to delay their removal. The primary goals of treatment for women with POF that develops after secondary sexual maturation is complete generally are relief from symptoms of estrogen deficiency and prevention of premature bone mineral depletion. Standard regimens of cyclic or combined continuous estrogen/progestin hormone replacement therapy or a low-dose OCP will meet the need, and either is appropriate. In vitro fertilization with use of donor oocytes offers the possibility of pregnancy and the prognosis for success in most such women is excellent. Women in whom POF is diagnosed before age 30 and in whom karyotype reveals a mosaic 45,X cell line must be carefully evaluated and counseled on the potential risks of aortic dissection and rupture during pregnancy, as described earlier. Monitoring or treatment should be offered for any associated autoimmune disorders. Pituitary Disorders The overwhelming majority of pituitary tumors are prolactin-secreting adenomas or nonfunctioning tumors. TSH-, ACTH-, or GH-secreting adenomas generally are rare. A pituitary tumor often will be discovered when imaging is performed in women with hyperprolactinemia and occasionally in euprolactinemic women with hypogonadotropic hypogonadism that cannot be confidently attributed to weight loss, eating disorders, or regular strenuous exercise ( Fig. 34.3). It may be difficult to differentiate a prolactin-secreting adenoma from a nonfunctioning tumor that disrupts normal hypothalamic dopamine delivery, except perhaps when prolactin levels are markedly elevated. However, because treatment options in hyperprolactinemic amenorrheic women with small intrasellar tumors are the same in either case, a specific diagnosis is not critical. The nature of a tumor may become clear only after treatment with a dopamine agonist restores normal prolactin levels; the majority of prolactin-secreting adenomas shrink in size during treatment and those that do not are more likely to be nonfunctioning tumors. In euprolactinemic amenorrheic women, a pituitary tumor also may be only an incidental finding, having no clinical significance.

FIG. 34.3. A pituitary tumor. Magnetic resonance imaging of a prolactin-secreting pituitary tumor ( three large arrows). Note the lateral location and elevation of the diaphragm sella. However, the tumor does not compress the optic chiasm ( single arrow) and thus no visual disturbances would be anticipated. Whereas transsphenoidal surgery was commonly performed in the past to remove pituitary tumors, it is now generally reserved for women with rare TSH-, ACTH-, or GH-secreting adenomas, those with large nonfunctioning tumors associated with complaints of headache or visual disturbances, and those in whom medical treatment (described subsequently) fails or is poorly tolerated. Surgery achieves immediate reduction of prolactin levels and restores cyclic menses in approximately 30% of women with prolactin-secreting macroadenomas and in up to 70% with microadenomas. However, residual or recurrent tumors and hyperprolactinemia are common and surgery may be complicated by cerebrospinal fluid leakage, meningitis, diabetes insipidus, or other trophic pituitary hormone deficiencies that require further treatment. Postoperative monitoring involves periodic serum prolactin determinations and repeated imaging. Radiation therapy for pituitary tumors is even less attractive because the response to radiation is slow and such treatment risks development of panhypopituitarism over time. As a result, radiation generally is reserved for postoperative recurrence of large tumors. Dopamine-agonist therapy is the mainstay of treatment for hyperprolactinemia, independent of whether a pituitary tumor is identified. Bromocriptine is highly efficacious but side effects (nausea, headache, orthostatic hypotension, dizziness, nasal congestion) are common. The dose should be titrated incrementally to normalize serum prolactin levels, beginning with a small dose (1.25–2.5 mg) at bedtime, with a second dose taken with breakfast or lunch when required; most patients will require 5.0 mg per day or less. Cabergoline may be an effective alternative when bromocriptine treatment cannot be tolerated, although its lower frequency of side effects and less frequent dosing (0.25–3.0 mg every 3–7 days) also make it an attractive initial therapeutic choice. Vaginal administration of bromocriptine or cabergoline is effective and may help to reduce side effects when oral treatment is poorly tolerated. Dopamine-agonist treatment restores menses and ovulatory function in up to 80% of amenorrheic hyperprolactinemic women, generally within approximately 6 to 8 weeks; reduction or cessation of galactorrhea typically requires somewhat longer-term treatment. Medical treatment with a dopamine agonist (bromocriptine, cabergoline) promotes shrinkage of pituitary macroadenomas in the majority of cases, usually within 6 to 12 weeks. Large tumors that fail to shrink despite effective suppression of excess prolactin secretion generally are nonfunctioning adenomas that cause hyperprolactinemia by interfering with delivery of dopamine from the hypothalamus and require surgery. Regardless of the treatment they receive, women with macroadenomas should be monitored with serum prolactin determinations and imaging every 6 to 12 months for at least 2 years, less often thereafter if tumor size and prolactin levels remain stable, and sooner if prolactin levels rise significantly or symptoms of headache or visual disturbances emerge or recur. Similar monitoring with serial prolactin determinations (every 6 to 12 months) and periodic imaging (annually for 2 years) is

recommended for those with microadenomas. Dopamine-agonist therapy is the best initial choice for persistent or recurrent tumor or hyperprolactinemia. For the few patients who may require radiation therapy, ongoing surveillance must be implemented to detect any evidence of developing panhypopituitarism. Pituitary tumors generally grow slowly or not at all, even during pregnancy. No more than approximately 5% of women with pituitary microadenomas will experience tumor growth during pregnancy, even fewer develop signs or symptoms as a result, and only a rare patient will require surgical intervention. The risk is greater but generally still modest for those with macroadenomas (approximately 15%). Routine serial visual field examinations and serum prolactin determinations therefore are unnecessary, but the patient and her physician must be alert to symptoms that may emerge in any trimester. Headaches generally precede visual disturbances and if either appears, monitoring with serum prolactin measurements, visual field examinations, and imaging should begin. With rare exceptions, dopamine-agonist therapy will arrest and reverse tumor expansion and eliminate associated symptoms and poses no significant risk to mother or fetus. Dopamine-agonist therapy clearly is the treatment of choice for anovulatory hyperprolactinemic women who desire pregnancy or have significant breast tenderness or troublesome galactorrhea, with or without a pituitary adenoma. In the absence of other coexisting causes of infertility, approximately 80% of anovulatory hyperprolactinemic women treated with dopamine agonists may be expected to conceive. In the few who cannot tolerate medical treatment with bromocriptine or cabergoline but desire pregnancy, ovulation induction with exogenous gonadotropins is an effective alternative as prolactin does not alter the effect of the gonadotropins on the ovarian follicles. Women not seeking pregnancy may be offered hormone replacement therapy or even a low-dose OCP that will restore menses and effectively eliminate or prevent the consequences of any associated estrogen deficiency. The uncommonly encountered empty sella syndrome generally is entirely benign and does not progress to pituitary failure. Treatment and subsequent surveillance are the same as in women with a pituitary adenoma. In the rare patient whose amenorrhea results from Sheehan syndrome and its related pituitary insufficiency, more than simple sex steroid replacement therapy may be required. Depending on the extent and scope of pituitary damage, such women also may require glucocorticoid and thyroid hormone replacement. For those affected women who desire a subsequent pregnancy, ovulation induction may be achieved, but only with exogenous gonadotropins. Hypothalamic Disorders Chronic anovulation resulting from abnormal patterns of pulsatile hypothalamic GnRH secretion is the cause of amenorrhea most commonly encountered in practice. A hypothalamic etiology can be established only after first excluding other peripheral, ovarian, and pituitary disorders (normal serum TSH and prolactin levels, low or normal serum FSH, negative imaging, where indicated). Most such women are estrogenized and can also be obese or androgenized (PCOS). Some have a more profound hypogonadism with grossly low levels of estrogen production (hypothalamic amenorrhea), often in association with history of emotional, nutritional, or physical stress. Treatment of women with PCOS is based on whether pregnancy is an immediate goal. Those who desire pregnancy are candidates for ovulation induction. For those not seeking pregnancy, the goals of treatment are to establish regular menses and prevent the consequences of chronic unopposed estrogen stimulation on the endometrium (dysfunctional uterine bleeding, endometrial hyperplasia, endometrial adenocarcinoma), to prevent the emergence or progression of hirsutism, and to reduce the longer-term risks of diabetes and cardiovascular disease associated with the disorder. The wide range of therapeutic options for ovulation induction in anovulatory women with PCOS who desire pregnancy is only summarized here. When obese, weight loss should always be encouraged and may be all that is required to restore menstrual cyclicity. Even if weight loss does not restore ovulatory function, it may be expected to improve the response to ovulation-inducing agents. The substantial costs, risks, and logistic demands of more aggressive forms of treatment might thus be avoided, but weight loss should never be a prerequisite for treatment. Clomiphene citrate is the initial treatment of choice as it is safe, inexpensive, and has few serious side effects. Clomiphene is a competitive estrogen receptor antagonist that acts centrally to deplete hypothalamic estrogen receptors, thereby interfering with estrogen-negative feedback and resulting in a compensatory alteration in pulsatile GnRH secretion that stimulates increased pituitary gonadotropin release and, in turn, drives ovarian follicular activity. Treatment should begin with a low dose (50 mg/d cycle days 3–7 or 5–9 after a spontaneous or progestin-induced menses) and subsequently increase in increments (100 mg/d, 150 mg/d) as needed to achieve ovulation. Most women with PCOS will respond to clomiphene, but many prove resistant and ultimately require alternative treatment. Among clomiphene-resistant anovulatory women with PCOS, a significant majority have insulin resistance. Metformin (1,000–2,000 mg/d in divided doses) is an insulin-sensitizing agent that can restore spontaneous menses and cyclic ovulation in many amenorrheic women with PCOS. Consequently, metformin can be used as the first treatment option for ovulation induction, adding clomiphene in those who fail to respond or, as is more commonly done, metformin can be reserved for those who first prove resistant to clomiphene. In either case, many who fail to ovulate in response to either alone will respond when the two are used in combination. Although the safety of metformin treatment in pregnancy has not been established, preliminary evidence suggests that it may reduce the incidence of spontaneous abortion and gestational diabetes in the subset of women with PCOS and insulin resistance. Those who prove resistant to clomiphene and metformin are candidates for treatment with exogenous gonadotropins. Because such women often are very sensitive to low doses of gonadotropins, unifollicular ovulation can be difficult to achieve and the risk for both multiple pregnancy and ovarian hyperstimulation syndrome is increased. Consequently, treatment must be carefully monitored and is best provided only by clinicians having the necessary training or experience. Ovarian drilling is a contemporary version of the classic ovarian wedge resection and another treatment option for ovulation induction in clomiphene-resistant, hyperandrogenic, anovulatory women with PCOS ( Fig. 34.4). The technique involves laparoscopic cautery, diathermy, or laser vaporization of the ovaries at multiple sites, the objective being to decrease circulating and intraovarian androgen levels by reducing the volume of ovarian stroma. However, the effects of treatment are temporary and postoperative adhesions that may compromise fertility are a distinct possibility. Drilling also has the potential to destroy oocytes and shorten the reproductive lifespan and therefore is best limited to those for whom all other alternatives have been exhausted.

FIG. 34.4. The ovaries in polycystic ovary syndrome shown in a woman with polycystic ovary syndrome at laparotomy. Note the large size of the ovaries, relative to the uterus, with a smooth ovarian capsule without evidence of ovulatory events.

OCPs have long been the mainstay of treatment for amenorrheic women with PCOS who do not desire pregnancy because the combined actions of OCP offer a number of important benefits. Programmed cyclic OCP treatment restores regular, predictable menses, protects the endometrium from the adverse effects of otherwise unopposed estrogen stimulation, and provides effective contraception for the occasional spontaneous ovulation. OCPs decrease circulating androgens by suppressing typically increased serum LH levels that stimulate excess ovarian androgen production and further reduce levels of circulating free androgen by stimulating hepatic production of sex hormone binding globulin. Consequently, within 3 to 6 months after initiation of treatment hair typically grows more slowly, becomes both finer and lighter in color, and therefore is less noticeable and more easily managed. For those with more severe hirsutism in whom treatment with OCP alone fails to achieve the desired effect, spironolactone can be added (100–200 mg/d in divided doses) to provide more effective control. In the dose range indicated, spironolactone acts as a competitive androgen receptor antagonist and blocks androgen action on the pilosebaceous unit (a hair follicle, sebaceous glands, and arrector pili muscle). The multiple actions of combined treatment with OCP and spironolactone (suppression of excess androgen production, increase in androgen binding, and blockade of androgen action) are complementary and generally quite effective for the management of hirsutism. In extreme cases, a more profound suppression of excess androgen production can be achieved by treatment with a GnRH agonist (e.g., leuprolide acetate) with superimposed cyclic or combined continuous hormone replacement with standard doses of estrogen and progestin to prevent the otherwise inevitable emergence of estrogen deficiency symptoms. Suppression of adrenal androgen production by chronic glucocorticoid treatment risks suppression of the adrenal response to stress, immunosuppression, and progressive bone mineral depletion, should therefore be reserved only for those with documented nonclassic CAH, and must be carefully monitored when offered. Insulin resistance certainly contributes to the pathophysiology of PCOS by local amplification of ovarian androgen production in response to LH (directly or via IGF-1) and by reducing hepatic synthesis of sex hormone binding globulin (directly). Moreover, hyperandrogenism clearly is associated with an atherogenic lipid profile. There also is evidence that in individuals with glucose intolerance, metformin treatment can reduce the risk of progression to frank diabetes and, presumably, the increased risk of cardiovascular disease associated with that disease. However, these longer-term potential benefits of primary metformin therapy remain unproven. Moreover, the drug is relatively costly and often poorly tolerated (nausea, vomiting, diarrhea). Under such circumstances, whether primary metformin treatment offers sufficient benefits to justify its use as an alternative or complement to other already proven treatment strategies (weight loss, exercise, OCP, statin therapy) remains controversial. Based on the evidence available to date, primary metformin treatment cannot be recommended for all women with PCOS, but may merit serious

consideration in selected individuals. In the absence of obesity, hirsutism, or other evidence of hyperandrogenism that is characteristic of PCOS, the most likely cause of amenorrhea in women with a normal or low-serum FSH level is a functional disorder of the hypothalamus or higher centers. As for women with PCOS, treatment for women with hypothalamic amenorrhea depends on whether pregnancy is an immediate goal. Those who desire pregnancy clearly are candidates for ovulation induction. For those not seeking pregnancy, the goals of treatment are to establish regular menses, if desired, and to prevent the consequences of chronic estrogen deficiency (genitourinary atrophy, progressive bone demineralization) as most, but not all women with hypothalamic amenorrhea exhibit frankly low levels of endogenous estrogen production. The range of therapeutic options for ovulation induction for women with hypothalamic amenorrhea ( Fig. 34.5) is somewhat narrower than for those with PCOS and again is only summarized here. In women with a low BMI, weight gain should be encouraged but weight gain, like weight loss, can be very difficult to achieve. Women with a low BMI should be educated on the relationship between weight and menstrual function and advised that weight gain may be all that is needed to restore menstrual cyclicity and ovulation. At the least, weight gain may be expected to increase the likelihood of response to more conservative ovulation induction strategies (clomiphene citrate) involving less cost and risk than otherwise may be required. Those in whom symptoms (fasting, purging) or physical findings suggest a potentially serious underlying eating disorder (bulimia, anorexia) should be offered psychiatric evaluation and care before attempting pregnancy as such disorders pose unique risks to both mother and fetus.

FIG. 34.5. The ovary in hypothalamic amenorrhea. Laparoscopy of the pelvis demonstrates a normal-appearing uterus with the left ovary having a smooth capsule without evidence of follicular activity. This is consistent with the lack of gonadotropin stimulation with a normal but unstimulated ovary.

Similarly, women whose hypothalamic amenorrhea is associated with a high level of regular strenuous exercise should be educated on the pathophysiology involved. Reduced physical activity may restore menstrual cyclicity and ovulation and, at the least, less exercise may improve the probability that conservative methods of ovulation induction will succeed. However, once again, reduced levels of exercise should not be a prerequisite for treatment. For many, high levels of exercise are intrinsic to lifestyle or represent the preferred method for relieving stress and may not be desirable or even feasible. Ovulation induction should be offered to those who desire pregnancy. Because of its relatively low cost and risk, clomiphene citrate is again the initial drug of choice. However, unlike women with PCOS, those with hypothalamic amenorrhea often fail to respond to clomiphene treatment. Clomiphene response requires an intact and functional HPO axis, something that most women with hypothalamic amenorrhea do not have (as reflected by low or normal serum FSH levels despite frankly low estrogen production). Consequently, one of two alternative approaches to ovulation induction frequently is required. Exogenous synthetic GnRH, administered in a pulsatile fashion via a portable and programmable infusion pump can effectively restore normal levels of pituitary gonadotropin secretion and spontaneous ovulatory function. GnRH pump therapy requires relatively little monitoring once the effective dose has been established, allows ovulation to occur spontaneously, and is associated with only modest risk of a multiple pregnancy. However, many women object to having an indwelling intravenous catheter over an extended interval of time (approximately 3 weeks) and reject this option. Most often, treatment with exogenous gonadotropins that directly stimulate ovarian follicular development is the preferred method for ovulation induction. Exogenous human chorionic gonadotropin is administered to trigger ovulation when follicular maturation is complete. Treatment requires careful monitoring with serial determination of serum estrogen concentrations and transvaginal ultrasound examinations to determine the size, number, and maturity of developing follicles and to minimize the risks of excessive follicular development, the ovarian hyperstimulation syndrome, and multifetal gestation. Hormone replacement therapy is indicated for those women with hypothalamic amenorrhea who do not desire pregnancy and in whom weight gain or decreased exercise fails to restore menstrual function. In those who are simply anovulatory and not frankly hypoestrogenic or at risk for unwanted pregnancy, cyclic progestational therapy may be all that is required to restore predictable menses and to prevent the endometrial consequences of chronic unopposed estrogen stimulation. However, the level of estrogen production in such women often fluctuates unpredictably and is difficult to judge confidently on clinical or laboratory criteria. Consequently, treatment with cyclic or combined continuous hormone replacement or OCP is prudent to prevent any prolonged interval of hypoestrogenism and the depletion of bone mineral that may result. Given that spontaneous ovulation may occasionally occur, sexually active women with hypothalamic amenorrhea who do not desire pregnancy are best managed with OCP.

SUMMARY POINTS Amenorrhea is a common symptom resulting from an abnormality in the reproductive system that may be anatomic (developmental or acquired), organic, or endocrinologic in nature. Congenital anomalies of the müllerian duct system typically present with normal puberty and may be associated with abdominal pain if there is functional endometrium present. Gonadal dysgenesis represents the lack of any ovarian function prior to puberty leading to sexual infantilism. Premature ovarian failure presents as amenorrhea associated with an elevated FSH level and, if occurring prior to age 30, may be associated with the presence of a Y chromosome. Pituitary tumors are very common, with the majority secreting prolactin and associated with hypoestrogenic amenorrhea and galactorrhea. Hypothalamic amenorrhea is the general term used to describe hypoestrogenic amenorrhea when no anatomic abnormality is identified. It is commonly associated with weight loss or athletic stress. Polycystic ovarian syndrome is a very common clinical problem which may present with amenorrhea but is associated with excess androgen production. A significant majority of these women have insulin resistance. The most common cause of amenorrhea in women of reproductive age is pregnancy. SUGGESTED READINGS Differential Diagnosis of Amenorrhea 1.

Koutras DA. Disturbances of menstruation in thyroid disease. Ann NY Acad Sci 1997;816:280–284.

2.

Laufer MR, Floor AE, Parsons KE, et al. Hormone testing in women with adult-onset amenorrhea. Gynecol Obstet Invest 1995;40:200–203.

3.

Reindollar RH, Byrd JR, McDonough PG. Delayed sexual development: a study of 252 patients. Am J Obstet Gynecol 1981;140:371–380.

4.

Reindollar RH, Novak M, Tho SPT, et al. Adult-onset amenorrhea: a study of 262 patients. Am J Obstet Gynecol 1986;155:531–543.

5.

Styne DM. New aspects in the diagnosis and treatment of pubertal disorders. Pediatr Clin North Am 1997;44:505–529.

6.

Warren MP. Evaluation of secondary amenorrhea. J Clin Endocrinol Metab 1996;81:437–442.

Genital Tract Problems 7. American College of Obstetricians and Gynecologists Committee on Adolescent Health Care. Nonsurgical diagnosis and management of vaginal agenesis. ACOG Committee Opinion No. 274, July 2002. Obstet Gynecol 2002;100:213–216.

8.

Blask AR, Sanders RC, Rock JA. Obstructed uterovaginal anomalies: demonstration with sonography. Part II. Teenagers. Radiology 1991;179:84–88.

9.

Burgis J. Obstructive müllerian anomalies: case report, diagnosis, and management. Am J Obstet Gynecol 2001;185:338–344.

10.

Fujimoto VY, Miller JH, Klein NA, et al. Congenital cervical atresia: report of seven cases and review of the literature. Am J Obstet Gynecol 1997;177:1419–1425.

11.

Kim HH, Laufer MR. Developmental abnormalities of the female reproductive tract. Curr Opin Obstet Gynecol 1994;6:518–525.

12.

March CM. Intrauterine adhesions. Obstet Gynecol Clin North Am 1995;22:491–505.

13.

Quigley CA, De Bellis A, Marschke KB, et al. Androgen receptor defects: historical, clinical, and molecular perspectives. Endocr Rev 1995;16:271–321.

14.

Reinhold C, Hricak H, Forstner R, et al. Primary amenorrhea: evaluation with MR imaging. Radiology 1997;203:383–390.

15.

Williams DM, Patterson MN, Hughes IA. Androgen insensitivity syndrome. Arch Dis Child 1993;68:343–344.

Gonadal Dysgenesis 16.

Guidozzi F, Ball J, Spurdle A. 46,XY pure gonadal dysgenesis (Swyer-James syndrome)—Y or Y not?: a review. Obstet Gynecol Surv 1994;49:138–146.

17.

Ranke MB, Saenger P. Turner's syndrome. Lancet 2001;358:309–314.

18.

Rosenfield RL. Diagnosis and management of delayed puberty. J Clin Endocrinol Metab 1990;70:559–562.

19.

Simpson JL, Rajkovic A. Ovarian differentiation and gonadal failure. Am J Med Genet 1999;89:186–200.

20.

Verp M, Simpson JL. Abnormal sexual differentiation and neoplasia. Cancer Genet Cytogenet 1987;25:191–218.

Premature Ovarian Failure 21.

Aittomaki K, Lucena JL, Pakarinen P, et al. Mutation in the follicle-stimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 1995;82:959–968.

22.

Byrne J, Mulvihill JJ, Myers MH, et al. Effects of treatment on fertility in long-term survivors of childhood cancer. N Engl J Med 1987;317:1315–1321.

23. Clark ST, Radford JA, Crowther D, et al. Gonadal function following chemotherapy for Hodgkin's disease: a comparative study of MVPP and a seven-drug hybrid regimen. J Clin Oncol 1995;13:134–139. 24.

Conway GS, Kaltsas G, Patel A, et al. Characterization of idiopathic premature ovarian failure. Fertil Steril 1996;65:337–341.

25.

Gradishar WJ, Schilsky RL. Ovarian function following radiation and chemotherapy. Semin Oncol 1989;16:425–436.

26.

Guerrero NV, Singh RH, Manatunga A, et al. Risk factors for premature ovarian failure in females with galactosemia. J Pediatr 2000;137:833–841.

27.

Hoek A, Schoemaker J, Drexhage HA. Premature ovarian failure and ovarian autoimmunity. Endocr Rev 1997;18:107–134.

28.

Howell S, Shalet S. Gonadal damage from chemotherapy and radiotherapy. Endocrinol Metab Clin North Am 1998;27:927–943.

29.

Laml T, Schulz-Lobmeyr I, Obruca A, et al. Premature ovarian failure: etiology and prospects. Gynecol Endocrinol 2000;14:292–302.

30.

McDonough PG. The search for the gonadotropin-resistant ovary—a lost planet or a dinosaur? Fertil Steril 1999;71:1169–1170.

31. Morishima A, Grumbach MM, Simpson ER, et al. Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J Clin Endocrinol Metab 1995;80:3689–3698. 32. Saenger P, Wikland KA, Conway GS, et al. Fifth International Symposium on Turner Syndrome. Recommendations for the diagnosis and management of Turner syndrome. J Clin Endocrinol Metab 2001;86:3061–3069. 33. van Kasteren YM, Schoemaker J. Premature ovarian failure: a systematic review on therapeutic interventions to restore ovarian function and achieve pregnancy. Hum Reprod Update 1999;5:483–492.

Pituitary Disorders (Hyperprolactinemia) 34.

Colao A, Loche S, Cappa M, et al. Prolactinomas in children and adolescents. Clinical presentation and long-term follow-up. J Clin Endocrinol Metab 1998;83:2777–2780.

35.

Conner P. Fried G. Hyperprolactinemia; etiology, diagnosis and treatment alternatives. Acta Obstet Gynecol Scand 1998;77:249–262.

36.

Hodgson SF, Randall RV, Holman CB, et al. Empty sella syndrome. Report of 10 cases. Med Clin North Am 1972;56:897–907.

37.

Maguire GA. Prolactin elevation with antipsychotic medications: mechanisms of action and clinical consequences. J Clin Psychiatry 2002;63:56–62.

38.

Molitch ME. Disorders of prolactin secretion. Endocr Metab Clin North Am 2001;30:585–610.

39.

Tan SL. Hyperprolactinemia and the management of related tumors. Curr Opin Obstet Gynecol 1990;2:378–385.

40.

Touraine P, Plu-Bureau G, Beji C, et al. Long-term follow-up of 246 hyperprolactinemic patients. Acta Obstet Gynecol Scand 2001;80:162–168.

Hypothalamic Amenorrhea 41.

Berga SL. Behaviorally induced reproductive compromise in women and men. Semin Reprod Endocrinol 1997;15:47–53.

42.

Hall JE. Physiologic and genetic insights into the pathophysiology and management of hypogonadotropic hypogonadism. Ann Endocrinol 1999;60:93–101.

43.

Hurst BS. Ovulation induction in the nonestrogenized patient. Semin Reprod Endocrinol 1996;14:299–308.

44.

Laughlin GA, Dominguez CE, Yen SSC. Nutritional and endocrine-metabolic aberrations in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab 1998;83:25–32.

45.

Marshall JC, Eagleson CA, McCartney CR. Hypothalamic dysfunction. Mol Cell Endocrinol 2001;183:29–32.

46.

Miller KK, Klibanski A. Amenorrheic bone loss. J Clin Endocrinol Metab 1999;84:1775–1783.

47.

Warren MP. Stiehl AL. Exercise and female adolescents: effects on the reproductive and skeletal systems. J Am Med Wom Assoc 1999;54:115–120.

Polycystic Ovarian Syndrome 48.

Azziz R, Dewailly D, Owerbach D. Non-classic adrenal hyperplasia: current concepts. J Clin Endocrinol Metab 1994;78:810–815.

49. Bili H, Laven J, Imani B, et al. Age-related differences in features associated with polycystic ovary syndrome in normogonadotropic oligo-amenorrhoeic infertile women of reproductive years. Eur J Endocrinol 2001;145:749–755. 50.

Dunaif A, Thomas A. Current concepts in the polycystic ovary syndrome. Annu Rev Med 2001;52:401–419.

51. Glueck CJ, Phillips H, Cameron D, et al. Continuing metformin throughout pregnancy in women with polycystic ovary syndrome appears to safely reduce first-trimester spontaneous abortion: a pilot study. Fertil Steril 2001;75:46–52. 52.

Glueck CJ, Wang P, Kobayashi S, et al. Metformin therapy throughout pregnancy reduces the development of gestational diabetes in women with polycystic ovary syndrome. Fertil Steril

2002;77:520–525. 53. Hughes E, Collins J, Vandekerckhove P. Clomiphene citrate for ovulation induction in women with oligo-amenorrhoea. In: Cochrane Database of Systemic Reviews. The Cochrane Library 2000; Issue 2 (CD000056). Oxford: Update Software. 54. Hughes E, Collins J, Vandekerckhove P. Gonadotrophin-releasing hormone analogue as an adjunct to gonadotropin therapy for clomiphene-resistant polycystic ovarian syndrome. In: Cochrane Database of Systemic Reviews. The Cochrane Library 2000; Issue 2 (CD000097). Oxford: Update Software. 55. Hughes E, Collins J, Vandekerckhove P. Ovulation induction with urinary follicle stimulating hormone versus human menopausal gonadotropin for clomiphene-resistant polycystic ovary syndrome. Cochrane Database of Systemic Reviews. The Cochrane Library 2000; Issue 2 (CD000087). Oxford: Update Software. 56.

Legro RS. Polycystic ovary syndrome: the new millennium. Mol Cell Endocrinol 2002;186:219–225.

57.

Nestler JE, Jakubowicz DJ, Evans WS, et al. Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovary syndrome. N Engl J Med 1998;338:1876–1880.

Chapter 35 Abnormal Uterine Bleeding Danforth’s Obstetrics and Gynecology

Chapter 35 Steve N. London

Abnormal Uterine Bleeding

DIAGNOSIS Pregnancy Hormones Iatrogenic Mechanical Infection Cancer IMAGING AND SPECIAL STUDIES IN THE MANAGEMENT OF ABNORMAL UTERINE BLEEDING TVS and SIS in the Evaluation of Abnormal Uterine Bleeding The Role of MRI in the Evaluation of Women with Abnormal Uterine Bleeding MANAGEMENT OF SPECIAL SITUATIONS Acute Hemorrhage Menorrhagia Abnormal Bleeding on Oral Contraceptives Abnormal Bleeding in Perimenopause SUMMARY POINTS SUGGESTED READINGS Diagnosis Pregnancy Hormones Iatrogenic Mechanical Infection Cancer Imaging and Special Studies in the Management of Abnormal Uterine Bleeding Management of Special Situations

This chapter presents a simplified diagnostic and therapeutic approach to the woman with abnormal uterine bleeding, discusses imaging studies that facilitate accurate diagnosis and management, and presents therapeutic plans for management of acute hemorrhage, menorrhagia, bleeding on oral contraceptives, and the dysfunctional bleeding associated with perimenopause. This chapter does not address abnormal bleeding in the postmenopausal woman either on or off hormone replacement. This chapter is divided into three sections: the first is a simplified diagnostic approach with discussion about the differential diagnosis and therapeutic options for each condition. This is followed by section on imaging modalities available to aid in the diagnosis of abnormal uterine bleeding. Special emphasis is placed on the role of pelvic ultrasound, saline infusion sonogram, and magnetic resonance imaging (MRI). The third section presents a therapeutic approach in special clinical situations, including the patient presenting with acute hemorrhage, bleeding on oral contraceptives, menorrhagia, and perimenopause.

DIAGNOSIS The differential diagnosis for all women presenting with abnormal uterine bleeding can be encompassed by the mnemonic PHIMIC (pregnancy, hormones, iatrogenic, mechanical, infection, and cancer) ( Table 35.1). The use of this mnemonic simplifies the approach and allows the clinician to cover a vast array of possibilities in the most efficient and treatment focused manner. This category/group approach provides for patient safety by assuring that serious conditions such as pregnancy, cancer, or unrecognized infections are not overlooked.

TABLE 35.1. PHIMIC—differential diagnosis for abnormal uterine bleeding

Before evaluating what is abnormal uterine bleeding, we must begin with a definition of what is normal uterine bleeding. Normal menstrual flow is the result of a withdrawal of an estrogen-primed endometrium. This should occur at regular intervals that vary by no more than 3 days from the woman's mean menstrual interval. Typically, normal ovulatory menses occur every 24 to 35 days with most women experiencing a predictable interval. Ovulatory women will set their own mean cycle length, which is defined as from the first day of bleeding, cycle day 1, to the next first day of bleeding in the subsequent cycle and will not vary more than 3 days. For example, if a woman's mean cycle interval is 26 days; her menstrual calendar should reflect menses varying from between 23 and 29 days. A variation greater than 3 days is suggestive of either oligoovulation or estrogenized anovulation and is classified as abnormal. The mean duration of normal menstrual flow is 3 to 7 days. Bleeding that persists longer than 7 days or decreases hemoglobin levels is considered, by definition, excessive. Unfortunately, there is no simple clinical way to quantitate the amount of blood lost during a menstrual period. Counting the number of sanitary protectives used in a menstrual cycle has proven to be inaccurate based on differences in personal hygiene and cultural backgrounds. When accurately quantitated, blood loss during an entire normal ovulatory menstrual cycle is approximately 60 to 80 mL. This amount of blood loss will not cause a decrease in hemoglobin in women consuming a normal balanced diet. Traditionally, aberrations in menses have been classified as either menorrhagia, defined as an excessive amount of menstrual blood loss, metrorrhagia, defined as intermenstrual, irregular or otherwise noncyclic bleeding, or menometrorrhagia, which combines both menorrhagia and metrorrhagia leading to heavy irregular uterine bleeding ( Table 35.2). While these terms are traditionally used they require that the clinician make a cognitive decision based on the patient's description of her symptomatology. This has the potential to alter the ability to make an accurate diagnosis, as the differential diagnosis of a woman with menorrhagia is different than that of metrorrhagia. Since menstrual histories are sometimes difficult to obtain due to the frequent miscommunication between the patient and her physician, limiting the description of the symptomatology to abnormal uterine bleeding without narrowing it to the more scientific term is encouraged. By approaching every woman, irrespective of whether she has menorrhagia, metrorrhagia, or menometrorrhagia with a generalized mnemonic such as PHIMIC, the clinician avoids narrowing the differential diagnosis too quickly, thus assuring that all possibilities are considered. Therefore, when a patient presents with abnormal uterine bleeding, its etiology will be found within one of six broad diagnostic categories: pregnancy, hormonal, iatrogenic, mechanical, infection, or cancer. This allows the same diagnostic approach for every woman, regardless of her age.

TABLE 35.2. Definitions

Pregnancy

Pregnancy should be considered in any woman who presents with abnormal uterine bleeding unless she has had a hysterectomy or is over the age of 60. Even women who have had tubal ligations as well as women who are in their 50s and have not menstruated in 1 to 2 years may have the occasional pregnancy. It is safe, on a clinical basis, to say that a physician will never regret ordering too many pregnancy tests, but may regret not ordering enough. Screening for pregnancy is typically accomplished by using a highly accurate urinary pregnancy test. It is not necessary to do a serum quantitative human chorionic gonadotropin (hCG) evaluation on every woman who presents with abnormal uterine bleeding. Currently available urinary pregnancy tests are immunometric or “sandwich” assays that use antibodies directed to the intact hCG molecule. This makes the urine pregnancy test extremely sensitive and specific. It is important that every clinician know what the limits are in sensitivity of the pregnancy test and the reference preparation that their practice uses. The two reference preparations currently in use are the First and the Third International Reference preparations. The original reference preparation was the Second International Reference Preparation (2nd-IRP) which measured intact and free ß-hCG subunits. One unit of the 2nd-IRP is approximately equivalent to two units of the first or third IRP. If the woman with abnormal uterine bleeding is found to have a positive urinary pregnancy test, a serum quantitative ß-hCG assay should be obtained. While this result is pending, it is appropriate to perform a pelvic sonogram to determine whether there is an identifiable intrauterine pregnancy, an adnexal mass, or a negative exam (no pregnancy within the uterus and no adnexal masses). If the ß-hCG returns above the discriminatory zone, then the finding of an empty uterus is suggestive of an ectopic pregnancy and management of ectopic pregnancy can be instituted. If the ß-hCG is below the discriminatory zone and the patient is asymptomatic and reliable, she can return in 48 hours for a repeat ß-hCG assay to determine the probability of a viable intrauterine pregnancy. Upon return, the patient can be rescanned if her ß-hCG is expected to be over the discriminatory zone or a repeat ultrasound can be scheduled based upon the projected time period interval when the level will be anticipated to exceed the discriminatory zone maximum. It is important to remember that 10% to 15% of all ectopic pregnancies have normal ß-hCG doubling times and that rupture of ectopic pregnancies has occurred with falling ß-hCG values. If the level has remained constant or is falling this is suggestive of a nonviable pregnancy, whether in the uterus or in an ectopic location, this can be followed for another 48 hours. Many pregnancies are lost without being able to determine if it was an early intrauterine loss or a nonviable ectopic pregnancy which resolved without becoming clinically symptomatic. Hormones Hormonal disorders that can cause abnormal bleeding can be grouped under four broad categories: coagulation disorders, prolactin disorders, thyroid disorders, and chronic anovulation ( Table 35.3). Coagulation disorders are relatively rare. Coagulation disorders are most commonly found in girls who are within 1 year of their first menstrual cycle and have a falling hemoglobin level. Von Willebrand disease needs to be excluded in every adolescent presenting with excessive menstruation (menorrhagia). Women with coagulopathies may present with acute hemorrhage and hypovolemic symptoms. These women almost invariably present with anemia or frank hemorrhage to the emergency department within a short interval after the onset of menstruation. They also need to be evaluated for conditions associated with abnormal bone marrow function or rapid turnover of platelets such as leukemia, idiopathic thrombocytopenic purpura, or aplastic anemia.

TABLE 35.3. Hormonal causes: disorders to consider

While hyperprolactinemia is a significant cause of abnormal bleeding in women, the prolactin itself is not the direct cause of the abnormal bleeding. The high prolactin levels are the result of alterations in neurotransmitters within the hypothalamus and prolactin down-regulates gonadotropin-releasing hormone (GnRH) secretion resulting in altered gonadotropin release, ovarian dysfunction, and subsequent abnormal uterine bleeding. A prolactin-secreting pituitary adenoma is the most common pituitary tumor in women and should be the first concern of a health care provider when evaluating a patient with an elevated prolactin level. It is reasonable to perform an MRI of the sella turcica in all women with a prolactin level greater than 60 ng/mL. In general, prolactin excess should be treated with bromocriptine if pregnancy is desired. If pregnancy is not desired and the patient has enough endogenous estrogen to prevent osteoporosis, then the prolactin level can simply be followed. Whether or not the woman has enough estrogen to prevent osteoporosis can be determined by the patient's response to progestin administration. If the patient fails to menstruate in response to this progestin challenge then consideration of estrogen replacement or other drug for prevention of osteoporosis is necessary. However, the abnormal bleeding will not resolve unless the prolactin level is returned to normal by treatment with a dopamine agonist such as bromocriptine. Even after the prolactin level has returned to normal there may be enough hypothalamic dysfunction that ovulation induction by other means may be required in women desirous of pregnancy. If the patient has a small microadenoma and desires contraception then oral contraceptives are the treatment of choice. If the patient does not need or desire contraception, cyclic progestin therapy with medroxyprogesterone acetate 10 mg for 14 days every other month can be used. A woman's lifetime risk of developing thyroid disease is approximately 15% and excess or a deficiency of thyroid hormone can affect a woman's menstrual cycle. The most common disorder is hypothyroidism and approximately two-thirds of women with hypothyroidism will develop an aberration of menses. Any woman who develops abnormal uterine bleeding, postpartum depression, premenstrual syndrome, or depression should be screened for thyroid disease ( Table 35.4). The principal screening test for thyroid disease is a measurement of serum thyroid-stimulating hormone (TSH). No further testing is required if the TSH is within normal limits. If the TSH is low this suggests hyperthyroidism and measurement of free thyroxine (T4) and free triiodothyronine (T3) should be performed. If either one is elevated the diagnosis of hyperthyroidism is made. After diagnosing hyperthyroidism, a radioactive iodine uptake scan is performed. If the T4 and T3 are normal a diagnosis of subclinical hyperthyroidism may be entertained. If the TSH is high then a free T4 should be obtained. If the free T4 is low, replacement with synthetic thyroxine is begun. A dose of 25 µg per day of synthetic thyroxine is given and increased every 4 weeks by 25 µg per day until the patient's TSH normalizes and her clinical symptoms have resolved. Menstruation will usually return to normal 8 to 10 weeks after the patient has achieved a euthyroid state.

TABLE 35.4. Thyroid dysfunction: when to screen

The most common hormonal disorder causing abnormal uterine bleeding is estrogenized anovulation (World Health Organization type II anovulation). Anovulatory bleeding in the past has been described as dysfunctional uterine bleeding. Dysfunctional uterine bleeding refers to bleeding unrelated to mechanical factors, iatrogenic causes, infectious agents, cancer, or pregnancy. There is no one particular pattern of bleeding that unequivocally defines dysfunctional uterine bleeding. A woman may present with anything from complete secondary amenorrhea to frank hemorrhage. Therefore, characterizing the menstrual flow is of no diagnostic advantage. Syndromes of chronic anovulation can be grouped etiologically as either hypothalamic (central nervous system), adrenal (adrenal androgen production), ovarian (polycystic ovarian syndrome; PCOS), or perimenopausal in origin. None of these groups have a clear-cut laboratory profile and a precise classification often proves difficult. PCOS is one of the more confusing causes of chronic anovulation. The exact etiology of this syndrome remains elusive, although a large percentage of women with PCOS have been noted to have hyperinsulinemia secondary to insulin resistance. The National Institutes of Health (NIH) consensus conference has defined PCOS as amenorrhea or oligoovulation with clinical or biochemical evidence of hyperandrogenemia. While many authorities have frequently found insulin resistance in women with PCOS, there currently is no practical and reliable method to clinically diagnose insulin resistance. The use of insulin sensitizers in this group, despite being somewhat controversial, is enjoying widespread use, primarily for women desirous of pregnancy. Because of the controversy in the use of insulin sensitizers in the treatment of the abnormal uterine bleeding associated with PCOS, it is best to treat these patients with cyclic progestin, oral contraceptives, or clomiphene citrate if pregnancy is desired and reserve insulin sensitizers for those women in whom clomiphene citrate has failed and who want to become pregnant. Another common hormonal cause of anovulatory uterine bleeding is the perimenopausal transition. Most women experience a 5- to 8-year transition from regular menstrual cycles to frank amenorrhea with an interval of abnormal uterine bleeding in between. This interval is marked by intervals of hypoestrogenism with symptoms such as hot flashes, irritability, and some vaginal atrophy, alternating with intervals of estrogenized anovulation. Usually the first sign of the perimenopausal transition is a shortening of the menstrual cycle accompanied by an increase in premenstrual symptomatology. Women over the age of 35 who present with a change in their mean menstrual cycle interval irrespective of other symptomatology who desire future childbearing should be screened for impending ovarian failure with a day 3 follicle-stimulating hormone (FSH) level. There are two therapeutic options for women with abnormal bleeding in the perimenopausal transition. The first is the cyclic administration of progestins. Typically, this is accomplished by the administration of 10 mg of medroxyprogesterone acetate for 14 days each month or every other

month. This therapy may be associated with intermenstrual bleeding and occasionally may precipitate premenstrual symptoms. There is no place for the use of injectable progestins in the management of perimenopausal bleeding. The best choice to control abnormal bleeding in the perimenopausal transition, if the patient is a healthy nonsmoking woman, is the use of continuous combined oral contraceptives. The use of oral contraceptives in perimenopause will ease the transition by enforcing regular menstrual shedding, decrease blood loss, reduce premenstrual symptoms, preserve bone mass, and reduce the lifetime risks of endometrial and ovarian cancers without increasing the risk of breast cancer. The use of hormone replacement therapy, as used in frankly menopausal women, is to be discouraged as this provides additional estrogen and progestin without suppressing the endogenous production and thus increases the total amount of sex steroid exposure to the uterus. Iatrogenic Iatrogenic causes of abnormal uterine bleeding appear to be increasing as many drugs used for cancer, renal disease, and transplantation either mimic or modify endogenous hormones and cause abnormal uterine bleeding. Every woman presenting with a change in her menstrual cycle requires that a careful history be taken with respect to her general health as well as use of any prescription or over-the-counter medications. The physician should inquire about drugs prescribed within the past 6 months even if the patient is not currently taking any medications ( Table 35.5). Drugs that can effect the menstrual cycle include prednisone, tamoxifen, coumadin, heparin, and Depo-Provera.

TABLE 35.5. Drugs associated with abnormal uterine bleeding

Mechanical Uterine leiomyomata are one of the most common causes of abnormal uterine bleeding and one of the leading causes of hysterectomy in the United States. Approximately one fourth of all women have uterine leiomyomata, but fortunately only a small fraction of that number will be clinically symptomatic and come to medical attention. Submucous or intracavitary myomas most commonly cause abnormal uterine bleeding. Submucous myomas may cause bleeding by frank erosion and ulceration of the endometrial surface, and intracavitary myomas prevent the normal hemostatic mechanisms required to stop normal menstruation. An intramural myoma may cause bleeding by creating large venous lakes and preventing the uterus from appropriately constricting blood vessels supplying the endometrium at the time of menstruation ( Fig. 35.1). Bleeding from a myoma can vary from frank hemorrhage to slight intermenstrual spotting. Uterine leiomyomata are estrogen-dependent. Therefore, any agent that blocks estrogen's biosynthesis or action could theoretically cause regression of myoma. Currently, there is no medical therapy that truly causes regression of leiomyomata. GnRH agonists block gonadotropin secretion resulting in hypoestrogenism and, hence, cause a reduction in the size of the estrogen-dependent muscle cells within the myoma. However, there is rapid return of myoma volume and symptomatology once the GnRH agonist is discontinued and estrogen production by the ovary resumes. RU-486 and other antiprogesterone drugs have been shown to shrink myomas in clinical trials. The long-term effect of these agents on leiomyomata however is unknown. Depo-Provera is often effective in reducing the amount of abnormal bleeding and may even cause some regression in myoma size. Oral contraceptive agents do not cause regression of the myoma but they may stabilize the endometrium enough to effectively control the abnormal bleeding associated with uterine myoma. It is very reasonable to make an attempt to control abnormal uterine bleeding with a trial of oral contraceptive agents in women with leiomyomata before attempting more expensive and hazardous therapies. Despite the use of oral contraceptives, the abnormal bleeding in many women will not be controlled or they may not want to take oral contraceptives on a continuous basis. It is reasonable to consider a hysterectomy or a uterus-sparing operation such as myomectomy when more conservative medical methods have proven unsuccessful. There is not enough outcome evidence in the literature to recommend one surgical approach over the other. Until more data are available, management of uterine leiomyomata will remain empirical.

FIG. 35.1. Ultrasound of uterine leiomyoma. This is a transvaginal ultrasound in a woman who has an intramural myoma ( arrows). The myoma clearly does not impinge on the endometrial cavity with intervening myometrium. The endometrial stripe is defined by a double arrow.

Endometrial polyps are found in up to 50% of women presenting with abnormal uterine bleeding. There is no precise etiology for the formation of these polyps but they clearly originate from the endometrium. Detecting polyps can often be difficult. Endometrial polyps are best diagnosed by either hysteroscopy ( Fig. 35.2) or saline infusion sonogram (SIS). The SIS is extremely sensitive but may overdiagnose the polyps. It is not uncommon at the time of hysteroscopy and dilation and curettage (D&C) for removal of the polyps diagnosed by SIS to find none present. Additionally, many small polyps will regress spontaneously. Since small polyps are found in asymptomatic women, the role of small endometrial polyps in a woman complaining of abnormal uterine bleeding is unknown.

FIG. 35.2. Hysteroscopic visualization of an endometrial polyp. An endometrial polyp is present on the anterior fundal aspect of the endometrial cavity ( arrow).

Infection Infection is one of the most overlooked causes of abnormal uterine bleeding. Historically, endometritis has been considered a cause of abnormal uterine bleeding, despite a paucity of evidence in the literature. Endometritis is diagnosed by the finding of an increased number of plasma cells on endometrial biopsy. Culture of the endometrial cavity is not clinically useful and is complicated by the high rate of contamination by vaginal organisms. Studies have failed to show any significant endometrial infection associated with abnormal uterine bleeding. The absence of culture-positive studies and the subjective diagnostic criteria used casts doubt whether chronic endometritis truly causes abnormal uterine bleeding. However, endocervicitis is a known cause of abnormal bleeding, particularly intermenstrual and postcoital bleeding. It is appropriate for women with abnormal bleeding to have cultures or DNA probe tests performed for chlamydia and gonorrhea to rule out those organisms as a cause. Due to the false-negative rate of these diagnostic procedures it is often prudent to treat empirically for chlamydia, if no other cause is found. Cancer Endometrial and cervical cancers are uncommon yet important causes of abnormal uterine bleeding. Cervical cancer often presents with postcoital bleeding or irregular menstruation. Even if the patient has a normal Pap smear within the previous year it is prudent to repeat the smear, particularly if the patient has other risk factors for cervical cancer. Endometrial cancer remains a possibility in any woman presenting with abnormal uterine bleeding and endometrial biopsy should be performed in all

women over the age of 35, the age at which the incidence begins to rise. Therapy will depend upon the staging of the disease. Complex endometrial hyperplasia with atypia is usually managed by simple hysterectomy. Other forms of hyperplasia may be safely managed with cyclic or continuous high-dose progestin therapy. If cyclic progestin is chosen, the minimum effective dose is 10 mg of medroxyprogesterone acetate for 14 days each month. Continuous progestin therapy is usually 20 to 40 mg of medroxyprogesterone acetate daily for 90 days. The use of a test-of-cure endometrial biopsy or hysteroscopy with curettage is controversial. However, it is reassuring to the patient and physician to document regression and uncover the rare progesterone receptor–negative hyperplasia.

IMAGING AND SPECIAL STUDIES IN THE MANAGEMENT OF ABNORMAL UTERINE BLEEDING A transvaginal sonogram (TVS) has become an integral component of the evaluation of women with abnormal uterine bleeding. This is true even if the patient has a normal pelvic exam as the sonogram is focusing on the internal architecture of the uterus not necessarily the overall size. The use of TVS identifies small myoma, adenomyosis, and excessive endometrial thickness. The presence of these factors requires further evaluation with either biopsy or SIS. While TVS is necessary even if the patient has a normal pelvic exam, the TVS does not replace the health care provider performing a thorough and careful pelvic exam. TVS and SIS in the Evaluation of Abnormal Uterine Bleeding Hysteroscopy and D&C are not often used as primary diagnostic and therapeutic modalities. While office hysteroscopy does not require anesthesia, the procedure is expensive and requires an experienced operator as compared to TVS and endometrial biopsy. TVS allows acoustic visualization of the endometrium, ovaries, and the myometrium, which is more information than obtained by hysteroscopy. TVS is very sensitive for identification of leiomyomata that are greater 2 cm in diameter. TVS also has the ability to identify adenomyosis and endometrial polyps but its sensitivity and specificity for these conditions is lower. TVS has a 60% sensitivity and a 93% specificity for identifying pathology within the endometrial cavity. Still, TVS should be the first- line diagnostic modality when evaluating disorders of the myometrium and endometrium because of its simplicity and lack of risk. The sensitivity and specificity of TVS can be enhanced with the infusion of saline creating an SIS. Many other names have been used for this procedure including hydrosonogram, sonohysterosalpingography, and saline hysterosonography. No consensus for the appropriate terminology has been reached, but “saline infusion sonography” appears to be gaining acceptance due to its acronym SIS, which avoids confusion with old acronyms of HSG for hysterosalpingogram. The endometrial thickness in women of reproductive age varies greatly. Still, it should be measured to enhance the sensitivity of TVS in predicting intrauterine pathology. When the endometrial thickness is greater than 8 mm consideration should be given to SIS. The accuracy and precision of SIS can be further enhanced with the addition of an endometrial biopsy. One study has suggested the sensitivity and specificity for the detection of intrauterine pathology at 97% and 70%, respectively, and a positive and negative predictive value of 82% and 94% compared with hysteroscopy/D&C or hysterectomy. Therefore, the use of endometrial biopsy and SIS to identify intrauterine pathology has displaced hysteroscopy and D&C as the primary procedure in the evaluation of abnormal uterine bleeding. Even though surgical evaluation is now a second-line procedure, the health care provider must still remember that hysteoscopy with curettage is more accurate than the less invasive office-based procedures and still has a place in the diagnosis and management of the patient with abnormal uterine bleeding. The Role of MRI in the Evaluation of Women with Abnormal Uterine Bleeding With the development of new high-field strength magnets the evaluation of the uterus by MRI has become much quicker and accurate. MRI has the advantage over computed tomography (CT) by not being limited by the surrounding bony structures. Compared to TVS, the MRI is less dependent on operator experience, uterine position, and size. The uterus on MRI is best seen on the T2 weighted sequence. Three regions within the uterus are normally identified: a central bright area that corresponds to the endometrium, another area of intermediate signal intensity that corresponds to the myometrium, and a third layer between the endometrium and myometrium that is hormonally sensitive and termed the junctional zone. The thickness signal characteristics of these three areas are evaluated independently when looking for uterine pathology. The greatest advantage of MRI is its ability to distinguish between adenomyosis and uterine leiomyomata. While both of these conditions can cause pelvic pain, typically increasing dysmenorrhea as well as abnormal bleeding, there are patients who desire future fertility or for other reasons desire to preserve their uterus and the conservative operations are different for each condition. The MRI image has a sensitivity of 86% and a positive predictive value of 65% in detecting the presence of adenomyosis when junctional zone thickening is greater than 12 mm. The MRI can also be used to determine whether the adenomyosis is diffuse or superficial. If the adenomyosis is superficial it can successfully be treated with roller-ball endometrial ablation, whereas deep adenomyosis is best treated by hysterectomy. Additionally, the MRI has the ability to accurately define myoma less than 2 cm, the limit of TVS. This is important for women undergoing in vitro fertilization. MRI is becoming an efficient and cost-effective complementary tool and may soon supplant hysteroscopy as the second-line diagnostic modality in abnormal uterine bleeding.

MANAGEMENT OF SPECIAL SITUATIONS Acute Hemorrhage It is not uncommon for women with chronic anovulation or uterine leiomyomata to present to the emergency department with acute hemorrhage and symptoms of hypovolemia. These patients need rapid control of their hemorrhage and a clinical decision has to be made quickly as to whether to proceed to the operating room or to make an attempt at medical control. If the patient is hemodynamically unstable, a D&C is indicated as it will result in immediate cessation of the bleeding and allow time for transfusion and correction of the underlying pathology. If the patient is stable enough to attempt medical management, fluid resuscitation is begun. Blood products are then given as necessary and high-dose parenteral estrogen therapy initiated. The high dose of estrogen will initiate endometrial proliferation and increase fibrinogen as well as factors V and IX, but most important, it promotes platelet aggregation and clotting at the capillary level. It is appropriate to use i.v. estrogen even when the patient has uterine leiomyomata. Both i.v. and p.o. estrogen work equally as well but the parenteral route is usually chosen if the clinical situation requires rapid cessation of the blood loss. Intravenous conjugated estrogen is given 25 mg every 4 hours until the bleeding stops. The oral regimen is equally efficacious and consists of 2.5 mg conjugated estrogen or 2 mg of micronized estradiol every 4 hours for 24 hours followed by a single daily dose for 7 to 10 days. If the bleeding persists for more than 12 to 24 hours with either one of these regimens, a D&C is indicated. Whether the estrogen is given i.v. or p.o., a progestin should be started at the same time and continued for 5 to 7 days at which time the patient should expect a withdraw bleed. If the patient is not having acute hemorrhage but rather is having heavy menstrual flow and becoming anemic then the use of high-dose oral contraceptives may be considered. In these cases the patient is given 4 pills containing 30 to 35 µg of ethinyl estradiol for 4 days followed by 3 pills for 7 days followed by 2 pills for 11 days then 1 pill a day thereafter. Hysterectomy should be considered in women who have completed childbearing and when medical management has failed. Menorrhagia Menorrhagia affects approximately 15% to 20% of women presenting with abnormal uterine bleeding and causes significant social inconvenience as well as the potential for significant anemia. The first line of therapy for the treatment of women with menorrhagia is antiprostaglandin agents, the cyclooxygenase inhibitors or nonsteroidal antiinflammatory drugs (NSAIDs). Women with menorrhagia have significantly higher levels of prostaglandin E (PGE) when compared to prostaglandin F 2a. The antiprostaglandin agents affect the balance between PGE and thromboxane. Many prospective trials have shown use of NSAIDs beginning on the day of menstruation and continued throughout menses significantly decrease blood loss. The use of the cyclooxygenase-1 (COX-1) class of agents is contraindicated in women with peptic ulcer disease, asthma, and allergy to aspirin. Numerous regimens have been used all with similar efficacy: mefenamic acid 500 mg t.i.d., ibuprofen 400 mg t.i.d., and naproxen sodium 275 mg q6h after a loading dose of 550 mg. All these regimens have demonstrated a significant decrease in blood loss. The newer COX-2 inhibitors have markedly fewer problems and inhibit the same COX enzyme, but have not been tested in clinical trials. They may improve the overall success of this approach with fewer side effects. Another strategy, using the levonorgestrel intrauterine device (IUD), has recently been introduced. Several controlled trials have shown that its use significantly decreases the amount of menstrual blood lost per cycle by causing an intense atrophy of the endometrium. The levonorgestrel IUD should be considered in any woman who needs contraception and who is experiencing menorrhagia. If these therapies fail to control the problem, oral contraceptives can also be considered. Surgical therapy should be considered when medical management has failed or is medically contraindicated. Surgical therapy consists of either endometrial ablation or hysterectomy. Endometrial ablation by any technique is reported to provide 90% to 95% patient satisfaction, with amenorrhea being achieved in approximately 50% of patients. Unfortunately, the long-term problems associated with endometrial ablation are not known. With regard to hormone therapy after menopause, the patient having endometrial ablation must be regarded as having fully responsive endometrium. There is a theoretical risk of concealed endometrial cancer, but this has not been reported to date. Hysterectomy results in amenorrhea 100% of the time and has a higher patient satisfaction rate than endometrial resection 2 years after the procedure. Abnormal Bleeding on Oral Contraceptives Abnormal bleeding on oral contraceptives is usually due to an insufficient amount of estrogen in a particular pill to stabilize the endometrium. Strategies to control the bleeding involve increasing the estrogen content of the pill, changing to a different birth control pill which increases the estrogen/progestin ratio, or giving supplemental estrogen. Since most noncompliance with oral contraceptives is related to the progestin content and if a patient is comfortable with the side effects in her particular pill, an attempt should be made to continue with that level of progestin and add supplemental estrogen, either 1.25 mg conjugated estrogen or 2 mg of estradiol starting at

the time the patient bleeds and continuing for 7 days. This regimen is repeated at the same time for two consecutive months whether or not the patient has spotting in the two subsequent cycles. Alternatively, the patient can simply use a pill with a higher estrogen/progestin ratio or a higher-dose estrogen pill to determine if that will be sufficient to control the bleeding. If the patient fails to respond to these regimens then other factors need to be assessed. Abnormal bleeding on pills is significantly increased in women who smoke, due to enhanced estrogen clearance, and when taking antibiotics because of alterations in absorption. All efforts should be made to have the patient stop smoking. Additionally, cultures or DNA probe for chlamydia should be obtained as endocervicitis-related chlamydia is a significant cause of abnormal bleeding in women on birth control pills. Abnormal Bleeding in Perimenopause The most common cause of abnormal uterine bleeding in the perimenopausal woman is estrogenized anovulation. Still, every woman in this category needs a complete evaluation as detailed earlier in this chapter, particularly as they are now in the age range where the incidence of endometrial cancer is beginning to be significant. Attributing the abnormal uterine bleeding to the perimenopausal transition is a diagnosis of exclusion and can only be made after a through evaluation. Therapy with oral contraceptives should be initiated provided the patient has no contraindication to their use, primarily smoking. Oral contraceptives have been shown to decrease the amount of blood loss associated with menstruation providing cycle regulation and relief from menorrhagia. Additionally, oral contraceptive agents preserve bone mass during a time when occult bone loss may be occurring. Most significantly, oral contraceptive agents provide reduction in the risk of endometrial and ovarian cancer for 10 to 15 years following their discontinuation. Every effort should be made to encourage the use of oral contraceptives in all perimenopausal woman.

SUMMARY POINTS Use of the mnemonic PHIMIC (pregnancy, hormones, iatrogenic, mechanical, infection, and cancer) offers a simplified yet complete approach to the evaluation of abnormal uterine bleeding. Patients presenting with abnormal uterine bleeding should have a pregnancy test, luteinizing hormone, follicle-stimulating hormone, thyroid-stimulating hormone, prolactin, Pap smear, DNA probe for gonorrhea and chlamydia, and a transvaginal ultrasound. Estrogenized anovulation is the most common cause of abnormal uterine bleeding. Anovulation may be caused by a variety of conditions, including elevated prolactin, hyper- or hypothyroidism, adrenal disorders, and polycystic ovarian syndrome (hyperandrogenic anovulation). Von Willebrand disease should be considered in any young woman experiencing menorrhagia within 1 year of menarche. No therapeutic advantage of intravenous estrogen over oral estrogen in the control of acute hemorrhage is known. Saline infusion sonography and endometrial biopsy, when performed together, have a 97% sensitivity and 70% specificity in determining intrauterine pathology. MRI can reliably differentiate between deep and superficial adenomyosis. MRI can accurately identify leiomyomata less that 2 cm in size and determine their location. SUGGESTED READINGS Diagnosis 1.

Cote, I, Jacobs P, Cumming D. Work loss associated with increased menstrual loss in the United States. Obstet Gynecol 2002;100:683–687.

2.

Farquhar CM, Steiner CA. Hysterectomy rates in the United States 1990–1997. Obstet Gynecol 2002;99:229–234.

3.

Livingstone M, Fraser IS. Mechanisms of abnormal uterine bleeding. Hum Reprod Update 2002;222:619–627.

4.

Long C, Gast M. Menorrhagia in menstrual cycle disorders. In: London SN, Chihal HJ, eds. Obstet Gynecol Clin North Am 1990;17:343–359.

5.

Munro MG. Dysfunctional uterine bleeding: advances in diagnosis and treatment. Curr Opin Obstet Gynecol 2001;13:475–489.

6.

Robins JC. Therapies for the treatment of abnormal uterine bleeding. Curr Womens Health Rep 2001;1:196–201.

Pregnancy 7. Storring PL, Gaines-Das RE, Bangham DR. International reference preparation of human chorionic gonadotropin for immunoassay: potency estimates in various bioassays and protein binding assay systems; and international reference preparations of the alpha and beta subunits of human chorionic gonadotropin for immunoassay. J Endocrinol 1980;84:295. 8.

Stovall TG, Ling FW. Ectopic pregnancy: diagnostic and therapeutic algorithms minimizing surgical intervention. J Reprod Med 1993;38:807.

Hormones 9.

ACOG Committee Opinion. Von Willebrands disease in gynecologic practice. Int J Gynaecol Obstet 2002;76:336–337.

10.

Ayala AR, Danese MD, Ladenson PW. When to treat mild hypothyroidism. Endocrinol Metab Clin North Am 2000;29:399–415.

11.

Falcone T, Desjardins C, Bourque J, et al. Dysfunctional bleeding in adolescents. J Reprod Med 1994;39:761–764.

12.

Freda PU, Wardlaw SL. Clinical review 110: diagnosis and treatment of pituitary tumors. J Clin Endocrinol Metab 1999;84:3859–3866.

13.

Goodman-Gruen D, Hollenbach K. The prevalence of von Willebrand disease in women with abnormal uterine bleeding. J Womens Health Gend Based Med 2001;10:677–680.

14.

Taylor AE. Insulin-lowering medications in polycystic ovary syndrome. Obstet Gynecol Clin North Am 2000;27:583–595.

Iatrogenic 15.

Shwayder JM. Pathophysiology of abnormal uterine bleeding. Obstet Gynecol Clin North Am 2000;27:219–234.

Mechanical 16.

DeWaay DJ, Syrop CH, Nygaard IE, et al. Natural history of uterine polyps and leiomyomata. Obstet Gynecol 2002;100:3–7.

17.

Myers ER, Barber MD, Gustilo-Ashby T, et al. Management of uterine leiomyomata: what do we really know. Obstet Gynecol 2002;100:8–17.

18.

Santen RJ. To block estrogen's synthesis or action: that is the question. J Clin Endocrinol Metab 2002;87:3007–3012.

Infection 19.

Ferenczy A, Gelfand M. The biologic significance of cytologic atypia in progestogen-treated endometrial hyperplasia. Am J Obstet Gynecol 1989;160:126–131.

20.

Pizarro AR, Jamison R, Matthews-Greer J, et al. Organism-specific chronic endometritis is not a cause of abnormal uterine bleeding. J Pelvic Surg 7:147–150.

Cancer 21.

Clark T, Mann CH, Shah N, et al. Accuracy of outpatient endometrial biopsy in the diagnosis of endometrial cancer: a systematic quantitative review. Br J Obstet Gynecol 2002;109:313–321.

22.

Clark TJ, Mann CH, Shah N, et al. Accuracy of outpatient endometrial biopsy in the diagnosis of endometrial hyperplasia. Acta Obstet Gynecol Scand 2001;80:784–793.

23.

Clark TJ, Voit D, Gupta JK, et al. Accuracy of hysteroscopy in the diagnosis of endometrial cancer and hyperplasia: a systematic quantitative review. JAMA 2002;288:1610–1621.

Imaging and Special Studies in the Management of Abnormal Uterine Bleeding

24.

Bradley LD, Falcone T, Magen AB. Radiographic imaging techniques for the diagnosis of abnormal uterine bleeding. Obstet Gynecol Clin North Am 2000;2:245–276.

25. de Vries LD, Dijkhuizen PHLJ, Mol BWJ, et al. Comparison of transvaginal sonography, saline infusion sonography, and hysteroscopy in premenopausal women and abnormal uterine bleeding. J Clin Ultrasound 2000;28:217–223. 26. Dueholm M, Lundorf E, Olesen F. Imaging techniques for evaluation of the uterine cavity and endometrium in premenopausal patients before minimally invasive surgery. Obstet Gynecol Surv 2002;57:388–403. 27.

Mihm LM. The accuracy of endometrial biopsy and saline sonohysterography in the determination of the cause of abnormal uterine bleeding. Am J Obstet Gynecol 2002;186:858–860.

Management of Special Situations Acute Hemorrhage 28.

DeVore GR, Owens O, Kase N. Use of intravenous Premarin in the treatment of dysfunctional uterine bleeding—a double-blind randomized control study. Obstet Gynecol 1982;59:285–291.

Menorrhagia 29.

Fedele L, Bianchi S, Raffaelli R, et al. Treatment of adenomyosis-associated menorrhagia with a levonorgestrel releasing intrauterine device. Fertil Steril 1997;68:426–429.

Abnormal Bleeding on Oral Contraceptives 30.

Jones GL. Health-related quality of life measurement in women with common benign gynecologic conditions: a systematic review. Am J Obstet Gynecol 2002;187:501–511.

Abnormal Bleeding in Perimenopause 31.

Jensen JT. Health Benefits of Oral Contraceptives. Obstet Gynecol Clin North Am 2000;27:705–721.

Chapter 36 Premenstrual Syndrome Danforth’s Obstetrics and Gynecology

Chapter 36 Robert L. Reid

Premenstrual Syndrome

MOLIMINA, PREMENSTRUAL SYNDROME (PMS), AND PREMENSTRUAL DYSPHORIC DISORDER (PMDD) Diagnosis Etiology Therapy Medical Interventions SUGGESTED READINGS Introduction, Diagnosis, and Etiology Therapy: Education, Counseling, and Lifestyle Modification Antidepressant Therapy Medical Ovarian Suppression Surgical Therapy

MOLIMINA, PREMENSTRUAL SYNDROME (PMS), AND PREMENSTRUAL DYSPHORIC DISORDER (PMDD) During the reproductive years, up to 80% to 90% of menstruating women will experience symptoms (breast pain, bloating, acne, constipation) that forewarn them of impending menstruation, so-called molimina. Available data suggest that as many as 30% to 40% of these women are sufficiently bothered by such molimina that they would seek relief if it were readily accessible, simple, and safe. Since the term premenstrual syndrome, or PMS, has become so ingrained in lay culture most women describe the symptoms of molimina as PMS. However moliminal symptoms show a variable association with the more severe psychological symptoms typically required to meet research diagnostic criteria for PMS. Researchers in the field have argued that the term PMS should be reserved for a more severe constellation of symptoms that affects closer to 5% of women during their reproductive years. To capture the true severity of PMS in this population, the definition describes not only the symptoms but also their functional impact. PMS has been defined as “the cyclic recurrence in the luteal phase of the menstrual cycle of a combination of distressing physical, psychological, and/or behavioral changes of sufficient severity to result in deterioration of interpersonal relationships and/or interference with normal activities.” The American Psychiatric Association, following years of debate about whether to include PMS in their Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), included as an appendix (meaning that the subject requires further study) a condition labeled premenstrual dysphoric disorder (PMDD) ( Table 36.1). Such a designation was intended to alert physicians to a combination of specific severe symptoms (mostly psychiatric in nature) that could mimic other psychiatric disorders with the exception that the symptoms recurred in the premenstrual phase of the menstrual cycle.

TABLE 36.1. Diagnostic criteria for premenstrual dysphoric disorder (PMDD)

Diagnosis History and Charting A typical woman suffering from PMS may describe herself as a well-adjusted person for most of the month who is productive at work and in other endeavors, and, if she has children, a good mother. However starting 7 to 10 days prior to menstruation she awakes in the morning with feelings of anger, anxiety, or sadness. At work she may have difficulty concentrating on the tasks at hand and she may overreact to otherwise typical actions of co-workers, friends, partner, or children. She feels depressed but she cannot understand why because she typically enjoys life and is happy with most of its aspects. Occasionally depression, anger and aggression, or anxiety may be extreme and result in concerns for the welfare of the affected woman or those around her. A “black and white” diagnostic cutoff for determining if someone suffers from PMS is not presently possible. Although many women report feeling bloated and irritable before menstruation, further questioning usually reveals that these symptoms have little substantive impact on their lives. In part, the distinction between troublesome premenstrual moliminal symptoms and PMS may have to do with the duration and severity of symptoms. Particularly in the first few years after symptoms appear, the severity of PMS may vary dramatically from month to month. The nature of symptoms and their functional impact is best established by self-report using a prospective calendar record. Any calendar used for this purpose must obtain information on four key areas: specific symptoms, severity, timing in relation to the menstrual cycle, and baseline level of symptoms in the follicular phase. Retrospective accounts about symptoms often fail to provide critical information on timing of symptoms or baseline level of symptoms during the follicular phase of the cycle. For example, many individuals with underlying psychiatric disease will experience premenstrual exacerbation of their symptoms. They will recall the severe symptoms prior to menstruation and may give a history that sounds typical for PMS. However when they complete a 2-month calendar record it usually becomes apparent that symptoms are present all the time and are at their worst premenstrually. This has been termed “premenstrual magnification” of an underlying psychiatric disorder. Typically PMS symptoms appear after ovulation, worsen progressively leading up to menstruation, and are relieved at varying rates after onset of menstruation. In some women there is almost immediate relief from psychiatric symptoms with the onset of bleeding while for others the return to normal is more gradual. The most severely affected women report symptoms beginning shortly after ovulation (2 weeks before menstruation) and resolving at the end of menstruation. Such individuals typically report having only one “good week” per month. If this pattern is longstanding then it becomes increasingly difficult for interpersonal relationships to rebound during the good week with the result that symptoms may start to take on the appearance of a chronic mood disorder. (Whenever charting leaves the diagnosis in doubt, a 3-month trial of medical ovarian suppression [discussed later in the chapter] will usually provide a definitive answer.) About 5% to 10% of women who suffer from PMS experience a brief burst of symptoms (coincident with the midcycle fall in estradiol that accompanies ovulation) with a return to normal until later in the luteal phase when symptoms recur ( Fig. 36.1). The different patterns of PMS symptoms are shown in Figure 36.2.

FIG. 36.1. Circulating gonadotropin, estradiol, and progesterone concentrations correlated to symptom severity in a woman with midcycle and premenstrual syndrome (PMS) symptoms—so-called pattern C PMS. (From Reid RL. Premenstrual syndrome. In: DeGroot L, ed. Female Reproductive Endocrinology. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.)

FIG. 36.2. Four common patterns of premenstrual syndrome symptomatology in relation to the menstrual cycle. Note that in every case symptoms commence after ovulation. (From Reid RL. Premenstrual syndrome. In: DeGroot L, ed. Endotext.com. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.) Information should be sought about stresses related to the woman's occupation and personal life as these may tend to exacerbate PMS. Past medical and psychiatric diagnoses may be relevant in that a variety of medical and psychiatric disorders may show premenstrual exacerbation. A prospective symptom record should usually be maintained for a minimum of two complete cycles prior to establishing a diagnosis since an unfavorable life event occurring during the follicular phase of a single month could give the erroneous impression that symptoms were continuously present. Key components of a prospective symptom record are listed in Table 36.2.

TABLE 36.2. Key elements of a prospective symptom record used for the diagnosis of premenstrual syndrome One example of such a calendar record, the PRISM calendar (prospective record of the impact and severity of menstrual symptoms) ( Fig. 36.3) allows rapid visual confirmation of the nature, timing, and severity of menstrual cycle–related symptomatology and at the same time provides information on life stressors and use of PMS therapies. Although symptoms are rated in severity on a scale from 1 to 3, the actual interpretation of the calendar requires no mathematical calculations. An arm's length assessment of the month-long calendar usually allows a rapid distinction to be made between PMS and other more chronic conditions ( Fig. 36.4).

FIG. 36.3. One example of a prospective record of the impact and severity of menstrual symptoms—the PRISM calendar. (From Reid RL. Premenstrual syndrome. In: DeGroot L, ed. Endotext.com. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.)

FIG. 36.4. Arm's length assessment of a completed PRISM (prospective record of the impact and severity of menstrual symptoms) calendar reveals a pattern of symptoms consistent with premenstrual syndrome. (From Reid RL. Premenstrual syndrome. In: DeGroot L, ed. Endotext.com. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.) Physical Findings History should usually be accompanied by physical and gynecologic exam since many women with apparent PMS may have coexisting medical conditions. The summation of premenstrual and menstrual symptoms often determines the overall experience of women who suffer from PMS and therapy directed at gynecologic disorders such as dysmenorrhea (associated with endometriosis) or heavy flow (associated with fibroids) can dramatically reduce the perceived severity of PMS. Organic causes of PMS-like symptoms must be ruled out. Marked fatigue may result from anemia, leukemia, hypothyroidism, or diuretic-induced potassium deficiency. Headaches may be due to intracranial lesions. Women attending PMS clinics have been found to have brain tumors, anemia, leukemia, thyroid dysfunction, gastrointestinal disorders, pelvic tumors including endometriosis, and other recurrent premenstrual phenomena such as arthritis, asthma, epilepsy, and pneumothorax. Efforts to rule out such causes for premenstrual distress are important before choosing any PMS therapy. Blood Tests Contrary to some claims there is no blood test that will establish a diagnosis of PMS. Blood work is only helpful to rule out conditions such as anemia, leukemia, or thyroid dysfunction. Etiology Many theories have attempted to explain the diverse manifestations of PMS but until recently no single theory has received widespread acceptance. Many of the early theories lacked biologic plausibility and appeared to have emerged as a means to market specific therapeutic products ( Table 36.3). More recently a theory that links gonadal steroid levels and central serotonergic activity has started to emerge.

TABLE 36.3. List of proposed etiologies for premenstrual syndrome

The current consensus seems to be that in some predisposed women normal fluctuations in the gonadal hormones estrogen and progesterone trigger central biochemical events related to PMS symptomatology. Of all the neurotransmitters studied to date, serotonin seems to be the most promising central target. It is likely that serotonin levels dip premenstrually in most women and that susceptible individuals will show varying degrees of psychiatric symptomatology. Many women report a menstrual or immediate postmenstrual state of mental well-being (sometimes bordering on euphoria) that may reflect a rebound surge of serotonin activity. The theoretic effects of gonadal hormones on serotonin activity ( Fig. 36.5) form the basis for most of the current interventions for PMS and PMDD.

FIG. 36.5. A hypothetical depiction of the interrelationship between gonadal steroid fluctuations and central changes in serotonin activity to explain the timing of symptoms in premenstrual syndrome. When serotonin levels or activity fall below an arbitrary level (that may be influenced by stress, heredity, or other factors) symptoms of anger, anxiety, or depression may emerge. (From Reid RL. Premenstrual syndrome. Solid line, estrogen; dotted line, progesterone. In: DeGroot L, ed. Endotext.com. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.)

Several lines of evidence from clinical medicine support this interrelationship between estrogen or lack of estrogen effect (perhaps mediated by progestin-induced depletion of estrogen receptors) and central serotonergic activity. The midcycle appearance of PMS symptoms in women with pattern C PMS coincides with the periovulatory fall in estradiol, whereas the reappearance and ever-increasing severity of symptoms in the late luteal phase may result from progesterone-induced depletion of central estrogen receptors coincident with the decline of estradiol leading up to menstruation (see Fig. 36.1). Estrogen has been shown to alleviate clinical depression in hypoestrogenic perimenopausal women in double-blind clinical trials. The addition of sequential progestin therapy to estrogen replacement triggers characteristic PMS-like mood disturbance in some susceptible postmenopausal women. Anti-estrogens given for ovulation induction may, at times, provoke profound mood disruption. Women with premenstrual syndrome show a surprisingly high frequency of premenstrual and menstrual hot flashes (85% of those with PMS vs. 15% of non-PMS controls) that are typical of those experienced by menopausal women. Selective serotonin reuptake inhibitors (SSRIs) have been shown to relieve hot flashes in breast cancer survivors made menopausal by chemotherapy. In each of these circumstances, a decrease in exposure to estrogen has been linked to mood disturbance and in each case a decrease in serotonin activity (inferred from the response to SSRIs) appears to be the proximate cause. Therapy Education and Counseling Women who present with severe premenstrual complaints often benefit from reassurance that they are not “losing their minds” and that PMS is a real condition that affects many women of reproductive age. Reassurance that, although we do not yet have all the answers about PMS, we have available a range of very effective therapies is equally important information. The prospective documentation of symptoms is as much a part of therapy as it is the key to diagnosis (see Table 36.1). The very use of a calendar record (listing the range of PMS symptoms) may provide much needed reassurance that other women experience similar symptoms. The completed chart may demonstrate to the patient for the first time the cyclicity of her symptoms supporting a clear link to hormonal shifts of the menstrual cycle. While it is useful for women with PMS to learn to anticipate times in the month when vulnerability to emotional upset and confrontation may be greatest, the strategy of making important decisions “only on the good days,” as espoused in some PMS clinics, falls apart if premenstrual symptoms last for more than just a few days per month. For some women premenstrual symptoms may last for a full 3 weeks and advising them to restrict their important activities to the remaining days of the month is neither helpful nor warranted. Interventions aimed at reducing symptoms are more appropriate in this circumstance. Lifestyle Modification Communication Strategies When an individual is suffering to a degree that requires more than simple counseling and reassurance, measures aimed at lifestyle modification should first be explored. Women with such symptoms should be encouraged to discuss the problem with those individuals who are central to their life including partner, other family members, or friends. Often confrontations can be avoided if an understanding partner or friend recognizes the cause for a woman's upset and defers discussion of the controversial subject until another time. Strategies for stress reduction can be helpful. Communication skills and assertiveness may be improved with counseling. Group counseling in a program supervised by a clinical psychologist may be invaluable. Diet While there have been many books written describing specific “PMS diets” few of the recommendations contained therein are based on scientific fact. Several simple dietary measures may afford relief for women with PMS. Reduction of the intake of salt and refined carbohydrates may help prevent edema and swelling in some women. Although a link between methylxanthine intake and premenstrual breast pain has been suggested, available data are not convincing. Nevertheless, a reduction in the intake of caffeine may prove useful in women where tension, anxiety, and insomnia predominate. Anecdotal evidence suggests that small, more frequent meals may occasionally alleviate mood swings. Based on evidence that cellular uptake of glucose may be impaired premenstrually, there is, at least, some theoretic basis for this dietary recommendation. Several lines of evidence indicate that there is a tendency toward increased consumption of alcohol premenstrually and women should be cautioned that excessive use of alcohol is frequently an antecedent factor in relationship discord. Calcium supplementation to reduce symptoms has been shown to be marginally superior to placebo in a randomized placebo-controlled trial. Exercise Exercise is reported to reduce premenstrual molimina in women running in excess of 50 km per cycle. The effects of exercise in women with well-characterized PMS have not been rigorously evaluated. As part of an overall program of lifestyle modification, exercise may reduce stress by providing a time away from the home and by providing a useful outlet for any anger or aggression. Some women who suffer from PMS report that exercise promotes relaxation and helps them sleep at night. Medical Interventions The primary factor directing the selection of therapy should be the intensity and impact of premenstrual symptoms. Symptoms that are causing major disruption to quality of life rarely respond to lifestyle modification alone and efforts to push this approach often do nothing more than delay effective therapy. Conversely, minor symptoms or symptoms that are short-lived each month seldom justify major medical interventions. Attention should always initially be directed to symptoms for which simple, established treatments exist. For example, dysmenorrhea or menorrhagia may be satisfactorily relieved with prostaglandin synthetase inhibitors or oral contraceptives. Mefenamic acid (500 mg t.i.d.) in the premenstrual and menstrual weeks has outperformed placebo for the treatment of PMS in some, but not all, clinical trials. It is likely that many of the end-stage mediators of PMS symptomatology are prostaglandins, which indicates this prostaglandin synthetase inhibitor may be working through a general inhibition of prostaglandin activity. Based on this, mefenamic acid is an ideal adjunct for any woman with coexisting dysmenorrhea and menorrhagia. In practice however, its effectiveness for premenstrual symptomatology, particularly psychological symptoms, seems quite variable. Mefenamic acid is contraindicated in women with known sensitivity to aspirin or those at risk for peptic ulcers. Trials comparing oral contraceptive therapy to placebo have not shown a beneficial effect on mood in most circumstances. However, when contraception is required in a woman with PMS and coexisting dysmenorrhea or menorrhagia, the net effect of an oral contraceptive pill may be positive. Although oral contraceptives by no means guarantee relief from PMS they may afford sufficient relief from associated physical symptoms, so that for some women, the remaining premenstrual manifestations become tolerable. When an oral contraceptive user presents with PMS it is common to find that symptoms begin earlier in the cycle. This effect may result from exposure to progestin in the oral contraceptive 3 weeks before onset of menstruation rather than progesterone exposure 2 weeks before menses as occurs in natural cycles. In the woman presenting with distressing PMS while on an oral contraceptive, a switch to an alternate means of contraception is often attended by a substantial reduction in symptoms. Published data in regard to the efficacy of pyridoxine (vitamin B 6 ) have been contradictory, however, this medication in proper doses (100 mg daily) is, at worst, a safe placebo that becomes one part of an overall management plan for women with distressing molimina that should include lifestyle modification and changes in diet. Patients should be cautioned that these medications do not work for all women and that increasing the dose of pyridoxine in an effort to achieve complete relief of symptoms may lead to peripheral neuropathy. Pyridoxine should be discontinued if there is evidence of tingling or numbness of the extremities. Neither progestin therapy nor oil of evening primrose has been shown to be efficacious for PMS in controlled clinical trials. Premenstrual mastalgia which affects up to 10% of women in reproductive age may occur in isolation from other PMS symptoms and, as such, should be considered a moliminal symptom. Low-dose danazol (100 mg OD) or luteal phase–only danazol (200 mg OD) can bring about dramatic relief of mastalgia in most women, however higher doses (400 mg OD) may be required to relieve other PMS symptoms. Mastalgia may also respond to tamoxifen (10 mg daily), but has not been shown to

respond to diuretics, medroxyprogesterone acetate, or pyridoxine. The routine use of diuretics in the treatment of PMS should be abandoned. Most women show only random weight fluctuations during the menstrual cycle despite the common sensation of bloating. In the absence of demonstrable weight gain it is likely that this symptom may result from constipation or bowel wall edema rather than from an overall fluid accumulation. In rare cases, ingestion of salt and refined carbohydrates has been shown to result in true fluid retention. In cases where a consistent increase in weight can be documented or where edema is demonstrable, limitation of salt and refined carbohydrate intake should be attempted first. If such dietary restrictions fail to relieve premenstrual fluid accumulation, use of a potassium-sparing diuretic, such as spironolactone, may be considered. Continued use of a diuretic activates the renin–angiotensin–aldosterone system resulting in rapid rebound fluid accumulation as soon as the diuretic is discontinued. Weight takes approximately 2 to 3 weeks to return to normal after discontinuation of a diuretic in some people. Unfortunately this leaves the affected woman with the impression that she has to have a diuretic to maintain normal fluid balance. Some women report overriding symptoms of anxiety, tension, and insomnia in the premenstrual week. New short-acting anxiolytics or hypnotics such as alprazolam or triazolam, respectively, may be prescribed sparingly for such individuals. Buspirone has also proven useful in preliminary trials. Estrogen withdrawal has been implicated in menstrually-related migraines and evidence indicates that estrogen supplementation commencing in the late luteal phase and continued through menstruation may alleviate headaches in some women. As discussed subsequently, if headaches are severe and are unrelieved by short-term estrogen supplementation they can often be controlled by intramuscular or oral sumatriptan therapy or by medical ovarian suppression with gonadotropin-releasing hormone (GnRH) agonists and continuous combined hormone replacement therapy. Antidepressant Therapy A range of newer antidepressant medications that augment central serotonin activity (SSRIs or selective serotonin and norepinephrine reuptake inhibitors [SNRIs]) have been shown to alleviate severe premenstrual syndrome. The theoretic effect of these agents on serotonin levels throughout the menstrual cycle is depicted in schematic fashion in Figure 36.6. Since these agents will also relieve endogenous depression, a pretreatment diagnosis, achieved by prospective charting, is very important. Practically speaking, many women who attend a gynecology clinic to seek relief from premenstrual symptoms express reservations about taking an antidepressant, particularly if a short-term end point (3–6 months away) is not likely. Long-term therapy may be required to control symptoms of PMS for women in their 30s until menopause. Rapid reappearance of PMS has been reported after cessation of SSRI therapy by premenopausal women.

FIG. 36.6. Hypothetical depiction of how drugs that elevate serotonin activity can alleviate premenstrual syndrome. Solid line, estrogen; dotted line, progesterone. (From Reid RL. Premenstrual syndrome. In: DeGroot L, ed. Endotext.com. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.) For patients in whom PMS psychiatric symptoms predominate, antidepressant therapy may provide excellent results. SSRIs, such as fluoxetine, sertraline, paroxetine, fluvoxamine, and velafaxine (an SNRI) have all been successfully employed. Symptom profiles may help in selecting the most appropriate agent (i.e., fluoxetine in patients where fatigue and depression predominate; sertraline if insomnia, irritability, and anxiety are paramount). SSRIs have been associated with loss of libido and anorgasmia, which are particularly distressing to this patient population. Appropriate pretreatment counseling is essential. Tricyclic antidepressants (TCAs) have not generally been effective with the exception of clomipramine, a TCA with strong serotonergic activity. Intolerance to the side effects of TCAs is common. Most women who suffer from PMS would prefer to medicate themselves only during the symptomatic phase of the menstrual cycle. Studies have demonstrated that luteal phase therapy may be effective for many women with PMS. Practically speaking, a trial of SSRI therapy should be commenced with continuous use. After a woman has determined the optimal dosage to achieve the desired response it is reasonable to test luteal phase–only therapy to determine if the benefit is maintained. Medical Ovarian Suppression Suppression of cyclic ovarian function may afford dramatic relief for the woman with severe and long-lasting symptoms ( Fig. 36.7). In each case therapy should be directed to the suppression of cyclic ovarian activity while ensuring a constant low level of estrogen sufficient to prevent menopausal symptomatology and side effects.

FIG. 36.7. Hypothetical depiction of how medical or surgical elimination of ovarian steroid fluctuations can stabilize central serotonin activity and eliminate premenstrual syndrome. (From Reid RL. Premenstrual syndrome. In: DeGroot L, ed. Endotext.com. Available at: http://www.endotext.com/. Accessed January 16, 2003, with permission.) As a rule, although oral contraceptives suppress ovulation they seldom relieve the psychiatric symptoms of PMS, perhaps because ovarian steroidogenesis is replaced by exogenous synthetic steroids. Danazol 200 mg b.i.d. will effect ovarian suppression in approximately 80% of women with prompt relief from symptoms. Higher doses up to 200 mg q.i.d. may be required in the remaining women in order to achieve complete menstrual suppression. Danazol, at a dose of 200 mg b.i.d., has relatively few side effects; these may include hot flashes, muscle cramps, and occasional cases of epigastric pain, fine tremor, or insomnia. In the woman without preexisting hair growth, hirsutism is rarely a problem at these dosages when treatment is limited to 1 year or less. In the individual with preexisting acne or increased body hair, alternative interventions should be considered. The use of danazol causes a shift to a more unfavorable lipid profile which is likely to have little impact when danazol is used on a short-term basis. In the patient who tolerates danazol well, and in whom symptomatic relief is dramatic, this effect is the primary concern that will influence decision making about long-term treatment. GnRH agonists effect rapid medical ovarian suppression thereby inducing a pseudomenopause and affording relief from PMS. This approach is unsatisfactory in the long term not only because of the troublesome menopausal symptoms it evokes but also because if creates an increased risk for osteoporosis and ischemic heart disease. When combined with continuous combined hormone replacement therapy, GnRH agonists afford excellent relief from premenstrual symptomatology without the attendant risks and symptoms resulting from hypoestrogenism. The major drawback to this therapeutic approach is the expense of medication and the need for the patient to take multiple medications on a long-term basis. A simpler and less expensive approach involves the use of depo-medroxyprogesterone acetate (DMPA). This approach results in rapid suppression of cyclic ovarian function without attendant menopausal symptomatology. The major drawback to this approach is that a substantial percentage of women will experience irregular bleeding and gradual weight gain. The ideal situation, in which DMPA induces amenorrhea, may be problematic for the woman who wishes future fertility. Patients should always be counseled about the potential for protracted anovulation following use of this medication. Surgical Therapy The fact that medical approaches to PMS should be exhausted prior to considering surgery (hysterectomy and oophorectomy) for debilitating PMS has been emphasized in an American College of Obstetricians and Gynecologists' committee opinion on premenstrual syndrome (No. 155, April 1995). Clinical trials, however, have clearly shown this therapy to be effective. For the woman in whom there is unequivocal documentation that premenstrual symptoms are severe and disruptive to lifestyle and relationships, and in whom conservative medical therapies have failed (either due to lack of response, intolerable side effects, or prohibitive cost), the effect of medical ovarian suppression should be tested. At times this therapeutic approach (a GnRH agonist and continuous combined hormone replacement therapy) can be maintained until menopause with satisfactory symptom control. Some women, despite complete relief of symptoms, cannot afford or choose not to take this combination of medications for prolonged intervals (as long as 10–15 years from diagnosis until menopause in some cases). In these specific circumstances a surgical option may be considered. In the circumstance where family is complete and permanent contraception is desired, the pros and cons of oophorectomy for lasting relief from premenstrual symptomatology should be discussed with the patient. In many women the progestin component of hormone replacement therapy, when given sequentially, may induce an apparent recrudescence of PMS-like symptoms and, when given continuously, may result in unwanted irregular bleeding. Accordingly, hysterectomy at the time of oophorectomy is a consideration that allows subsequent hormone replacement with low-dose estrogen alone. SUGGESTED READINGS Introduction, Diagnosis, and Etiology 1.

American College of Obstetricians and Gynecologists. Premenstrual syndrome. ACOG Committee Opinion No. 155, April 1995.

2.

Case AM, Reid RL. Effects of the menstrual cycle on medical disorders [Review]. Arch Intern Med 1998;158:1405–1412.

3.

Diamond M, Simonson CD, DeFronzo RA. Menstrual cyclicity has a profound effect on glucose homeostasis. Fertil Steril 1989;52:204–208.

4.

Hahn PM, Wong J, Reid RL. Menopausal-like hot flushes reported in women of reproductive age. Fertil Steril 1998;70:913–918.

5. Premenstrual syndrome and premenstrual dysphoric disorder: scope, diagnosis, and treatment. In: FW Ling, JF Mortola, SF Pariser, et al, eds. APGO Educational Series on Women's Health 1998. Washington, DC: Association of Professors of Obstetrics and Gynecology and Obstetrics, 1998. 6.

Reid RL. Premenstrual syndrome. Curr Prob Obstet Gynecol & Fertil 1985;8:1–57.

7.

Steiner M. Premenstrual dysphoric disorder: an update. Gen Hosp Psychiatry 1996;18:244–250.

Therapy: Education, Counseling, and Lifestyle Modification 8.

Budeiri D, Li Wan Po A, Dornan JC. Is evening primrose oil of value in the treatment of premenstrual syndrome? Control Clin Trials 1996;17:60–68.

9.

De Lignieres B, Vincens M, Mauvais-Jarvis P, et al. Prevention of menstrual migraine by percutaneous oestradiol. Br Med J 1986;293:1540.

10.

Kleijnen J, Ter Riet G, Knipschild P. Vitamin B6 in the treatment of the premenstrual syndrome—a review. Br J Obstet Gynecol 1990;97:847–852.

11.

Mira M, McNeil D, Fraser IS, et al. Mefenamic acid in the treatment of premenstrual syndrome. Obstet Gynecol 1986;68:395–398.

12.

Salonen R, Saiers J. Sumatriptan is effective in the treatment of menstrual migraine; a review of prospective studies and retrospective analyses. Cephalgia 1999;19:16–19.

13.

Steege JF, Blumenthal JA. The effects of aerobic exercise of premenstrual symptoms in middle aged women: a preliminary study. J Psychosom Res 1993;37:127–133.

14.

Wyatt K, Dimmock P, Jones P, et al. Efficacy of progesterone and progestogens in management of premenstrual syndrome: systematic review. BMJ 2001;323:776–780.

15.

Wyatt KM, Dimmock PW, Jones PW, et al. Efficacy of vitamin B6 in treatment of premenstrual syndrome: systematic review. BMJ 1999;318:1375–1381.

Antidepressant Therapy 16. Berger CP, Presser B. Alprazolam in the treatment of two subsamples of patients with late luteal phase dysphoric disorder: a double blind placebo controlled crossover study. Obstet Gynecol 1994;84:379–385. 17.

Dimmock PW, Wyatt KM, Jones PW, et al. Efficacy of selective serotonin-reuptake inhibitors in premenstrual syndrome: a systematic review. Lancet 2000;356:1131–1136.

18.

Rubinow DR, Schmidt PJ, Roca CA. Estrogen-serotonin interactions: implications for affective regulation. Biol Psychiatry 1998;44;839–850.

19.

Steiner M, Korzekwa M, Lamont J, et al. Fluoxetine in the treatment of premenstrual dysphoria. N Engl J Med 1995;332:1529–1534.

20.

Steiner M, Korzekwa M, Lamont J, et al. Intermittent fluoxetine dosing in the treatment of women with premenstrual dysphoria. Psychopharmacol Bull 1997;33:771–774.

Medical Ovarian Suppression 21.

Graham CA, Sherwin BB. A prospective treatment study of premenstrual symptoms using a triphasic oral contraceptive. J Psychosom Res 1993;36:257–266.

22.

Hahn PM, Van Vugt DA, Reid RL. A randomized placebo controlled crossover trial of danazol for the treatment of premenstrual syndrome. Psychoneuroendocrinology 1995;20:193–209.

23.

Mezrow G, Shoupe D, Spicer D, et al. Depot leuprolide acetate with estrogen and progestin add back for long-term treatment of premenstrual syndrome. Fertil Steril 1994;62:932–937.

24.

Muse KN, Cetel NS, Futterman L, Yen SSC. The premenstrual syndrome: effects of “medical ovariectomy.” N Engl J Med 1984;311:1345–1349.

Surgical Therapy 25.

Casper RF, Hearn MT. The effect of hysterectomy and bilateral oophorectomy in women with severe premenstrual syndrome. Am J Obstet Gynecol 1990;162:105–109.

26.

Casson P, Hahn P, VanVugt DA, et al. Lasting response to ovariectomy in severe intractable premenstrual syndrome. Am J Obstet Gynecol 1990;162:99–102.

Chapter 37 Androgen Excess Disorders Danforth’s Obstetrics and Gynecology

Chapter 37 Richard S. Legro and Ricardo Azziz

Androgen Excess Disorders

NORMAL ANDROGEN PRODUCTION AND METABOLISM Androgen Production and Action Androgen Clearance Bioavailability of Androgens Androgens, Age, and Menopause Androgens and Obesity SIGNS AND SYMPTOMS OF ANDROGEN EXCESS Androgen Excess and the Pilosebaceous Unit Androgens and Hypothalamic–Pituitary–Ovarian Axis Dysfunction Androgens and Hypothalamic–Pituitary–Adrenal Axis Dysfunction Virilization and Masculinization THE POLYCYSTIC OVARY SYNDROME Clinical, Biochemical, and Metabolic Features of PCOS Sequelae of PCOS OTHER ANDROGEN EXCESS DISORDERS The HAIRAN Syndrome Nonclassic Adrenal Hyperplasia Androgen-Secreting Tumors Rare Causes of Androgen Excess IDIOPATHIC HIRSUTISM EVALUATION OF ANDROGEN EXCESS Laboratory Evaluation of Androgen Excess Laboratory Assessment for Metabolic Abnormalities Radiologic Imaging in the Evaluation of the Patient with Androgen Excess TREATMENT OF ANDROGEN EXCESS Regulation of Uterine Bleeding Treatment of Hirsutism, or Androgen Excess-Associated Acne or Alopecia Treatment of Associated Metabolic Abnormalities Treatment of Hyperandrogenic Oligoovulatory Infertility SUMMARY POINTS SUGGESTED READINGS Normal Androgen Metabolism The Polycystic Ovary Syndrome Other Androgen Excess Disorders Evaluation of Androgen Excess Treatment of Androgen Excess

Androgen excess (i.e., hyperandrogenism) is a common, albeit heterogeneous, endocrine disorder of women. The clinical presentation of androgen excess may range from mild hirsutism, acne, or subtle ovulatory dysfunction, to the rare patient with frank virilization and masculinization. Androgen excess disorders include the polycystic ovary syndrome (PCOS), nonclassic adrenal hyperplasia (NCAH), the hyperandrogenic–insulin resistant–acanthosis nigricans (HAIRAN) syndrome, and androgen-secreting neoplasms. Although the most recognizable clinical feature of androgen excess is hirsutism, it should be noted that not all patients with hirsutism have evidence of androgen excess, as in the patient with idiopathic hirsutism (IH). Likewise, not all patients with an androgen excess disorder have clinically evident hirsutism, as in the Asian patient with PCOS. Because of its frequent association with PCOS, clinically evident androgen excess (e.g., hirsutism) is a useful marker for the presence of metabolic abnormalities in these women, including insulin resistance and glucose intolerance resulting in an increased risk for type 2 diabetes mellitus (DM) and cardiovascular disease (CVD). With these new insights, the burden of care for the treating physician has increased beyond that of solely treating the presenting complaint to include the detection and, if possible, prevention of these metabolic consequences.

NORMAL ANDROGEN PRODUCTION AND METABOLISM Androgen Production and Action Androgens are C19 steroids (that is they contain 19 carbons) produced from circulating low-density lipoprotein (LDL) cholesterol (a C27 molecule). As in most metabolic pathways, the first step is rate limiting (i.e., the conversion of cholesterol to the weak progestogen pregnenolone, a C21 steroid) by cytochrome P450scc ( Fig. 37.1). The zona reticularis of the adrenal cortex, and the theca and stroma of the ovaries, secrete androgens produced through de novo synthesis from cholesterol. In addition, circulating steroid precursors can be metabolized further in these organs or in peripheral tissues, including the liver, adipose tissue stroma, and the pilosebaceous unit (PSU) in skin, to more potent androgens (e.g., the conversion of testosterone to dihydrotestosterone [DHT]), to estrogens via the action of aromatase, or they can be inactivated and readied for excretion ( Fig. 37.2). The origins of the principal circulating plasma androgens in pre- and postmenopausal women are summarized in Table 37.1. Androstenedione is the most important precursor of testosterone and DHT, while dehydroepiandrosterone (DHEA) accounts for only 5% to 13% of circulating testosterone in normal women.

FIG. 37.1. Pathways of androgen synthesis. The ovary, the testis, and the adrenal gland produce six common core steroids. The core d5 steroids are pregnenolone, 17-hydroxypregnenolone, and dehydroepiandrosterone. The core d4 steroids are progesterone, 17-hydroxyprogesterone, and androstenedione (androstene-3,17-dione). The core steroids are important precursors for the production of sex steroids, glucocorticoids, and mineralocorticoids.

FIG. 37.2. Principal pathways of androgen metabolism. Enzymes: 1, 3ß-hydroxysteroid dehydrogenase; 2, 17ß-hydroxylase; 3, 17,20-lyase (17,20-desmolase); 4, 17ß-hydroxysteroid dehydrogenase (17-keto reductase); 5, aromatase; 6, 5a-reductase; 7, 5ß-reductase; 8, 3ß-oxoreductase; and 9, 3a-oxoreductase. a: In the gonads the 17ß-hydroxysteroid dehydrogenase reaction predominantly produces 17ß-dehydrogenated products (androstenediol and testosterone), while the reverse is true in peripheral tissues. b: Aldosterone, etiocholanolone, and dehydroepiandrosterone are the principal urinary metabolites of androgens. (From Azziz R. Reproductive endocrinologic alterations in female asymptomatic obesity. Fertil Steril 1989;52:703–725, with permission.)

TABLE 37.1. Percent origin of androgens in women

Androgens, either directly or through metabolites, can act systemically in a classic endocrine fashion, or locally in a paracrine and intracrine fashion (e.g., in the PSU). Androgens exert their genomic effects through interaction with the androgen receptor, a member of the nuclear receptor superfamily. The unbound androgen receptor is a cytoplasmic protein and, upon ligand binding, it translocates into the nucleus. In the nucleus it influences the transcription of target genes through a complex process which includes interactions with other transcription factors and coactivators. As noted, androgens can also exert their effects indirectly, though metabolites such as the aromatization to estrogens. Androgen Clearance The clearance of androgens is accomplished by hepatic extraction and peripheral metabolism, which are highly dependent on the unbound portion of circulating steroid. Approximately 10% of testosterone and 50% of androstenedione are metabolized peripherally in women. Clearance of androgens by the hepatic splanchnic circulation involves mainly catabolism via 5a- and 5ß-reductase. Androgen metabolites are further conjugated by the liver (95% glucuronic and 5% sulfuric), facilitating their urinary excretion (see Fig. 37.2). Some 15% of androgen sulfates are excreted in bile, of which 80% are reabsorbed in the gut. Peripheral metabolism of androgens also occurs in the various target tissues including skin, muscle, brain, and adipose tissue. A-ring (aromatization), 17ß-hydroxysteroid dehydrogenization, 5aand 5ß- A-ring reduction, and 3a- and 3ß-oxo-reduction give rise to potent androgens such as DHT, or to weaker metabolites such as 5a-androstane-3a,17ß-diol, androsterone, and etiocholanolone. These are subsequently conjugated by the liver and excreted in the urine and bile. Measurement of these metabolites (e.g., 3a-androstanediol glucuronide) was proposed as a circulating marker of peripheral androgen excess, but its clinical use currently is negligible. Bioavailability of Androgens Androgens circulate in the body bound by a variety of proteins, including albumin, cortisol-binding globulin, acid a2-glycoprotein and, most important, sex hormone binding globulin (SHBG, formerly known as testosterone or testosterone-estradiol binding globulin [TeBG]). Because of its much higher concentration and total amount, albumin has a much greater overall binding capacity for androgens than SHBG, but the affinity of androgens for SHBG is several orders of magnitude higher, thus binding the largest portion of circulating testosterone. Androgens bound to SHBG are essentially not bioavailable, although those complexed with albumin are more readily available for tissue interaction due to the much lower affinity of this protein for these steroids. The liver produces SHBG, and production is stimulated by estrogen, particularly oral forms; and inhibited by androgens, and most importantly, insulin. These factors lead to lower levels of SHBG and higher bioavailable androgens in men and in women with androgen excess disorders compared to healthy women. As noted in the following sections, the production, metabolism, and circulating levels of androgens are affected by age, menopausal status, and the presence of obesity, among other factors. Androgens, Age, and Menopause Androgen metabolism and circulating levels can be significantly altered by age. Most notably, adrenal androgen production clearly decreases with age in premenopausal and postmenopausal women. Serum dehydroepiandrosterone sulfate (DHEA-S) concentrations decrease linearly with age beginning at about 20 years, independent of menopause. An impaired secretion of 17 a-hydroxyprogesterone, 17 a-hydroxypregnenolone, DHEA, and androstenedione has also been observed in postmenopausal women following acute adrenal stimulation. In normal postmenopausal women plasma concentrations of androstenedione are about half that observed in premenopausal women, although there is no observable difference in the clearance rate of this steroid between premenopausal and postmenopausal women. After menopause androstenedione levels continue to gradually decrease with age. Testosterone levels are less affected by age and menopause, and it is clear that the ovary continues to produce a significant amount of testosterone in postmenopause. Most of the decrease in circulating testosterone with age and menopause is explained by the lower androstenedione levels. Androgens and Obesity In eumenorrheic obesity the production rate and metabolic clearance rate (MCR) of ovarian and adrenal-secreted androgens are increased while circulating levels remain normal. The increased MCR may be due to an obesity-related decrease in the plasma concentration of the carrier protein SHBG. Steroid sequestration by fat may also increase steroid clearance, leading to an extremely large pool of sex hormones in obese individuals. Adipose tissue metabolism of steroids, including aromatization and 17ß-hydroxysteroid dehydrogenation, also contributes to the increased MCR of androgens while alterations in hepatic conjugation and extraction can also be a contributing factor. The increased ovarian and adrenal production noted in obesity may reflect changes in the intraglandular or circulating concentration of androgens, estrogens, prolactin, growth hormone and other growth factors, and most importantly, insulin. Although weight reduction corrects any abnormality noted in steroid levels, it is not known whether the elevated production and MCRs of androgens also normalize following weight reduction. The degree of androgen excess that can result in the appearance of clinical features most commonly has its source in the increased production (or intake) of androgens, which can be altered by age and degree of obesity. However, other factors such as the potency of the androgen produced (DHT>testosterone>androstenedione>DHEA), the amount of androgen that is free or weakly bound in serum (i.e., the amount bioavailable), and the degree and type of central and peripheral metabolism also play a role in determining and modifying the phenotype of androgen excess.

SIGNS AND SYMPTOMS OF ANDROGEN EXCESS Androgen excess results in various clinical signs and symptoms, including abnormalities of the PSU (hirsutism, acne, and androgenic alopecia), the hypothalamic–pituitary–ovarian axis (i.e., ovulatory and menstrual dysfunction), and the hypothalamic–pituitary–adrenal axis (adrenal androgen excess). If the androgen excess is very severe, virilization or masculinization can also be apparent. Androgen Excess and the Pilosebaceous Unit The pilosebaceous unit (PSU) is the common skin structure that gives rise to both hair follicles and sebaceous glands, which are found everywhere on the body except

the palms, soles, and lips. The density is greatest on the face and scalp (400–800 glands/cm 2 ) and lowest on the extremities (50 glands/cm 2 ). The number of PSUs does not increase after birth (about 5 million), but they can become more prominent through activation and differentiation. Generally, three phases of the hair growth cycle can be considered. The period of active growth is termed anagen, after which the hair follicle enters a resting or catagen phase of varying length of time ( Fig. 37.3). During this transition the hair shaft separates from the dermal papillae at the base. The separated hair is then shed during the telogen phase. The period of anagen varies from 3 years on the scalp to 4 months on the face. For corresponding parts of the skin, men have longer anagen phases than women do, which may be partially due to their higher circulating androgen levels.

FIG. 37.3. Diagram of cyclic hair follicles, a club hair (white) in old telogen and early anagen (I–IV) follicles, and a new hair (black) in new anagen (V–VI) follicles. (From: Uno H. Seminars in reproductive endocrinology. Vol 4. New York: Thieme, 1986, with permission.)

The effects of androgens are most visible on the PSU. Androgens stimulate the transformation of fine, unpigmented vellus hairs to coarse, pigmented, thickened terminal hairs, a process known as terminalization, in skin areas sensitive to the effects of androgens. The peripheral effects of androgens are determined primarily by the intracellular actions of the enzymes 17ß-ketosteroid reductase (converting androstenedione to testosterone) and 5a-reductase (converting testosterone to the more potent androgen, DHT), and the androgen receptor content. Before puberty body hair is primarily composed of fine, short unpigmented vellus hairs. The increase in androgen production observed with pubertal development transforms some of these, mainly in androgen-sensitive areas of skin such as the axilla and the genital triangle, into the coarser, longer, pigmented terminal hairs. It should be noted that not all skin areas are androgen-sensitive. For example, the development of terminal hairs in body areas such as the eyebrows, eyelashes, and the temporal and occipital scalp is relatively androgen-independent. Excess androgen-dependent hair growth present in women can result in clinically evident hirsutism (see the following section). Paradoxically, androgens can exert opposite effects on the hair follicles of the scalp, causing conversion of terminal follicles to vellus-like follicles, a process termed miniaturization. This effect may lead to the development of androgenic alopecia in women (see “ Androgenic Alopecia”) or male-pattern baldness characterized by frontal and sagittal scalp hair loss. Androgens can also cause increased sebum production and abnormal keratinization of the PSU, contributing to the development of seborrhea and acne evident at puberty and in women with androgen excess. Ethnic and genetic differences can modify the effects of androgens on skin, as demonstrated by the lesser degree of hirsutism present in Asian women with PCOS ( Fig. 37.4).

FIG. 37.4. Features in women with polycystic ovary syndrome (75 women, 25 each from the United States, Italy, and Japan). Note the much lower prevalence of hirsutism among Japanese women. (From Carmina et al, Am J Obstet Gynecol 1992;167:1807–1812, with permission.) Hirsutism Hirsutism is the presence of terminal (coarse) hairs in females in a male-like pattern. Excessive growth of coarse hairs of the lower forearms and lower legs alone does not constitute hirsutism, although women suffering from hirsutism may note an increase in the pigmentation and growth rate of hairs on these body areas. Hirsutism should be viewed much as polycystic ovaries, as a sign rather than a diagnosis. Most commonly hirsutism is associated with androgen excess. Although the term “idiopathic hirsutism” was coined to identify the presence of hirsutism without other identifiable cause or abnormality, this may actually reflect our limited ability to assess androgen action in the peripheral compartment or even in the circulation (see Evaluation of Androgen Excess). Acne The PSU, in addition to the hair follicle, also contains a sebaceous gland that produces an oily protective secretion, sebum. The excessive production of sebum, in response to androgen action, may lead to oily skin, clogged hair follicles, folliculitis, and the development of acne. Elevations in serum androgen levels have been noted in patients with acne, particularly in those with concurrent hirsutism, although not all investigators agree. Although the persistence or appearance of acne in adulthood has been suggested to be more frequently associated with androgen excess, we observed evidence of hyperandrogenemia in the majority of 30 consecutive nonhirsute, acneic patients regardless of age. Some investigators have observed that while serum androgen levels may be relatively normal in acneic patients, the conversion of androgens to DHT via 5a-reductase was increased in the affected skin. These data suggest that androgen suppression may be useful in treating acne in many of these patients, and it is clear that acne frequently improves following treatment with antiandrogens, oral contraceptives, or glucocorticoid suppression. Thus, empirical treatment with oral contraceptives, or even glucocorticoids, may be justified in acneic patients without other evidence of overt androgen excess. Androgenic Alopecia Scalp hair loss as a consequence of androgen excess can take two forms. In severe cases, where massive androgen excess and virilization/masculinization are present, patients can demonstrate the typical pattern of balding found in men (i.e., premature male-pattern balding). More common, however, is the so-called androgenic (also termed androgenetic, as an inherited etiology is often suspected) alopecia of women (i.e., female-pattern balding). In female androgenic alopecia a diffuse thinning of hair throughout the sagittal scalp is primarily noted and approximately 40% of women with androgenic alopecia have some form of hyperandrogenemia. However, if only nonhirsute women with androgenic alopecia are considered, then only 20% of these patients are found to be hyperandrogenemic. Androgens and Hypothalamic–Pituitary–Ovarian Axis Dysfunction Androgens, indirectly, through conversion to estrogens, and directly may alter the secretion of gonadotropins in women. However in most patients without severe forms of androgen excess, the direct effect of androgens on the hypothalamic–pituitary–ovarian axis, independent of their aromatization to estrogens, appears to be limited. Excessive androgen levels may also directly inhibit follicular development at the ovarian level, which may result in the accumulation of multiple small cysts within the ovarian cortex, the so-called “polycystic” ovary. Androgens and Hypothalamic–Pituitary–Adrenal Axis Dysfunction Adrenal androgen excess (i.e., elevated levels of DHEA and DHEA-S, and of the adrenal fraction of androstenedione) is a concomitant finding in many women with androgen excess. However, it is also possible that extraadrenal androgens (e.g., ovarian) may alter adrenocortical steroidogenesis and androgen secretion. For example, a 20% to 25% decrease in mean DHEA-S levels follows long-acting gonadotropin-releasing hormone-a (GnRH-a) suppression in women with PCOS with elevated levels of this adrenal androgen, although elevated adrenal androgen levels in these women rarely normalize with GnRH-a suppression. Thus, it appears that the secretion of adrenal androgen can be increased by extraadrenal hyperandrogenemia, although the clinical relevance of such an effect is unclear. Virilization and Masculinization Virilization includes the appearance of sagittal and frontal balding, clitoromegaly, and severe hirsutism. Furthermore, if androgen levels are extremely elevated for a substantial period of time the features of virilization may be accompanied by masculinization of the body habitus, with atrophy of the breasts, an increase in muscle mass, a redistribution of body fat, and a deepening of the voice. Premenopausal patients with virilization or masculinization are almost always amenorrheic. In general, virilization or masculinization should raise the suspicion of an androgen-secreting neoplasm or classic (but not nonclassic, as discussed later) adrenal hyperplasia. Occasionally girls suffering from severe insulin-resistance syndrome (discussed later in the chapter) may exhibit a moderate degree of virilization.

THE POLYCYSTIC OVARY SYNDROME By far the most common, although the least understood, cause of androgen excess is polycystic ovary syndrome (PCOS), accounting for a vast majority of patients seen. PCOS affects 4% to 6% of women in the developed world. There is no firm consensus as to the definition of PCOS; criteria arising from a 1990 National

Institutes of Health conference on the subject have proven clinically and investigationally useful. These criteria note that PCOS should include the presence of: clinical or biochemical evidence of hyperandrogenism ovulatory dysfunction the exclusion of other causes of androgen excess or ovulatory dysfunction, such as including adrenal hyperplasia, hyperprolactinemia, thyroid dysfunction, and androgen-secreting neoplasms. Thus, PCOS represents unexplained functional hyperandrogenic chronic anovulation and is a diagnosis of exclusion. Other diagnostic criteria for PCOS have been proposed, including the finding of a polycystic ovary morphology on ultrasound, with multiple 2-to 8-mm subcapsular preantral follicles forming a “black pearl necklace” sign ( Fig. 37.5). However, it should be noted that “polycystic ovaries” on sonography or at pathology are simply a sign of androgen excess and possibly PCOS. For example, this ovarian morphology is frequently seen in patients with adrenal hyperplasia ( Table 37.2), and up to 25% of unselected women have polycystic ovaries on ultrasound, many of which are normoandrogenic with regular menstrual cycles. Hence, we consider the appearance of polycystic ovaries, particularly on sonographic exam, to be a sign, albeit nondiagnostic, of androgen excess and PCOS.

FIG. 37.5. Transvaginal ultrasound visualization of a polycystic ovary. Note the string of subcapsular follicles measuring 2 to 8 mm in diameter, with increased central stromal mass.

TABLE 37.2. Syndromes or diseases entities associated with polycystic ovaries

Clinical, Biochemical, and Metabolic Features of PCOS Approximately 70% to 80% of women with PCOS demonstrate frank elevations in circulating androgens, particularly free testosterone, and 25% to 50% will have elevated levels of the adrenal androgen metabolite, DHEA-S. Prolactin levels are usually normal, although they may be slightly elevated (generally 80 µU/mL) or glucose-stimulated (>300–500 µU/mL) insulin levels, although exact diagnostic criteria are lacking, in part due to the heterogeneity of the syndrome. Patients can be severely hyperandrogenemic, with testosterone levels reminiscent of patients with androgen-secreting neoplasms, resulting in the development of severe hirsutism and even virilization. In addition to significant hirsutism, patients also develop the characteristic dermatologic finding of acanthosis nigricans (see Fig. 37.6). Because of their high androgen levels, gonadotropin levels in these patients may be somewhat suppressed resulting in persistent endometrial atrophy and amenorrhea in spite of the administration of a cyclic progestogen or an oral contraceptive. Because of the mitogenic effect of insulin on ovarian theca cells, the ovaries of many patients with the HAIRAN syndrome will become hyperthecotic. On ultrasound and histology the ovaries morphologically have a paucity of cortical cysts, and demonstrate a thickened and enlarged theca/stroma compartment. In addition, these patients are at significant risk for dyslipidemia, type 2 DM, hypertension, and CVD. These patients can be particularly difficult to treat although the selected use of long-acting GnRH analogs has been promising. Some patients may necessitate surgery, either ovarian wedge resection or oophorectomy. Nonclassic Adrenal Hyperplasia Nonclassic adrenal hyperplasia (NCAH), also referred to as late-onset congenital adrenal hyperplasia, is a homozygous recessive disorder due to mutations in the CYP21 gene which results in an abnormal (or absent) cytochrome P450c21 with relatively deficient 21-OH activity. Overall, between 1% and 8% of women with androgen excess have 21-OH–deficient NCAH depending on ethnicity, with the highest rates reported among Ashkenazi Jews. Due to the lack of 21-hydroxylation the progestogenic precursors to cortisol, 17 a-hydroxyprogesterone (17 a-HP) and to a certain degree 17 a-hydroxypregnenolone, accumulate in excess. These steroids are then metabolized to C19 products, principally androstenedione and testosterone (see Fig. 37.1, Fig. 37.2). However, clinically and biochemically these patients are very difficult to distinguish from other hyperandrogenic patients, particularly patients with PCOS. Patients with NCAH may present only with persistent acne or may have moderate degrees of hirsutism and oligoamenorrhea. Frank virilization or even severe hirsutism is relatively rare. The levels of the exclusive adrenal androgen metabolite DHEA-S are not any higher than that of other hyperandrogenic women. Although the frequency is relatively low, all patients with unexplained androgen excess should be screened for NCAH due to CYP21 mutations, as this diagnosis has a different prognosis, a different treatment regimen, and requires genetic counseling regarding the risks of congenital transmission. The measurement of a basal 17 a-HP in the follicular phase and in the morning can be used to screen for this disorder. In addition, these patients can be treated with corticosteroid replacement, although many of the patients who are diagnosed in adulthood tend to also demonstrate a PCOS-like pattern requiring additional ovarian suppression. While it is theoretically possible that NCAH may be due to mutations in other genes determining steroidogenic enzymes, such CYP11B1 (encoding for P450c11a and 11ß-hydroxylase (CYP11B1) or HSD3B (for 3ß-hydroxysteroid dehydrogenase), mutations in the genes coding for these enzymes are rarely noted in adult women presenting with androgen excess. Androgen-Secreting Tumors Androgen-secreting neoplasms are relatively rare, affecting between 1 of 300 and 1 of 1,000 hirsute patients. These tumors usually originate in the ovary, and rarely the adrenal cortex. The most common androgen-producing tumor in a premenopausal woman is a Sertoli-Leydig cell tumor, with thecomas and hilus cell tumors being less frequent. Hilus cell tumors are often small, and can be less than 1 cm in diameter, theoretically below the range of ultrasound visualization. Another rare neoplasm may be a granulosa cell tumor, 10% of which primarily secrete androgens instead of estrogens. Also, any large tumor within the body of the ovary (i.e., benign cystic teratomas, dysgerminomas, epithelial tumors) can produce androgens indirectly by causing hyperplasia of the surrounding normal stroma. Fortunately, the vast majority of ovarian androgen-secreting neoplasms are benign.

Alternatively, adrenal androgen-secreting tumors are generally malignant (i.e., adrenocortical carcinoma) and associated with a high mortality rate, although androgen-secreting adrenal adenomas have also been reported. Fortunately, adrenal androgen-secreting neoplasms are rare with an estimated incidence of two cases per one million persons per year. The age of onset in adults peaks in the fifth decade. Virilization can accompany both tumors primarily producing androgens and tumors primarily producing cortisol (Cushing syndrome). A long history of symptoms, as in the case with an ovarian tumor does not exclude the presence of an adrenocortical neoplasm. The presence of androgen-secreting neoplasms should be suspected when the onset of androgenic symptoms is rapid and sudden, when androgen excess initially presents in late life, when they lead to virilization and masculinization, or are associated with cushingoid features. Nonetheless, it should be remembered that some of younger patients with virilization suspected of having an androgen-secreting neoplasm actually suffer from the HAIRAN syndrome (discussed earlier). Suppression and stimulation tests can be misleading and are not encouraged for the diagnosis of these neoplasias. Overall, the single best predictor of an androgen-producing tumor is clinical presentation, and not biochemical markers. For example, in one study we noted that the positive predictive value of a repeat total testosterone above 250 ng/dL was only 9%, as most women with total testosterone levels above this cutoff have other abnormalities, such as the HAIRAN syndrome or PCOS. Rare Causes of Androgen Excess Clinical signs of androgen excess may also result from the presence a corticotropin-secreting tumor, due to either a pituitary adenoma (i.e., Cushing's disease) or an extrapituitary (ectopic) source, although patients are usually also cushingoid. In one study of young women with Cushing's syndrome all had stigmata of hyperandrogenism (hirsutism, acne, or balding), but only 70% had oligomenorrhea. However, as Cushing's syndrome has an extremely low prevalence in the population (1 in a million) and screening tests do not have 100% sensitivity/specificity, routine screening of all women with androgen excess for Cushing's syndrome is not indicated. Nonetheless, clinical signs more commonly found in Cushing's syndrome, such as ecchymoses, proximal muscle weakness, centripetal reddened striae, facial rubor and swelling, and perhaps hypertension and glucose intolerance should signal screening tests. Screening for cortisol excess can be conducted with a 24-hour urine collection for free cortisol. The use of exogenous androgens, often found among body builders, or the excessive use of androgens among postmenopausal women should also figure in the differential diagnosis. Severe hirsutism, and even virilization, that begins during pregnancy has its own unique differential including benign ovarian sources such as hyperreactio luteinalis (i.e., gestational ovarian theca–lutein cysts) or luteomas, and extremely rare fetoplacental sources, such as aromatase deficiency resulting in androgen excess due to the placental inability to convert precursor androgens into estrogens. Finally, gonadal dysgenesis associated with an abnormal Y chromosome can initially present as peripubertal androgenization.

IDIOPATHIC HIRSUTISM Although idiopathic hirsutism (IH) is not actually an androgen excess disorder, it is included among them and should be considered in the differential diagnosis of androgen excess. Overall, approximately 5% to 15% of hirsute women will have the diagnosis of “idiopathic” hirsutism. The diagnosis of IH is, by exclusion, made in a patient who is obviously hirsute, but in whom the circulating androgens and ovulatory function appear to be normal, and no other disorders can be identified. While some clinicians use regular menses as diagnostic of normal ovulatory function, we should note that approximately 40% of eumenorrheic hirsute women are actually anovulatory, and hence do not suffer from IH. Although biochemically these patients do not have an obvious elevation in circulating androgen levels, it also is likely that many of these patients simply demonstrate degrees of hyperandrogenemia that may not be detectable with routine clinical androgen assays. Nonetheless, in some of these women the 5a-reductase activity in the skin and hair follicles is probably overactive and leads to hirsutism in spite of normal circulating androgen levels.

EVALUATION OF ANDROGEN EXCESS The history and physical examination are essential to making a diagnosis of the underlying cause of androgen excess. History taking should focus on the onset and duration of the various stigmata of androgen excess, should include a thorough menstrual history, and a discussion of concomitant medications. The physical examination should carefully assess the patient for cushingoid features, acanthosis nigricans, balding, acne, and degree and type of body hair distribution. One commonly used scale to score the degree of excess hair growth is based on a modification of the Ferriman-Gallwey scale ( Fig. 37.8). Signs of virilization and masculinization should be sought, although their recognition is usually obvious. Clitoromegaly is usually defined as a clitoral index greater than 35 mm 2 , where the clitoral index is the product of the sagittal and transverse diameters of the glans of the clitoris in millimeters. In normal women these diameters are in the range of 5 mm.

FIG. 37.8. Modified Ferriman-Gallwey scale for assessing hirsutism. (Modified from Hatch, et al. Hirsutism: implications, etiology, and management. Am J Obstet Gynecol 1981;140:815–830.)

Evidence and risk factors for insulin resistance should be sought out as they are critical to the long-term health of the women affected. Family history of diabetes and CVD, especially of first-degree relatives with premature-onset heart disease (males 38), or in women with an elevated day 3 FSH. Most commonly, zona drilling is performed with acid Tyrode solution approximately 72 hours after oocyte retrieval, on the day of embryo transfer. This “drilling” creates a defect in the zona pellucida approximately 30 µm in diameter (see Fig. 39.4). Indications for AH include:

FIG. 39.4. Assisted hatching. The zona pellucida is opened with acid Tyrode solution. To the left is the holding pipette. The pipette on the right delivers acid Tyrode next to the zona pellucida.

age greater than 38 elevated day 3 FSH a prior failed IVF cycle with suspected implantation failure increased zona thickness on microscopy excess oocyte fragmentation. AH for the selected indications may increase implantation and pregnancy rates but randomized trials are needed in order to determine the best candidates for AH.

EMBRYO TRANSFER Embryo transfer melds the efforts of the clinician and laboratory. The success of implantation hinges in part on the success of the embryo transfer. The controversy remains as to the optimal timing of embryo transfer. Embryos can be transferred up to 6 days after fertilization. However, most embryos are transferred at 3 days, at the 8 to 10 cell stage ( Fig. 39.5). Proponents of later embryo transfer argue that implantation and pregnancy rates are higher at the blastocyst stage as the embryo is further developed and the embryos that progress in vitro to this stage are healthier and more likely to implant. This has the theoretic advantage of decreasing the need to transfer more than one or two embryos, and thereby reduce the multiple pregnancy rate.

FIG. 39.5. Grade I embryo ready for transfer. This excellent embryo has 10 to 12 even cells and less than 15% fragmentation.

Embryos can be transferred with a variety of soft catheters. A soft-tipped catheter conforms to the contour of the cervix and endometrial cavity and reduces tissue trauma. The catheter is attached to a 1-mL syringe containing approximately 30 µL of transfer medium in a continuous column with the embryos placed toward the tip. Patients should have a full bladder prior to transfer which helps straighten out the angle between the uterus and cervix in women with an anteverted uterus. The cervix and vagina are prepped and draped and care is taken not to introduce blood, mucus, or bacteria with the catheter tip as these contaminants produce a poor environment for implantation. Using abdominal ultrasound guidance, embryos are deposited between 1 and 1.5 cm from the fundus without actually contacting it. Touching the fundus during transfer may cause uterine contractions thereby increasing the intrauterine pressure and expelling the embryos through the cervix, which reduces the pregnancy rate. Ultrasound guidance facilitates embryo transfer and confirms that the embryos are placed correctly high in the uterine fundus. There appears to be a significant increase in implantation and pregnancy rates using ultrasound-guided embryo transfer when compared to touching the fundus prior to transfer to determine the catheter tip location in the uterine fundus. After transfer, the catheter is checked for any retained embryos that were not expelled. The best quality day 3 embryos are selected for fresh embryo transfer. The morning of transfer, embryos are graded and selected for transfer in order of their grade. They are graded as I, II, or III depending on the number and evenness of the dividing blastomeres, the percentage of extracellular fragmentation, and their cleavage rates. Ideally, grade I embryos (see Fig. 39.5) are selected first for transfer. These are “excellent” embryos with 8 even cells and less than 15% fragmentation, while grade II embryos have 6 to 8 even or slightly uneven cells with up to 25% fragmentation. Grade III embryos have less than 6 uneven cells and excessive fragmentation and are typically not transferred fresh or are cryopreserved because their obvious degeneration renders them virtually unable to develop further. If the patient has a significant number of excellent embryos or a uterine anomaly she may be offered day 5 embryo transfer with one or two embryos. Patients 35 years and under are routinely transferred with two embryos; three embryos are transferred in women over age 35.

OOCYTE DONATION Oocyte donation consists of oocyte retrieval with subsequent embryo transfer to a third party (the infertile patient or surrogate carrier). Oocyte donors can be either known or anonymous. Oocyte donation is used for women with: POF a history of surgical castration diminished ovarian reserve a risk of transmitting a heritable disease to an offspring older menopausal patients. Donor-recipient IVF cycles typically have the highest pregnancy rates among all ARTs and the number of donor-recipient cycles is increasing. In 1999, the ASRMSART (American Society for Reproductive Medicine/Society for Assisted Reproductive Technology) registry reported that 10% (9,066) of all ART cycles used donor oocytes. The success of donor recipient cycles is based on the tenet that the capacity to conceive is not based on uterine aging but on the quality of the oocyte. Younger donors have higher implantation and pregnancy rates when compared to older donors, with a decreased incidence of preclinical and clinical pregnancy loss. There does not

appear to be a decrease in the per cycle or cumulative pregnancy rates as endometrial receptivity is maintained and not altered by advancing maternal age. In addition, there was no association between cumulative pregnancy rates and the etiology of the infertility of the recipient. Patients with Turner syndrome are a special class of patient using donor oocytes. They are infertile due to gonadal dysgenesis, with natural pregnancies occurring in approximately 2% of these women, usually in patients with mosaicism (XX,XO). Small studies comparing oocyte donation in patients with Turner syndrome to controls report conflicting pregnancy and implantation rates. There may be an inherent endometrial receptivity problem in Turner syndrome with lower implantation and pregnancy rates. Women with Turner syndrome should be carefully screened as cardiovascular abnormalities and arterial hypertension are common, with coarctation of the aorta seen in 10% of patients, and bicuspid aortic valve noted in about one-third of patients without coarctations. Even fatal aortic dissection has been reported in pregnancies established by oocyte donation in women with Turner syndrome. Donor Oocyte Recipient Endometrial Preparation Embryos are transferred into the recipient's uterus after exogenous endometrial preparation using sequential estrogen and progesterone and gonadotropin down-regulation if they are still cycling. This hormonal manipulation primes the endometrium for embryo transfer by mimicking the natural menstrual cycle. Leuprolide acetate (0.1 µg) is started on cycle day 21 for gonadotropin down-regulation. Bleeding will typically begin after adequate gonadotropin suppression and the decline in the luteal support. After a normal baseline ultrasound and low endogenous estradiol level is created, estrogen therapy is initiated, based on the date of anticipated embryo transfer. Transdermal estradiol patches (0.1 mg) are used in increasing doses that mimic the progressively increasing estradiol level in the follicular phase. By day 11, most women are using four patches. Most IVF programs initiate aspirin (81 mg) at the start of stimulation in the hope of improving uterine and ovarian blood flow, implantation, and pregnancy rates. In an optimum fresh embryo transfer cycle, Lupron is maintained while transdermal estradiol is adjusted according to endometrial thickness and plasma estradiol levels. The recipient will receive between 2 and 3 weeks of estrogen for endometrial preparation. Once the endometrial lining is =8 mm (14 days) the recipient is ready for the pharmacologic conversion to luteal phase in coordination with the status of the oocyte donor. Both intramuscular (50 mg of progesterone in oil) and vaginal progesterone (100 mg b.i.d.) is started along with Medrol and tetracycline. The majority of fresh embryos are transferred on day 3 which translates into a day 18 transfer for cryopreserved embryos. After embryo transfer, the estradiol, the aspirin, and progesterone in oil are continued until the first pregnancy test, at which time the dose of progesterone is adjusted upward. Oocyte donors must be screened prior to donation as outlined in the ASRM guidelines for gamete and embryo donation. Donors should be between the ages of 21 to 34 years. Screening for a personal genetic history as well as a sexual history is complemented by testing for infectious and genetic diseases. Psychological assessment by a qualified mental health professional for the donor and her partner is required. Minimal genetic laboratory screening is typically required. Once prospective donors complete this initial evaluation, they are matched with potential recipients based on both physical and personal characteristics. Prior to starting a donor–recipient cycle, recipients have a precycle evaluation that includes an evaluation of the uterine cavity by sonohysterography or office hysteroscopy. If an intracavitary lesion or a hydrosalpinx is noted, surgical removal is conducted prior to embryo transfer. Oocyte donation is not without risks to the donor. Donors are often concerned about the risks of ovulation-inducing drugs and ovarian cancer, although the aggregate of studies does not support any association. With COH, the risk of OHSS is always present, although it is decreased in donor oocyte cycles since conception does not ensue. Oocyte retrieval places the donor at risk for anesthetic complications, pelvic infection, and intraperitoneal hemorrhage. These complications are real, though exceedingly rare. There are also potential psychological risks that donors undertake. With appropriate pretreatment screening, we can hopefully minimize donors who may have feelings of ambivalence or regret after donation. Recipients over 45 years of age are at an increased risk for obstetric complications including gestational diabetes mellitus, hypertensive disorders, cesarean delivery, intrauterine growth restriction, abruptio placentae, and preterm labor. These patients are referred to maternal–fetal medicine specialists for consultation prior to and after conception with IVF. Since there is an increasing risk of adverse pregnancy outcome with advanced maternal age, many centers do not perform donation for women over 50 years of age.

CRYOPRESERVATION After COH and fresh embryo transfer, 60% of stimulated IVF cycles will produce excess viable embryos which are available for cryopreservation. Cryopreserved or frozen embryos can be thawed and transferred back into the uterus, during a subsequent frozen embryo transfer cycle. This allows for higher overall pregnancy rates per attempted IVF cycle. The indications for embryo cryopreservation include: storing excess embryos for future use after a fresh embryo transfer decreasing the risk of OHSS in a fresh embryo transfer cycle at very high risk of OHSS uterine conditions that are unfavorable for fresh embryo transfer after retrieval (e.g., uterine bleeding, polyps, leiomyomas, severe cervical stenosis, or a thin endometrial lining). Uterine polyps, leiomyomas, and cervical stenosis are usually identified by sonohysterography, ultrasound, or at trial transfer prior to oocyte retrieval and corrected prior to IVF. Cryopreservation techniques attempt to minimize cell damage to embryos during the freezing and thawing process with the aid of cryoprotectants. Embryos are frozen at a slow rate with the cryoprotectant. A gradient is induced that allows intracellular water to leave the cell. The embryo is dehydrated to avoid the formation of cytotoxic intracellular ice crystals. Once they are frozen, the embryos are loaded into cryostraws and stored in liquid nitrogen at -196°C. When embryos are needed for transfer, they are thawed rapidly to avoid formation of intracellular ice crystals. Typically, cryopreservation results in an 80% survival rate after thawing frozen embryos. Patients should be extensively counseled prior to oocyte retrieval with regard to cryopreserving excess embryos. Informed consent is obtained as outlined in the ASRM committee opinion on elements to be considered in obtaining informed consent for ART. Semen is cryopreserved in men who are not able to produce a sample on the day of oocyte retrieval (due to performance anxiety), who are oligospermic, or azoospermic (post-MESA, PESA, or TESA). Even with the use of cryoprotectants, freezing and thawing sperm samples can decrease motility by 50%. This is usually not a problem as original ejaculate samples contain large numbers of sperm. In men with azoospermia, epididymal or testicular samples are usually cryopreserved until the day of oocyte retrieval. On the day of oocyte retrieval, the semen sample(s) are thawed, the cryoprotectant is removed, and ICSI is performed.

FROZEN EMBRYO TRANSFER The morning of “frozen embryo transfer,” embryos are thawed and surviving embryos are again graded. Pregnancy rates for fresh embryo transfer and frozen embryo transfer are very similar when similar grade embryos are transferred. This is due to selection criteria for embryo freezing. As discussed previously, day 3 embryos are graded as excellent (grade I), average (grade II), and poor (grade III). After the highest graded embryos are transferred fresh, the remaining grade I and II embryos are cryopreserved, while grade III embryos are not usually transferred or cryopreserved. The number of frozen embryos that are transferred is based on the same criteria as for fresh embryo transfer. Prior to frozen embryo transfer, the recipient will receive between 2 and 3 weeks of estrogen for endometrial preparation. Once the endometrial lining is =8 mm (14 days), intramuscular and vaginal progesterone is started for endometrial support.

RISKS OF IVF Couples who present for ARTs are usually healthy with no significant medical history aside from their infertility. In the treatment of their infertility, they are asked to take significant risks. The two major complications of ARTs are OHSS and multiple gestation. There has been a proposed association between ovulation-induction drugs and gynecologic cancers, although this is not supported by well-controlled retrospective trials. Prior to initiating an IVF cycle, the risks and benefits of ARTs should be discussed, and formal written consent is obtained. In addition, patients should be counseled on fetal reduction of higher-order multiple pregnancies to decrease perinatal morbidity of infants and mother.

OHSS OHSS is a potentially life-threatening complication of gonadotropin-induced COH. Although the signs and symptoms of impending OHSS usually become evident during stimulation, this syndrome does not become fully manifested until after hCG administration and oocyte retrieval. Patients typically resolve the OHSS within 10 to 14 days after onset of their initial symptoms. The syndrome is prolonged if a triggering dose of hCG is followed by a supplemental dose of hCG or if the patient

becomes pregnant and produces her own hCG. Patients who develop OHSS complain of abdominal bloating and pain due to ovarian enlargement. The pathophysiology of OHSS includes increased capillary permeability with “third spacing” of protein-rich fluid. This can result in hemoconcentration and ascites with dramatic fluid accumulations in the abdomen, pleura, or pericardial spaces. Patients can develop nausea, vomiting, diarrhea, and a decrease in appetite if OHSS progresses. As significant amounts of “third spacing” ensue, we often see shortness of breath and decreased urine output. The precise pathophysiology of OHSS is still not completely understood, though vascular endothelial growth factor (VEGF) also known as “vascular permeability factor” has been implicated. It has been shown that follicular fluid from patients with OHSS produce VEGF in increased quantities which, in a dose-dependant manner, increase vascular permeability. OHSS can be staged based on clinical signs and symptoms, ultrasound features, and laboratory findings and is used to help predict which women require hospitalization. Patients with mild OHSS present with abdominal discomfort, distention and pain, with ovaries enlarged up to 12 cm. Patients may have nausea, vomiting, or diarrhea. Moderate OHSS includes all the features of mild OHSS, but ultrasonographic evidence of ascites as well. Severe OHSS complicates less than 2% of stimulations. It includes features of mild and moderate OHSS as well as clinical evidence of ascites or hydrothorax and difficulty breathing. In addition, severe OHSS patients are at risk for hemoconcentration, coagulation and electrolyte abnormalities, diminished renal perfusion and function, and even adult respiratory distress syndrome. Risk factors for developing OHSS include young age, low body mass index, and elevated estradiol levels with an increased number of stimulated follicles during COH. In addition, women with a previous pregnancy complicated by a history of OHSS are at great risk of recurrence. If early manifestations of OHSS appear (i.e., abdominal pain, rapidly increasing estradiol levels [>3,000 pg/mL], and extensive follicular recruitment), preventive treatment strategies are employed. On rare occasions, when follicular development and estradiol levels are excessively high early in the stimulation protocol, all stimulation medications are stopped and the cycle cancelled. More often, gonadotropin dosages are adjusted so that estradiol levels increase more slowly, or “freeze all” the embryos rather than transfer them as fresh embryos. Alternatively, using a “coast” by withdrawing gonadotropins and withholding hCG until the estradiol level decreases often allows retrieval and transfer of fresh embryos. Once OHSS develops, a complete physical exam and ultrasound are performed including blood tests for sodium, potassium, creatinine, and hematocrit. On physical exam, patients often have abdominal distension and tenderness. Decreased breath sounds may be heard at the bases with pleural effusions. Ultrasound is performed to note the presence of ascites and ovarian size. Patients with OHSS can develop hyponatremia, hemoconcentration, hyperkalemia and even decreased renal perfusion. These patients are monitored as outpatients daily until the condition improves significantly. Patients with severe OHSS are admitted to the hospital for inpatient management using intravenous albumin. Patients are instructed to weigh and measure their abdominal girth daily, drink eight glasses of a high sodium drink, and maintain a high protein diet. In addition, they monitor their urine output, and are instructed to report any changes. An increase in abdominal pain or girth, nausea, vomiting, or decreased appetite necessitates a prompt phone call to the health care provider and an office visit. Fertility Drugs and Cancer The potential association between ovulation induction drugs and ovarian cancer remains controversial. The data suggesting an association have been based on meta-analyses, but more recent, well-done case-control studies reassuringly have not found any association when evaluating treatment of infertile women with fertility medications compared with infertile women not treated with ovulation-induction drugs. Patients should be informed about the available studies on this issue. It is appropriate to minimize the number of ovulation-induction cycles, both with clomiphene citrate and gonadotropins, as long as there is any concern surrounding ovulation-induction agents and ovarian cancer. Multiple Pregnancy Multiple pregnancies occur with a higher frequency in pregnancies resulting from ARTs than spontaneous conception. Since there is a low implantation rate per embryo (10%–25%) with IVF, more than one embryo is often placed into the uterus. The goal is to minimize the number of multiple pregnancies while maintaining good pregnancy rates. Multifetal gestations confer significant morbidity and mortality to both the mother and fetuses because of a high rate of prematurity and in utero mortality. Maternal risks include preterm labor, placental abruption and previa, cesarean section, postpartum hemorrhage, gestational diabetes, and preeclampsia. According to the U.S. Department of Health/Centers for Disease Control and Prevention's 1999 National Summary and Fertility Clinic Reports, 37% of all ART births using fresh nondonor eggs were multiple births (84% twins and 16% triplets and higher), while less than 3% of births in the general population are multiples. Patients under 34 years of age with a history of a full-term pregnancy and no history of female infertility have the highest implantation rates with IVF. This includes women with previous tubal ligations, normal ovarian reserve testing, donor oocyte recipients, and couples with male factor infertility. In contrast, women older than 34 years of age, with no pregnancies, and extended histories of infertility, have lower implantation rates. Quite often these patients have a history of endometriosis, poor ovarian reserve, and previous failed IVF cyclesembryo transfers. During the initial IVF consultation, the IVF success rates and the patient's infertility history are reviewed. Based on this information, the number of embryos we recommend transferring and the number of embryos a couple is willing to accept is defined. The goal is that each patient will have two excellent quality embryos replaced, adjusting (up or down) the number of embryos according to embryo quality and each patient's infertility history. Multifetal pregnancy reduction should be discussed as an option for couples with high-order multiple pregnancy. The aim is to reduce preterm deliveries by decreasing the number of fetuses a woman carries. Besides being an invasive procedure, pregnancy reduction, either transabdominal or transvaginal, risks the loss of the entire pregnancy. In experienced centers, pregnancy loss after transabdominal fetal reduction occurs in 6% to 8% of these pregnancies. In addition, the data suggest that pregnancies reduced to twins proceed as if the fetuses were naturally conceived as twins. There is a psychological toll on couples who are making decisions regarding multifetal reduction, especially in women who have experienced difficulty becoming pregnant. If fetal reduction is not acceptable to a patient under 40, a maximum of three embryos is transferred. High-order multiple pregnancies (triplets and higher) can no longer be viewed as an acceptable risk of IVF and transferring two embryos will decrease the risk. Proponents of two-embryo transfers argue that pregnancy rates in large series are equivalent when compared to three-embryo transfers. Multiple pregnancy rates in patients 30 to 35 years of age are significantly increased from approximately 29% when two embryos were transferred to 40% with the transfer of three embryos. Interestingly, the twin birth rate with transfers of two and three embryos is 26% and 29%, respectively. Thus, a decrease in embryos transferred from three to two decreases the incidence of high-order pregnancies, but does not reduce the twinning rate. Preimplantation Genetic Diagnosis Preimplantation genetic diagnosis (PGD) enables couples who are carriers of genetic diseases to test embryos in vitro for genetic abnormalities prior to embryo transfer and select the embryos free of the genetic error for transfer. Traditionally, carrier couples have used chorionic villus sampling and amniocentesis, during the first and second trimesters respectively, to determine if they have a genetically abnormal fetus. The major disadvantage of this type of prenatal diagnosis is the need for second trimester pregnancy termination when genetically abnormal fetuses are detected. PGD provides couples the option to select genetically normal embryos for embryo transfer prior to transfer. PGD involves performing a blastomere biopsy and then genetic testing to determine the genetic complement of the embryo. Blastomeres can be removed from oocytes and embryos at three different stages, polar body analysis, 6- to 10-cell cleavage stage biopsy, and blastocyst stage biopsy. Cleavage stage biopsy is the most widely used technique. In order to biopsy a blastomere or polar body a hole is made in the zona pellucida using micromanipulation instruments with either acid Tyrode or laser. Since these embryos are transferred back into the uterus on day 4 or 5, a genetic diagnosis is determined within 48 hours. First and second polar body biopsies evaluate only maternal chromosomes. Crossover recombination events between homologous chromosomes can create improper diagnoses of heterozygous genetic defects. Presently, polar body biopsy is not performed by most centers offering PGD. Cleavage stage biopsies are performed on one or two blastomeres of 6- to 10-cell stage embryos 3 days after insemination. These cells are used for diagnosis with polymerase chain reaction (PCR) or fluorescent in situ hybridization (FISH). PCR amplifies fragments of DNA for specific single gene defects. DNA amplification requires careful monitoring to avoid the problems of accidental contamination and allele dropout (ADO) leading to an erroneous genetic diagnosis. ADO or preferential amplification occurs when one of two alleles amplifies preferentially over the other. In a heterozygous cell, the normal allele may be amplified over the abnormal allele and the embryo would be diagnosed as “normal,” and then transferred. PGD has been applied to many monogenic diseases including autosomal recessive, dominant, and X-linked diseases. Examples include cystic fibrosis, Tay–Sachs, spinal muscular dystrophy, Huntington disease, Marfan syndrome, and fragile X syndrome. Defining the number and structure of metaphase chromosomes is carried out by FISH, not PCR. FISH is used to examine chromosomes in embryos for aneuploidy in patients with male factor infertility and women of advanced maternal age, as well as to detect chromosomal translocations. Common aneuploidies occurring and creened for with FISH include defects in chromosomes 13, 16, 18, 21, 22,X, and Y. FISH uses fluorochrome-labeled probes that hybridize to complementary DNA sequences. The fluorescence produced by the probes allows a color and number notation of these probes which aids in diagnosis. PGD is successful in about 80% to 90% of biopsied embryos, with 50% of these embryos being unaffected and suitable for transfer. Amniocentesis and chorionic villus sampling continues to be recommended because of possible errors in cell sampling or technical difficulties with PCR or FISH. Despite this concern, the risk of carrying a genetically abnormal fetus is limited to the risk of a testing error which is far less than the spontaneous risk of the genetic disease.

SUCCESS RATES The Fertility Clinic Success and Certification Act of 1992 required all clinics performing ARTs in the United States to annually report their success rates to the Centers for Disease Control and Prevention (CDC). Through SART, the CDC obtains clinic data. The CDC compiles these data and publishes a yearly report which is also available on the CDC website. The first year this clinical data became available for review was 1995. There is a 3-year lag between the time clinics report to the CDC and the time that these reports are available to the public which reflects the time for the last delivery in a calendar year and the subsequent time needed to compile and publish the data. Each clinic reports on their “pregnancy success rates” based on the type of cycle performed (fresh embryos from nondonor eggs, frozen embryos from nondonor eggs, and donor eggs) as well as the age of the women (40.). The number of ART cycles in the United States has steadily increased as well as the success rates. In 1995, 45,906 fresh nondonor IVF cycles were performed as compared to 65,751 in 1999 ( Table 39.1). For patients using their own oocytes, pregnancy rates decrease with increasing age. It is difficult to directly compare the success rates between individual ART programs as they each treat different populations of infertile couples, they may focus on specific infertility factors rather than others (e.g., male factor, preimplantation genetic diagnosis, etc.), and different judgments are often made as to the number of embryos to transfer, IVF protocols, and success rates.

TABLE 39.1. 1999 CDC/SART assisted reproductive technology success rates

SUMMARY POINTS IVF was initially developed to treat tubal factor infertility but now represents the final therapy for virtually all infertile couples, regardless of the etiology. There is a significant maternal age-related decline in IVF pregnancy and delivery rate which should be taken into account when discussing infertility therapies. Day 3 FSH ± estradiol and clomiphene challenge test is a more accurate predictor of IVF success than simply maternal age. IVF with ICSI has allowed procreation in men with severe oligospermia and in azoospermic men after retrieval of sperm from the epididymis and testes. High-order multiple gestation significantly increases maternal and fetal morbidity and mortality. Careful consideration should be given to transferring the fewest number of embryos possible. Donor oocyte IVF is very successful for women with ovarian failure, limited oocyte reserve, advanced maternal age, and genetic disorders. Preimplantation genetic diagnosis is being increasingly performed using blastomere biopsy and genetic screening to avoid transferring embryos with aneuploidy and autosomal recessive or autosomal dominant gene mutations. SUGGESTED READINGS Patient Selection for IVF Guzick DS, Sullivan MW, Adamson GD, et al. Efficacy of treatment for unexplained infertility. Fertil Steril 1998;70:207–213. Nestler JE, Jakubowicz DJ, Evans WS, et al. Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovarian syndrome. N Engl J Med 1998;338:1876–1880. Paulson RJ, Hatch IE, Lobo RA, et al. Cumulative conception and live birth rates after oocyte donation: implications regarding endometrial receptivity. Hum Reprod 1997;12:835–839.

Testing Prior to Assisted Reproduction American Society for Reproductive Medicine. Guidelines for gamete and embryo donation. ASRM Practice Committee Report, 1997.

Ovarian Reserve Testing Levi AJ, Raynault MF, Bergh PA, et al. Reproductive outcome in patients with diminished ovarian reserve. Fertil Steril 2001;76:666. Navot D, Rosenwaks Z, Margalioth EJ. Prognostic assessment of female fecundity. Lancet 1987;2:645–647. Toner JP, Philput CB, Jones GS, et al. Basal follicle stimulating hormone level is a better predictor of in vitro fertilization performance than age. Fertil Steril 1991;55:784–791.

Evaluation of the Uterus and Fallopian Tubes Camus E, Poncelet C, Goffinet F, et al. Pregnancy rates after in vitro fertilization in cases of tubal infertility with and without hydrosalpinx: a meta-analysis of published comparative studies. Hum Reprod 1999;14:1243–1249. Eldar-Geva T, Meagher S, Healy DL, et al. Effect of intramural, subserosal, and submucosal uterine fibroids on the outcome of assisted reproductive technology treatment. Fertil Steril 1998;70:687–691. Hayden HA, Li TC, Cooke ID. The septate uterus: a review of management and reproductive outcome. Fertil Steril 2000;73:1–14. Hurst BS, Tucker KE, Awoniyi CA, et al. Hydrosalpinx treated with extended doxycycline does not compromise the success of in vitro fertilization. Fertil Steril 2001;75:1017–1019. Lass A, Williams G, Abusheikha N, et al. The effect of endometrial polyps on outcome of in vitro fertilization (IVF) cycles. J Assist Reprod Genet 1999;16:410–415. Pal L, Shifren JL, Isaacson KB, et al. Impact of varying stages of endometriosis on the outcome of in vitro fertilization-embryo transfer. J Assist Reprod Genet 1998;15:27–31. Strandell A, Lindhard A, Waldenstrom U, et al. Hydrosalpinx and IVF outcome: a prospective, randomized multicentre trial in Scandinavia on salpingectomy prior to IVF. Hum Reprod 1999;14:2762–2769. Surrey ES, Lietz AK, Schoolcraft WB. Impact of intramural leiomyomata in patients with a normal endometrial cavity on in vitro fertilization-embryo transfer cycle outcome. Fertil Steril 2001;75:405–410.

Evaluation of the Male Factor Daya S. Overview analysis of outcomes with intracytoplasmic sperm injection. J SOGC 1996;18:645. Kim ED, Lipshultz LI. Evaluation and imaging of the infertile male. Infertil Reprod Med Clin North Am 1999;10:377–409.

Semen Analysis and Sperm Preparation Kruger TF, Acosta AA, Simmons KF, et al. Predictive value of abnormal sperm morphology in in vitro fertilization. Fertil Steril 1998;49:112.

Genetic and Hormonal Evaluation of the Infertile Male Jaffe T, Oates RD. Genetic abnormalities and reproductive failure. Urol Clin North Am 1996;21:389. Reijo R, Lee TY, Salo P, et al. Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNA-binding protein gene. Nat Genet 1995;10:383.

Preparation for IVF Mansour R, Aboulghar M, Serour G. Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertil Steril 1990;54:678–681.

COH Medications and Strategies Agrawal R, Holmes J, Jacobs HS. Follicle-stimulating hormone or human menopausal gonadotropin for ovarian stimulation in in vitro fertilization cycles: a meta analysis. Fertil Steril 2000;73:338–343. Benadiva CA, Davis O, Kligman I, et al. Withholding gonadotropin administration is an effective alternative for the prevention of ovarian hyperstimulation syndrome. Fertil Steril 1997;67:724–727. Hughes EG, Fedorkow DM, Daya S, et al. The routine use of gonadotropin-releasing hormone agonists prior to in vitro fertilization and gamete intrafallopian transfer: a meta-analysis of randomized controlled trials. Fertil Steril 1992;58:888–896. Olivennes F, Belaisch-Allart J, Emperaire JC, et al. Prospective, randomized, controlled study of in vitro fertilization-embryo transfer with a single dose of a leuteinizing hormone-releasing hormone (LH-RH) antagonist (cetrorelix) or a depot formula of an LH-RH agonist (triptorelin). Fertil Steril 2000;73:314–320. Stadtmauer LA, Toma SK, Riehl RM, et al. Metformin treatment of patients with polycystic ovary syndrome undergoing in vitro fertilization improves outcomes and is associated with modulation of the insulin-like growth factors. Fertil Steril 2001;75:505–509.

Luteal Progesterone Support after Oocyte Retrieval Soliman S, Daya S, Collins J, et al. The role of luteal phase support in infertility treatment: a meta-analysis of randomized trials. Fertil Steril 1994;61:1068–1076.

ICSI and Sperm Retrieval Techniques for Severe Male Factor Schlegel PN. Sperm retrieval techniques for assisted reproduction. Infertil Reprod Med Clin North Am 1999;10:539–553.

Assisted Zona Hatching Prior to Embryo T'ransfer Cohen J, Alikani M, Trowbridge J, et al. Implantation enhancement by selective assisted hatching using zona drilling of human embryos with poor prognosis. Hum Reprod 1992;7:685–691.

Embryo Transfer Schoolcraft WB, Surrey ES, Gardner DK. Embryo transfer: techniques and variables affecting success. Fertil Steril 2001;76:863–871.

Oocyte Donation Navot D, Drews MR, Bergh PA, et al. Age-related decline in female fertility is not due to diminished capacity of the uterus to sustain embryo implantation. Fertil Steril 1994;61:97–101.

Donor Oocyte Endometrial Preparation American Society for Reproductive Medicine. Guidelines for gamete and embryo donation. Fertil Steril 2002[Suppl 5];77:1S–18S.

Cryopreservation American Society for Reproductive Medicine. Elements to be considered in obtaining informed consent for ART. Committee Opinion, June 1997 Elder E, Dale B. Cryopreservation. In: In vitro fertilization, second ed. United Kingdom: Cambridge University Press, 2000:192–224.

Ovarian Hyperstimulation Syndrome (OHSS) Levin ER, Rosen GF, Cassidenti DL, et al. Role of vascular endothelial cell growth factor in ovarian hyperstimulation syndrome. J Clin Invest 1998;102:1978–1985.

Fertility Drugs and Cancer Mosgaard BJ, Lidegaard O, Kjaer SK et al. Infertility, fertility drugs, and invasive ovarian cancer: a case-control study. Fertil Steril 1997;67:1005–1012. Potashnik G, Lerner-Geva L, Genkin L, et al. Fertility drugs and the risk of breast and ovarian cancers: results of a long-term follow-up study. Fertil Steril 1999;71:853–859.

Multiple Pregnancy Berkowitz R, Lynch L, Stone J, et al. The current status of multifetal pregnancy reduction. Am J Obstet Gynecol 1996;174:1265–1272. Templeton A, Morris JK. Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. N Engl J Med 1998;339:573–577.

Preimplantation Genetic Diagnosis Kanavakis E, Traeger-Synodinos J. Preimplantation genetic diagnosis in clinical practice. J Med Genet 2002;39:6–11.

Success Rates US Department of Health and Human Services, Centers for Disease Control and Prevention. 1999 assisted reproductive technology success rates. National Summary and Fertility Clinic Reports, Dec 2000.

Chapter 40 Endometriosis Danforth’s Obstetrics and Gynecology

Chapter 40 Robert S. Schenken

Endometriosis

INTRODUCTION Pathogenesis Epidemiology Pathology Symptoms Diagnosis Classification Treatment SUGGESTED READINGS Pathogenesis Epidemiology Pathology Diagnosis Classification Treatment

INTRODUCTION Endometriosis is defined by the presence of endometrial glands and stroma outside the endometrial cavity and uterine musculature. The pelvis is the most common site of endometriosis, but endometriotic implants may occur nearly anywhere in the body. Although there are numerous theories to explain why women develop endometriosis, no one theory has been proven conclusively. Endometriosis is a common gynecologic problem in reproductive-age women who have pelvic pain, dyspareunia, or infertility. The management of endometriosis is controversial, but randomized clinical studies have substantiated some therapeutic approaches. Pathogenesis Several theories have been proposed to explain the histogenesis of endometriosis. The implantation theory proposes that endometrial tissue desquamated during menstruation passes through the fallopian tubes, where it gains access to and implants on pelvic structures. The incidence of retrograde menstruation is similar in women with and without endometriosis. Thus, the development of endometriosis could depend on the quantity of endometrial tissue reaching the peritoneal cavity, specific factors enhancing attachment of endometrial cells to the peritoneum and ovary, or the capacity of a woman's immune system to remove the refluxed menstrual debris. The direct transplantation theory is the probable explanation for endometriosis that develops in episiotomy, cesarean section, and other scars following surgery. Endometriosis in locations outside the pelvis likely develops from dissemination of endometrial cells or tissue through lymphatic channels or blood vessels. The coelomic metaplasia theory proposes that the coelomic (peritoneal) cavity contains undifferentiated cells or cells capable of dedifferentiating into endometrial tissue. This theory is based on embryologic studies demonstrating that all pelvic organs, including the endometrium, are derived from the cells lining the coelomic cavity. The induction theory, an extension of the coelomic metaplasia theory, postulates that the refluxed endometrial debris releases a product that activates undifferentiated peritoneal cells to undergo metaplasia. There is no conclusive proof that peritoneum can undergo spontaneous or induced metaplasia. Anatomic alternations of the pelvis that increase tubal reflux of menstrual endometrium increase a woman's chance of developing endometriosis. The incidence of endometriosis is increased in young women with genital tract obstructions that prevent expulsion of menses into the vagina and increase the likelihood of tubal reflux. Other studies have suggested that deficient cellular immunity results in an inability to recognize the presence of endometrial tissue in abnormal locations. Decreased natural killer cell activity resulting in decreased cytotoxicity to autologous endometrium has been reported in women with endometriosis. The presence of increased concentrations of leukocytes and their cytokine products in peritoneal fluid of women with endometriosis may play a role in the initiation and growth of the ectopic implants. The immune system clearly has an important, albeit unclear, role in the pathogenesis of endometriosis. The possibility of a familial tendency for endometriosis has been recognized for several decades. If a patient has endometriosis, a first-degree female relative has a 7% likelihood of being affected similarly. Epidemiology The true prevalence of endometriosis in the general population is unknown. Estimates of its prevalence are based on visualization of the pelvic organs. Pelvic endometriosis is present in approximately 1% of women undergoing major surgery for all gynecologic indications, 6% to 43% of women undergoing sterilization, 12% to 32% when laparoscopy is performed to determine the cause of pelvic pain in reproductive-age women, and 21% to 48% of women undergoing laparoscopy for infertility. Endometriosis is found in 50% of teenagers undergoing laparoscopy for evaluation of chronic pelvic pain or dysmenorrhea. The influence of age, socioeconomic status, and race on the prevalence of endometriosis remains controversial. The age at time of diagnosis is commonly 25 to 35 years, and endometriosis rarely is diagnosed in postmenopausal women. Many believe that endometriosis is more common in women of upper economic classes because they delay pregnancy, which is postulated to increase the risk of developing endometriosis. It is unknown whether this reflects a true increased incidence or results from greater access to medical care. Evidence indicates that blacks have a prevalence of endometriosis similar to that in whites when controlled for socioeconomic status. Pathology The most common sites of endometriosis, in decreasing order of frequency, are the ovaries, anterior and posterior cul-de-sac, posterior broad ligaments, uterosacral ligaments, uterus, fallopian tubes, sigmoid colon, appendix, and round ligaments. Other sites less commonly involved include the vagina, cervix, and rectovaginal septum. These latter lesions usually result from extension and invasion of posterior cul-de-sac implants. Uncommon locations include the inguinal canal, abdominal or perineal scars, ureters, urinary bladder, umbilicus, kidney, lung, liver, diaphragm, vertebrae, and extremities. Macroscopic Appearance Endometriotic implants have a variety of appearances. Superficial lesions on the ovarian or peritoneal surface are commonly reddish maculae or nodules similar in consistency to normal endometrium. These implants vary from 1 millimeter to several centimeters in size. Collection of hemosiderin results in yellow-brown or black discoloration (“powder-brown” lesions). Nonpigmented disease appears as whitish opacified peritoneum, translucent blebs, or pinkish polyploid implants. Scarring with retraction of adjacent peritoneum and peritoneal pockets may occur. Endometriosis also may appear as a deeply infiltrative disease. Tumorlike masses form from invasion, and diffuse fibrosis usually develops in the posterior cul-de-sac, pelvic sidewall, or posterior broad ligament and ovary and may extend deep into the retroperitoneal space, occasionally constricting the ureter. Lesions in the cul-de-sac may invade the rectovaginal septum. The rectosigmoid and small bowel may become adherent to these areas. Endometriotic foci on the ovarian surface may develop a fibrous enclosure and manifest cyst formation as a result of accumulation of fluid and blood. These endometriotic cysts (endometriomas) vary from several millimeters to over 10 centimeters in size. Bleeding with menses gives the cyst a dark red or bluish hemorrhagic color. The degradation of blood pigment over time results in thick, tarry contents, hence the term chocolate cysts. Occasionally, the content changes to a yellow straw color or clear fluid. Filmy or dense fibroid adhesions from these cysts to the pelvic sidewall and fallopian tubes are common and may obscure visualization of the cyst. Microscopic Appearance Endometriosis is histomorphologically similar to eutopic endometrium. The four major components of endometriotic implants are endometrial glands, endometrial stroma, fibrosis, and hemorrhage. The relative amount of each component is highly variable and dependent, in part, on the age and location of the lesions. Identifying the endometrial elements in individual implants requires an adequate tissue specimen, proper orientation, and often serial sections of the specimen. The endometrial glands in ectopic implants lack uniform size and shape. The glands may show normal cyclic change with mitotic figures and pseudostratification in response to estrogen, or vacuoles and intraluminal secretion in response to progesterone. The response to endogenous and exogenous hormones is inconsistent. This may imply differences in steroid hormone receptor content and function or a loss of the normal gland-stromal interaction. When glands are responsive, the epithelium becomes attenuated, and hemorrhage ensues at the time of menstruation. The stromal cell morphologies of ectopic and eutopic

endometrium are similar. Small arterioles, similar to the spiral arterioles of normal endometrium, usually are present in implants. Interstitial hemorrhage with accumulation of blood products and hemosiderin-laden macrophages is a frequent finding. Fibrosis may occur in older endometriotic implants. This is very common in the lining of endometriomas, where the only histologic finding may be fibroblast proliferation and hemosiderin pigment deposition. Symptoms The common signs and symptoms of endometriosis are pelvic pain, dysmenorrhea, dyspareunia, abnormal uterine bleeding, and infertility. The type and severity of symptoms are dependent on the extent of disease, the location, and the organs involved. Even limited amounts of disease may cause significant symptomatology. Endometriosis is present in approximately one third of patients with chronic pelvic pain. The pain may be described as crampy, dull, or sharp and usually increases around menses. The discomfort may be unilateral or bilateral, and many patients complain of rectal pressure or low backache. Acute abdominal pain may result from hemorrhage secondary to a ruptured endometrioma. Dysmenorrhea is a more frequent complaint than dyspareunia. There is some correlation between the extent of disease and the severity of pain. The morphologic appearance of an endometriotic implant appears to be unrelated to pain symptomatology. Dyspareunia is more common in women with invasive endometriotic nodules in the cul-de-sac, uterosacral ligaments, rectovaginal septum, and vagina. Abnormal uterine bleeding occurs in up to one third of women with endometriosis with symptoms of oligomenorrhea, polymenorrhea, and midcycle or premenstrual spotting. The abnormal bleeding likely results from conditions associated with endometriosis: oligoanovulatory, luteinized unruptured follicles, luteal phase defects, and other pathology such as uterine fibroids. Endometriosis involving the gastrointestinal or urinary tracts and extrapelvic sites causes symptoms characteristic of the location of disease. Bladder involvement is associated with frequency and urgency. Invasion of the mucosa results in hematuria. Ureteral and rare cases of renal endometriosis occasionally cause flank pain or gross hematuria. Symptoms suggestive of gastrointestinal involvement include, in decreasing order of frequency, diarrhea, rectal bleeding, constipation, and dyschezia. All symptoms usually are exacerbated catamenially. There are numerous case reports of extrapelvic endometriosis. Pulmonary endometriosis causes catamenial hemoptysis and dyspnea. Cutaneous lesions are associated with catamenial bleeding, tenderness, and swelling. It is estimated that 25% to 50% of infertile women have endometriosis, and 30% to 50% of women with endometriosis are infertile. Although the association of endometriosis and infertility is well recognized, the pathophysiologic mechanisms are poorly understood. Endometriomas and endometriosis with adhesions distort pelvic anatomy and impair tubal ovum pickup, which is an acceptable explanation for infertility. In less severe cases, there are several theories to explain the observed subfecundity ( Table 40.1).

TABLE 40.1. Proposed mediators and mechanisms of infertility

Research to explain the subfertility has focused on peritoneal fluid leukocytes and their cytokine products. Studies have suggested that constituents in the peritoneal fluid inhibit sperm function, fertilization, embryonic development, and implantation. The clinical significance of these findings has not been established. Diagnosis Endometriosis usually is diagnosed during the third and fourth decades of life. It has not been found in prepubertal girls and rarely is diagnosed in postmenopausal women unless they are taking replacement hormones. Endometriosis should be suspected in any woman having the classic symptoms of pelvic pain, dysmenorrhea, dyspareunia, abnormal menstrual bleeding, and infertility. These symptoms are present in other gynecologic disorders. No one constellation of signs or symptoms is pathognomonic of endometriosis. Many women with endometriosis are completely asymptomatic, and endometriosis should be considered in all reproductive-age women with infertility or an adnexal mass. Physical findings in women with endometriosis are variable and dependent on the location and severity of disease ( Table 40.2). Frequently there are no obvious findings on pelvic examination. When findings are present, the most common is tenderness when palpating the posterior fornix. Nodules of endometriosis on the uterosacral ligaments, enlarged ovaries as a result of endometriotic cysts, and a uterus fixed in the cul-de-sac by adhesions may be detected during a pelvic examination. Uterosacral implants are best palpated during a rectovaginal examination. On the other hand, some patients with these clinical findings turn out not to have endometriosis.

TABLE 40.2. Clinical signs

The optimal way to diagnose endometriosis is by direct visualization of the site of suspected involvement. Because endometriosis is located primarily in the pelvis, laparoscopy is the preferred technique to make an accurate diagnosis. A double-puncture technique is necessary to view adequately all structures that may contain implants. Peritoneal fluid should be aspirated to see the entire cul-de-sac. Adhesions should be lysed to view the entire surface of the ovaries and the fossa ovarica. These sites are commonly involved with endometriosis when the ovary is adherent to the pelvic sidewall. Suspected endometriomas should be aspirated and resected to confirm the diagnosis. Biopsy and histologic study of any suspicious areas are helpful when the diagnosis is questionable, but often the visual diagnosis by the surgeon is more accurate than histologic sections of small peritoneal biopsies. Transvaginal ultrasonography can be used to identify ovarian endometriomas, but it is of little utility to diagnose peritoneal implants. The use of other radiologic studies and blood tests to diagnose endometriosis rarely is required. Radioimmunoassay for the tumor marker CA-125 has been used, but the test is not sufficiently sensitive or specific, and patients having conditions other than endometriosis may have positive results. Classification A number of classifications have been developed for staging endometriosis. The most widely used system was introduced by the American Society for Reproductive Medicine (ASRM) in 1979 and revised in 1985 and in 1997. This system assigns a point score for the size and location of endometriotic implants and associated adhesions. The new ASRM endometriosis classification for infertility includes the morphologic appearance of the implant. There is a form published by the ASRM to assist in the management of endometriosis in the presence of pelvic pain. Endometriosis is classified as minimal, mild, moderate, and severe. Mild disease is characterized by superficial implants less than 5 square centimeters in aggregate scattered on the peritoneum and ovaries. Minimal or no adhesions are present. Moderate forms are characterized by multiple implants, both superficial and invasive. Peritubal and periovarian adhesions may be evident. Severe forms are characterized by multiple superficial and deep implants, including large ovarian endometriomas.

Filmy and dense adhesions are usually present. However, no staging system has been validated to correlate with the symptoms of pain or infertility. Treatment The treatment of endometriosis is dependent on (a) the severity of symptoms, (b) the extent of disease, (c) the location of disease, (d) the patient's desire for pregnancy, and (e) the age of the patient. Treatment options are presented in Table 40.3.

TABLE 40.3. Treatment options Expectant Management Avoiding specific therapy is considered when patients have minimal or no symptoms and have suspected minimal or mild endometriosis. Patients in this category may benefit from cyclic oral contraceptives to retard progression of the disease and protect against unwanted pregnancy. Minor pain may be controlled by nonsteroidal antiinflammatory drugs and/or analgesics. Infertile women having suspected limited disease may be observed without treatment. One study suggests that surgical treatment of mild endometriosis results in higher pregnancy rates than expectant management. Another study with fewer subjects failed to confirm the benefit of laparoscopic surgery for infertile women with endometriosis. If pregnancy occurs, regression or complete resolution of the disease is common. Perimenopausal women may be managed expectantly even when the disease is advanced, because endometriotic implants usually regress in the absence of ovarian hormone production after menopause. Medical Therapy Endometriotic implant growth is highly dependent on ovarian steroids. Medical therapy attempts to “induce” pseudopregnancy or menopause, the two physiologic states believed to inhibit or delay progression of endometriosis by interrupting cyclic ovarian hormone production. Progestins alone or in combination with estrogen hormonally mimic pregnancy. Danazol and gonadotropin-releasing hormone (GnRH) analogs induce a state of pseudomenopause. Medical therapy has the following advantages over surgery: (a) avoidance of the surgical risks of damaging pelvic organs and causing postoperative adhesions, and (b) treatment of implants that are not visible at surgery. Disadvantages of medical therapy are the associated side effects, high recurrence rates following discontinuation of treatment, lack of an effect on endometrioma and adhesions, and inability to conceive because of medically induced anovulation. Medical therapy never has been shown to enhance fertility ( Fig. 40.1). Thus, it is not appropriate for women with advanced stages of endometriosis and adhesions causing symptoms or women desiring pregnancy. The medications used most commonly to treat endometriosis are continuous oral contraceptives, progestins, including oral contraceptives, danazol, and GnRH analogs. They should be considered after a definitive diagnosis of endometriosis has been made by direct visualization of the implants.

FIG. 40.1. Meta-analysis of studies comparing medical treatment with no treatment. Progestins inhibit endometriotic tissue growth by a direct effect on the implants, causing initial decidualization and eventual pseudodecidual necrosis or atrophy (progestins). Progestins also inhibit pituitary gonadotropin secretion and ovarian hormone production. Treatment may consist of medroxyprogesterone acetate (10 mg 3 times a day) or norethindrone acetate (5 mg daily for 2 weeks), then increase by 2.5 milligrams per day every 2 weeks until a daily dose of 15 milligrams is reached. Depot-medroxyprogesterone may also be given as an injection (100 to 150 mg monthly). Treatment usually is continued for at least 6 months. Side effects include irregular menstrual bleeding, nausea, breast tenderness, fluid retention, and depression. The effectiveness of continuous oral contraceptives or progestins in eliminating implants and the risk of recurrent endometriosis following treatment are not precisely known. Over 80% of women have partial or complete pain relief. Low-dose cyclic oral contraceptives are effective in relieving dysmenorrhea in women with endometriosis, but they are less likely to relieve dyspareunia. Pregnancy rates in patients with less severe stages of disease are equivalent to those following expectant management. Danazol is the isoxazole derivative of 17a-ethinyltestosterone. Danazol has the following three mechanisms of action: (a) inhibition of pituitary gonadotropin secretion, (b) direct inhibition of endometriotic implant growth, and (c) direct inhibition of steroidogenic enzymes. Danazol is given orally in divided doses, from 400 to 800 milligrams daily, generally for 6 months. Most women taking danazol have side effects, but only a small percentage of patients discontinue the drug because of unwanted effects. Side effects, in decreasing order of frequency, include weight gain, muscle cramps, decreased breast size, acne, hirsutism, oily skin, decreased high-density lipoprotein levels, increased liver enzyme levels, hot flashes, mood changes, and depression. Danazol decreases the size of implants, especially in treating mild or moderate stages of disease. Endometriomas and adhesions do not respond well to danazol treatment. More than 80% of patients experience relief or improvement of pain symptoms within 2 months of treatment. Pregnancy rates following treatment approximate 40% and are independent of the disease severity. However, danazol is no more effective than expectant management for treating infertility. The GnRH analogs profoundly suppress ovarian estrogen production by inhibiting pituitary gonadotropin secretion. These medications are administered by nasal spray or depot injections. The usual dosage is 400 to 800 milligrams daily for nasal nafarelin, 3.6 milligrams for monthly subcutaneous goserelin, and 3.75 milligrams for monthly intramuscular leuprolide. Side effects of the hypoestrogenemia are common and include hot flashes, vaginal dryness, decreased libido, insomnia, breast tenderness, depression, headaches, and transient menstruation. In addition, GnRH analog treatment for the recommended 6-month period decreases bone density and total body calcium, but most of the bone loss is reversible. Hypoestrogenic side effects and bone loss may be prevented by “add-back” therapy with high-dose norethindrone (10 mg daily) or the daily combination of low-dose norethindrone (2.5 mg), sodium etidronate (400 mg), and calcium carbonate (500 mg) ( Table 40.4). The GnRH analogs effectively reduce the size of endometriotic implants, even with add-back therapy. Recurrence rates over 5 years range from 37% for patients with mild disease to 74% for severe disease. The GnRH analogs are as affective as other medical therapy in relieving pain symptoms, but they do not enhance fertility.

TABLE 40.4. Endometriosis “add-back” therapy Surgical Management Surgery for endometriosis is considered conservative when the uterus and as much ovarian tissue as possible are preserved. Definitive surgery involves hysterectomy with or without removal of the fallopian tubes and ovaries. Surgery is indicated when the symptoms are severe, incapacitating, or acute and when the disease is advanced. Surgery is preferred over medical therapy for advanced stages of disease with anatomic distortion of the pelvic organs, endometriotic cysts, or obstruction of the bowel or urinary tract. Women who are older than 35 years, infertile, or symptomatic following expectant or medical management should be treated surgically. Laparoscopy is the preferred approach to perform conservative surgery. Treatment of endometriosis is possible during the initial laparoscopy, used to diagnose the condition. This offers the advantage of ablating the implants and adhesions, while avoiding possible progression of disease or symptoms and the expense and side effects of medical therapy. Disadvantages include possible damage to bowel and bladder, infection, and mechanical trauma that may result in adhesion formation. Conservative surgery involves excision, fulguration, or laser ablation of endometriotic implants and removal of associated adhesions. The goal is to restore normal pelvic anatomy. Laparoscopic treatment offers advantages over laparotomy, including shorter hospitalization, anesthesia, and recuperation times. Laparotomy is advisable to deal with extensive adhesions or invasive endometriosis located near structures such as the uterine arteries, ureter, bladder, and bowel. Ancillary procedures include presacral neurectomy or uterosacral transection for interruption of sensory nerves innervating the pelvis to relieve midline pelvic pain. The Cochrane Review of three trials involving destruction of pelvic nerve pathways concluded that there is insufficient evidence to recommend these procedures. Uterine suspension may be performed to avoid adhesion formation from the cul-de-sac to the posterior surface of the uterus, tubes, and ovaries. Surgery effectively removes pathology and restores normal anatomy in most cases. The disease recurrence risk is estimated to be as much as 40% with 10 years of follow-up. Pain relief is achieved in 80% to 90% of patients. Presacral neurectomy provides additional pain relief, but its benefit is not lasting, and bladder dysfunction occasionally occurs after the procedure. The chance for pregnancy following surgery is related to the stage of disease and presence of other infertility factors. Approximate pregnancy rates after surgery in patients with mild, moderate, and severe endometriosis are, respectively, 60%, 50%, and 40%. Surgery is preferred over expectant or medical management for infertile women with endometriosis ( Fig. 40.2).

FIG. 40.2. Meta-analysis of studies comparing surgical treatment (operative laparoscopy or laparotomy) with nonsurgical treatment (medical treatment or no treatment). Definitive surgery for treatment of endometriosis is indicated when significant disease is present and pregnancy is not desired, when incapacitating symptoms persist following medical therapy or conservative surgery, and when coexisting pelvic pathology requires hysterectomy. The decision to perform hysterectomy is dependent primarily on the patient's interest in maintaining childbearing potential and the severity of her symptoms. The ovaries may be conserved in younger women to avoid the need for estrogen replacement therapy. Removal of both ovaries is appropriate when the ovaries are damaged extensively by endometriosis or when menopause is approaching. Endometriosis may recur even with castration, presumably from microscopic foci of disease not visible at surgery. Menopausal hormonal replacement is indicated when the ovaries are removed, even when surgery has not removed all endometriotic implants. The chance for symptomatic recurrence in these cases is small except when endometriosis involves the bowel. Combination Medical and Surgical Therapy Medical therapy is used before surgery to decrease the size of endometriotic implants and thus reduce the extent of surgery. When complete removal of implants is not possible or advisable, postoperative medical therapy is used to treat residual disease. Progestin, danazol, or GnRH analogs may be used in conjunction with conservative or definitive surgery. Preoperative medical therapy may decrease the amount of surgical dissection required to remove implants, but it does not prolong pain relief, increase pregnancy rates, or decrease recurrence rates. Postoperative treatment with GnRH analogs will delay somewhat the recurrence of pelvic pain, but there is no evidence to support its use in infertile patients.

SUMMARY POINTS The histogenesis of endometriosis is poorly understood, but emerging evidence supports the causative role of retrograde menstruation and implantation of endometrial tissue. Endometriosis is common in women with pelvic pain or infertility. Laparoscopy is the optimal technique to diagnose pelvic endometriosis. In most cases, surgical therapy at the time of initial diagnosis effectively relieves pain and may enhance fertility. Alternatively, medical therapy with progestins, danazol, or GnRH analogs will ameliorate pelvic pain, but they do not enhance fertility. Endometriosis is a recurrent disease, and definitive treatment with removal of pelvic organs may be necessary. SUGGESTED READINGS Pathogenesis Dmowski WP, Gebel HM, Braun DP. The role of cell-mediated immunity in pathogenesis of endometriosis. Acta Obstet Gynecol Scand Suppl 1994;159:7. Olive DL, Henderson DY. Endometriosis and múllerian anomalies. Obstet Gynecol 1987;69:412. Oosterlynck DJ, Cornillie FJ, Waer M, et al. Women with endometriosis show a defect in natural killer activity resulting in a decreased cytotoxicity to autologous endometrium. Fertil Steril 1991;56:45. Schenken RS. Pathogenesis. In: Schenken RS, ed. Endometriosis: contemporary concepts in clinical management. Philadelphia: JB Lippincott, 1989:1. Steele RW, Dmowski WP, Marmer DJ. Immunologic aspects of human endometriosis. Am J Reprod Immunol 1984;6:33.

Epidemiology Chatman DL, Ward AB. Endometriosis in adolescents. J Reprod Med 1982;27:156. Houston DE. Evidence for the risk of pelvic endometriosis by age, race, and socioeconomic status. Epidemiol Rev 1984;6:167. Olive DL, Schwartz LB. Endometriosis. N Engl J Med 1993;328:1759. Sangi-Haghpeykar H, Poindexter AN. Epidemiology of endometriosis among parous women. Obstet Gynecol 1995;85:983.

Pathology Jenkins S, Olive DL, Haney AF. Endometriosis: pathogenetic implications of the anatomic distribution. Obstet Gynecol 1986;67:335.

Diagnosis Schenken RS, Vancaillie TG, Riehl RM, et al. New developments in diagnostic techniques. In: Chadha DR, Buttram VC Jr, eds. Current concepts in endometriosis. New York: Alan R. Liss, 1990:137. Vercellini P, Trespidi L, De Giorgi O, et al. Endometriosis and pelvic pain: relation to disease stage and localization. Fertil Steril 1995;65:299.

Classification American Society for Reproductive Medicine. Revised American Society for Reproductive Medicine classification of endometriosis. Fertil Steril 1997;67:817–821.

Treatment Adamson GD, Hurd SJ, Pasta DJ, et al. Laparoscopic endometriosis treatment: Is it better? Fertil Steril 1992;59:35–44. Adamson GD, Pasta DJ. Surgical treatment of endometriosis-associated infertility: meta-analysis compared with survival analysis. Am J Obstet Gynecol 1994;171:1488. Barbieri RL, Evans S, Kistner RW. Danazol in the treatment of endometriosis: analysis of 100 cases with a 4-year follow-up. Fertil Steril 1982;37:737. Crosignani PG, Vercellini P, Biffignandi F, et al. Laparoscopy versus laparotomy in conservative surgical treatment for severe endometriosis. Fertil Steril 1996;66:706. Dawood MY, Ramos J, Khan-Dawood FS. Depot leuprolide acetate versus danazol for treatment of pelvic endometriosis: changes in vertebral bone mass and serum estradiol and calcitonin. Fertil Steril 1995;63:1177. Dlugi AM, Miller JD, Knittle J, Lupron Study Group. Lupron depot (leuprolide acetate for depot suspension) in the treatment of endometriosis: a randomized, placebo-controlled, double-blind study. Fertil Steril 1990;54:419. Gruppo Italiano per lo Studio dell'Endometriosi. Ablation of lesions or no treatment in minimal-mild endometriosis in infertile women: a randomized trial. Hum Reprod 1999;14:1332. Hughes E, Fedorkow D, Collins, J, et al. Ovulation suppression for endometriosis (Cochrane Review). In: Cochrane Database of Systemic Reviews. The Cochrane Library; Issue 1. Oxford: Update Software; 2002. Macroux S, Haheux R, Berube S, et al. Laparoscopic surgery in infertile women with minimal or mild endometriosis. N Engl J Med 1997;337:217. Olive DL. Medical treatment: alternatives to danazol. In: Schenken RS, ed. Endometriosis: contemporary concepts in clinical management. Philadelphia: JB Lippincott, 1989;189. Parazzini F, Fedele L, Busacca M, et al. Postsurgical medical treatment of advanced endometriosis: results of a randomized clinical trial. Am J Obstet Gynecol 1994;171:1205.

Proctor ML, Farquhar CM, Sinclair OJ, et al. Surgical interruption of pelvic nerve pathways for primary and secondary dysmenorrhoea (Cochrane Review). In: Cochrane Database of Systemic Reviews. The Cochrane Library; Issue 1. Oxford: Update Software; 2002. Redwine DB. Endometriosis persisting after castration: clinical characteristics and results of surgical management. Obstet Gynecol 1994;83:405. Surrey ES, Voigt B, Fournet N, et al. Prolonged gonadotropin-releasing hormone agonist treatment of symptomatic endometriosis: the role of cyclic sodium etidronate and low-dose norethindrone add-back therapy. Fertil Steril 1995;63:747. Vercellini P, Trespidi L, Colombo A, et al. A gonadotropin releasing hormone agonist versus a low-dose contraceptive for pelvic pain associated with endometriosis. Fertil Steril 1993:60:75. Waller KG, Shaw RW. Gonadotropin-releasing hormone analogues for the treatment of endometriosis: long-term follow-up. Fertil Steril 1993;59:511. Wheeler JM, Malinak LR. Recurrent endometriosis: incidence, management, and prognosis. Am J Obstet Gynecol 1983;146:247.

Chapter 41 Menopause Danforth’s Obstetrics and Gynecology

Chapter 41 Marcelle I. Cedars and Michele Evans

Menopause

REPRODUCTIVE STAGES Oocyte Depletion Endocrinology Systemic Effects of Declining Ovarian Function PREMATURE OVARIAN FAILURE Menstrual Cycle Menopausal Syndrome HORMONE REPLACEMENT Introduction Benefits Risks CONTROVERSIES Cardiovascular Disease Breast Cancer Selective Estrogen Receptor Modulators SUMMARY SUMMARY POINTS SUGGESTED READINGS General Premature Ovarian Failure Vasomotor symptoms Genitourinary Atrophy Osteoporosis Cardiovascular Disease Breast Cancer Cognitive Function Selective Estrogen Receptor Modulators Controversies Treatment Regimens

The ovary is unique in that the age associated with decline in function (to frank failure) appears to have remained constant despite the increase in longevity experienced by women over the last century. Because the loss of ovarian function has profound impact on the hormonal milieu in women and on the subsequent risk for the development of disease via the loss of estrogen production, improving our understanding of reproductive aging is critical to care for all women. Human follicles begin their development during the fourth gestational month. Approximately 1,000 to 2,000 germ cells migrate to the gonadal ridge and multiply, reaching a total of 5 to 7 million around the fifth month of intrauterine life. At this point, replication stops and follicle loss begins, declining to approximately 1 million by birth. In the human male, the dividing germ cells will become quiescent and maintain their stem cell identity. In women, between the weeks of 12 and 18, the germ cells will enter meiosis and differentiate. Thus, in the female, all germ stem cells have differentiated prior to birth. In the adult woman, the germ cells may remain quiescent, be recruited for further development and ovulation, or undergo apoptosis. Over time, the population of oocytes will be depleted, without regeneration, through recruitment and apoptosis until fewer than a thousand oocytes exist and menopause ensues. Approximately 90% of women experience menopause during the early 50s. The other 10% of women experience menopause prior to 46 years (often termed early menopause), with 1% of women experiencing menopause at an age less than 40 years (premature menopause or premature ovarian failure). Menopause occurs at a median age of 51.4 years, with the age range in normal women being 42 to 58 years. The age of menopause appears to be determined largely by genetics and is due to exhaustion of the oocyte pool. Menopause and the years preceding are characterized by hormonal changes, decline in reproductive potential, and increased risk for physical and psychological changes. The average age of menopause has remained constant throughout recorded history. It does not appear to be related significantly to race, parity, height, weight, socioeconomic status, or age at menarche. Evidence suggests that genetic and environmental factors influence age of menopause, although the specific nature of these relationships is poorly characterized. Given the strong association between age at menopause between mothers and daughters, this is likely a largely genetically determined trait. Environmental factors may not have a significant effect in themselves, but the interplay among environmental factors such as smoking (known to accelerate the age of menopause by 1.5–2 years), body mass index (BMI), alcohol use, and socioeconomic status and genetic risk, may be important. According to the 2000 census data, 35% of the population is age 45 or older and 21% are over 55 years of age. Of the population, 7.3% are women 65 years or older (approximately 20 million women in the United States). Over 50 million women in the United States are in the menopausal transition or menopause.

REPRODUCTIVE STAGES Reproductive aging is a continuum beginning in utero and ending with menopause. The stages along this continuum have been difficult to define. Numerous terms have been used clinically, including perimenopause, menopausal transition, climacteric, menopause, and postmenopause, to describe the end of this continuum. The Stages of Reproductive Aging Workshop (STRAW) was convened in July 2001 to address the lack of a pertinent reproductive staging system and to establish a nomenclature and guidelines, a consistent reproductive aging system for health practitioners, the medical research community, and the public. The staging system developed takes into account menstrual cyclicity, endocrinology and symptomatology beginning with menarche and ending with a woman's demise. The foundation of the staging system is the final menstrual period (FMP). Five stages precede the FMP and two follow it, for a total of seven stages. Stages -5 to -3 are called the reproductive interval, stages -2 to -1 are termed the menopausal transition, and stages +1 and +2 are known as the postmenopause ( Fig. 41.1).

FIG. 41.1. Stages of reproductive aging. (From Soules MR, Sherman S, Parrott E, et al. Stages of Reproductive Aging Workshop (STRAW). J Womens Health Gend Based Med 2001;10:843–848, with permission.)

The menopausal transition begins with variation in the menstrual cycle length (>7 days different from normal) in a woman with an elevated follicle stimulating hormone (FSH) level. This stage ends with the FMP, which cannot be determined conclusively until after 12 months of amenorrhea. Early postmenopause is defined as the first 5 years following the FMP. Late postmenopause is variable in length, beginning 5 years after the FMP and ending with the woman's death. Although this staging system is said to include endocrinologic aspects of ovarian aging, it still depends largely on menstrual cyclicity as a key indicator of ovarian aging. It does include measurement of FSH; however, by the time FSH is elevated, even with cyclic menstrual cycles, oocyte depletion already has proceeded to such an extent that fertility (as a marker of reproductive aging) is diminished significantly. As noted above, evidence suggests that genetic and environmental factors influence both age of menopause and the decline in fertility, although the specific nature of these relationships is characterized poorly. Premature menopause can be due to a failure to attain adequate follicle numbers in utero or to an accelerated depletion thereafter. Potentially, either of these factors could be affected by genetic and

environmental risk. The timing of menopause has a consistent impact on overall health with respect to osteoporosis, cardiovascular disease (CVD), and cancer risk. Over the next decade, it's estimated that more than 40 million women will enter menopause. Oocyte Depletion As discussed above, the leading theory regarding the onset of menopause relates to a critical threshold in oocyte number. Approximately 1,000 to 2,000 germ cells migrate to the gonadal ridge and multiply, reaching a total of 5 to 7 million around the fifth month of intrauterine life. At this point, replication stops and follicle loss begins, with a reduction to approximately 1 million by birth and 500,000 to 600,000 by menarche. As the number of oocytes in the reserve pool continues to decline, menstrual irregularity, followed by cessation, will occur. The theory that menopause is triggered primarily by ovarian aging is supported by the coincident occurrence of follicular depletion, elevated gonadotropin levels, and subsequent menstrual irregularity with ultimate cessation of bleeding. A mathematical model that predicts the rate of follicular decline has been developed ( Fig. 41.2). It utilized existing data, which ultimately showed a bi-exponential decline, with an acceleration in oocyte loss when the remaining oocyte number equaled approximately 25,000. In this model, the decline occurred at 37.5 years of age. At this point, the rate of follicular atresia accelerates. In the absence of this acceleration, the model suggests menopause would be delayed until age 71. The cause of this accelerated depletion is not well defined. It is also clear that if the factor influencing the rate of decline is follicle number and not age, other factors which might account for a diminished follicle number (genetic risk and possible toxic exposure) would lead to an earlier rate of accelerated decline and an earlier age of menopause.

FIG. 41.2. Bi-exponential model of declining follicle numbers in pairs of human ovaries from neonatal age to 51 years old. Data were obtained from the studies of Block (1952, 1953) ( x, n = 6; +, n = 43), Richardson and others (1987) (squares, n = 9), and Gougeon (unpublished) (*, n = 52). (From Faddy MJ, Gosden RG, Gougeon A, et al. Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 1992;7:1342–1346, with permission.)

Coincident with the decline in the number of follicles in the ovary, there appears to be an increase in random genetic damage within these structures. Evidence for this comes from an observed increase in aneuploidy in the offspring of older mothers and the observation that, in women over 40 years old, oocytes harvested for in vitro fertilization are karyotypically abnormal approximately 40% of the time. Further examples in nature, such as Turner syndrome, shed some light on the process of oocyte aging. Individuals with this syndrome are, by and large, born with dysgenetic gonads which are devoid of follicles. Ninety-five percent of these individuals are aborted spontaneously prior to birth. If one examines the ovaries of a 20-week abortus, a full compliment of oocytes is present. Two factors have been isolated from the ovary: oocyte maturation inhibiting factor (OMI) and luteinizing inhibitor, which may control the rate of maturation of follicles. It has been suggested that individuals with dysgenetic gonads may fail to produce adequate OMI, thus allowing all follicles to progress prematurely toward maturity. The control mechanisms are conceptual rather than factual at this juncture, and new information will have to accumulate before the factors governing human oocyte atresia are elucidated more clearly. Endocrinology The entire endocrine system changes with advancing age. The somatotrophic axis begins to decline during the fourth decade prior to the loss of ovarian function. This decline in growth hormone is accelerated during ovarian failure and may, itself, accelerate the ovarian failure. However, pituitary concentrations of growth hormone, as well as adrenocorticotropic hormone and thyroid stimulating hormone, remain constant into the ninth decade. Although the thyroid gland undergoes progressive fibrosis with age and concentrations of T3 decline by 25% to 40%, elderly patients remain clinically euthyroid. Beta-cell function also undergoes degeneration with aging such that, by age 65 years, 50% of subjects have an abnormal glucose tolerance test result. Frank diabetes is rare, however, occurring in only 7%. The female reproductive system, on the other hand, undergoes virtually complete failure at a relatively early age. As noted above, during the late fourth decade, FSH levels begin to rise even when cyclic menses continue. The most likely cause is a decrease in functional granulosa cells from the oocyte pool, with a decrease in inhibin B negative feedback allowing a monotropic rise in FSH. Early on, there is also a decline in luteal phase progesterone levels. As ovarian aging progresses, estradiol levels may be quite variable, with chaotic patterns and, occasionally, very high and very low levels. This dramatic variability may lead to an increase in symptomatology during the perimenopause (stages -2 to -1). As peripheral gonadotropin levels rise, luteinizing hormone (LH) pulsatile patterns become abnormal. There is an increase in pulse frequency with a decrease in opioid inhibition. Estrogens The main circulating estrogen during the premenopausal years is 17ß-estradiol. These levels are controlled by the developing follicle and resultant corpus luteum. Oophorectomy will reduce peripheral estradiol levels from 120 to 18 pg/mL, which suggests more than 95% of circulating estradiol is derived from the ovary. Other sources include the peripheral conversion of testosterone and estrone. Very small amounts are secreted by the adrenal gland. Because the two-cell theory requires aromatization of androgens produced by the theca in the granulosa cell, follicular exhaustion is associated with gradual declines in estradiol concentrations. The predominant estrogen in the postmenopausal woman is estrone, which has a biologic potency of approximately one third that of estradiol. Estrone is derived largely from peripheral conversion of androstenedione. Extraglandular aromatase is found in liver, fat, and some hypothalamic nuclei. This activity increases with aging and with a higher fat content (also an age-related change). Estrone and estradiol production rates during the postmenopause are 40 and 6 µg/day, respectively. This compares with 80 to 500 µg/day for estradiol during the reproductive years. Essentially, all the estradiol in the postmenopausal women is derived from conversion of estrone. Androgens Dehydroepiandrosterone (DHEA) and its sulfated conjugate, DHEAS, have been shown to decrease, along with adrenal corticotropin responsiveness, with aging. DHEAS levels decrease in both men and women. The decline is greater in women and may be due to the relative estrogen deprivation. Ovarian failure, at any age, accelerates this decline. Evidence suggests physiologic levels of DHEA may protect against neoplasia, enhance insulin action, protect against osteoporosis, increase immune competency, and offer some cardioprotection. Changes in DHEA levels also have been associated with alterations in body composition which, in themselves, appear to impact cardiac and breast cancer risk. DHEAS levels may also have an impact on “sense of mental well-being.” Androstenedione is the predominant androgen during the reproductive years and production declines from 1,500 to 800 pg/mL in postmenopausal women. The postmenopausal ovary contributes only 20% to the circulating androstenedione. Testosterone levels also decline after menopause, but not to the same extent as estradiol levels. Postmenopausal testosterone is derived from the ovary (25%), the adrenal gland (25%), and extraglandular conversion from androstenedione (50%). The postmenopausal ovary produces a larger percentage of testosterone (50%) than does the premenopausal ovary. Systemic Effects of Declining Ovarian Function The decline in ovarian function brings about profound changes in secondary sexual organs. The endometrium becomes atrophic and the uterus decreases in size. Evidence is accumulating in animal models that the uterus may be partially responsible for the initial decline in reproductive capacity. On the other hand, data from human oocyte donor programs have shown that transfer of ova from younger donors to menopausal recipients produced normal gestations and offspring. It should be noted that these women are stimulated with an artificial sequential overlapping regimen of estrogen and progesterone. This produces an endometrium that is indistinguishable from that of the premenopausal state. Our own data have shown high implantation rates in older women with a hormonally induced endometrium. Only those women who receive pelvic irradiation have responded poorly, suggesting an alteration of the microvascular system. The postmenopausal vagina, devoid of estrogen treatment, becomes smaller in both length and caliber. There is decreased elasticity of the vaginal wall, and the karyopyknotic index changes to show fewer superficial cells. The fallopian tubes contain both ciliated and secretory components. After age 60, cilia begin to disappear in the isthmic region, although they remain until a very old age in the ampulla infundibulum. The mammary gland develops secretory potential at the time of puberty and maintains this function until menopause. Like other secondary sex organs, it is dependent on female sex steroids for its maintenance. With cessation of the production of estrogen and progesterone, glandular, ductal, and stromal involution occurs. The basement membrane thickens and the luminal space becomes obliterated.

Connective tissue of the lobule becomes indistinguishable from other types of connective tissue. There is an accumulation of adipose tissue in the breast, which occurs simultaneously with this involutional process. Despite the involutional changes of the breast, 6% of patients with breast carcinoma are over the age of 50, with a median of 55. There is evidence for a bimodal distribution of breast carcinoma, with the first peak occurring at 45 years of age and the second at 65 years of age. The portion of estrogen receptor–positive breast cancers increases until ages 60 to 74 years. Therefore, a dichotomy appears to exist that ducts and glands, rapidly undergoing involution, the result of failing steroid production, become susceptible to malignant transformation while retaining a receptor molecule (E 2 ), which normally is self-induced.

PREMATURE OVARIAN FAILURE Premature ovarian failure (POF) is a unique entity in which a woman undergoes changes consistent with menopause, such as amenorrhea, elevated FSH levels, and depletion of ovarian follicles, prior to the age of 40. POF occurs in 0.1% of women under 30 years of age and 1% of women by age 40. Genetic factors are thought to have a strong relationship with POF. There is a higher incidence of family history of early menopause and a suggestive increase in the family history of infertility, as well as an increased incidence of familial cases of early menopause amongst patients with idiopathic POF. Twin studies have likewise noted a strong genetic component to the age of menopause. Although inheritance appears to be either X linked or autosomal dominant sex limited, paternal transmission cannot be excluded. Women who have idiopathic early menopause (between the ages of 40 and 45) have a genetic pattern similar to whose with POF. These observations support the hypothesis of common underlying genetic factors, which may lead to an early decrease in fertility, early menopause, and POF. POF may not be the same as age-appropriate menopause, which results from the depletion of the primordial follicle pool, because POF may be reversible, with follicles in the ovary, estradiol production, and even pregnancies long after the diagnosis. When using ultrasonography to evaluate the follicles in women with POF compared with age-appropriate menopausal and young women on oral contraceptives (OCPs), the mean ovarian volumes were smaller in patients with POF compared with women on OCPs, but not different from the women with age-appropriate menopause. Approximately 40% of patients with POF had follicles in the ovary, albeit fewer than in the normal premenopausal women. Menstrual Cycle Prior to the menopausal transition, the average length of a menstrual cycle ranges from 21 to 35 days. The menopausal transition is defined partially by menstrual irregularity that occurs in response to changes within the ovary, specifically, a dramatic decline in follicle number (and granulosa cell content). As a result, inhibin B levels fall, decreasing negative feedback on FSH, causing a monotrophic rise in FSH. This early cycle rise in FSH may shorten the follicular phase due to accelerated folliculogenesis. Estradiol levels remain relatively constant with age until the menopausal transition, when they initially rise in response to increased FSH levels. As the ovary fails, progesterone levels decline, leading to a shortened (or inadequate) luteal phase. Thus, an early sign of waning ovarian function may be a decreased intermenstrual interval. Precycle spotting may also signal deficiencies in progesterone production. These clinical signs of reproductive aging indicate a poor prognosis for those women still interested in reproduction. As the FMP approaches and the oocyte complement declines to a critical level, estradiol levels fall, leading to hot flashes, vaginal atrophy, and accelerated bone mineral density (BMD) loss. Also, as the FMP approaches, there is a steady trend toward an increased mean cycle length. In a woman's final 10 to 20 cycles, average cycle lengths are characteristically 40 to 42 days. As menstrual irregularity increases during the menopausal transition, many women seek medical care. Treatment can be approached in several ways. After a complete history and physical examination, bleeding irregularities can be treated with different hormonal regimens including OCPs, cyclic hormone replacement therapy (HRT), or progestin-only therapy. Many of these patients continue to ovulate, albeit irregularly, so the addition of cyclic progestin (without estrogen) may further increase cycle irregularity and does not offer contraceptive protection. Thus, treatment with OCPs or a combined, cyclic estrogen-progestin regimen is advisable. There is always the risk of endometrial hyperplasia in this age group. Patients considered high risk (history of chronic anovulation or obesity, or suspicious bleeding patterns such a watery, bloody discharge) should undergo endometrial sampling. Pelvic ultrasonography for measurement of endometrial stripe thickness is not a reliable predictor of risk in cycling women. For those who have a relatively new onset of bleeding irregularity (consistent with the menopausal transition in a previously ovulatory woman), consideration can be given for initial hormonal treatment, with endometrial biopsy reserved only for those whose cycles fail to normalize after 3 months of therapy. Postmenopausal bleeding always should be considered abnormal and must be evaluated accordingly. Bleeding can occur from the rectum, vagina, cervix, urethra, or uterus. A thorough history and physical examination is crucial. If the source of bleeding is uterine, a transvaginal ultrasonographic examination can be very helpful. If the endometrial stripe is thinner than 5 mm, the bleeding is typically the result of an atrophic endometrium. If the endometrium is 5 mm or thicker, it is imperative to perform a diagnostic test, either an endometrial biopsy or dilation and curettage, to sample the endometrium. Menopausal Syndrome Given the above endocrinologic changes with aging, many symptoms associated with aging in women are due to estrogen deficiency, but the decline in adrenal androgens and growth hormone may contribute. Symptoms definitely due to estrogen deprivation include vasomotor symptoms and urogenital atrophy. Osteoporosis is likely, largely due to estrogen deficiency, but this may be exacerbated by the relative growth hormone decline. The same may be said for the hormone-related changes of increasing atherosclerotic CVD and psychosocial symptoms including insomnia, fatigue, short-term memory loss, and depression. Both DHEAS and growth hormone may well have an impact on these age-related symptoms. Vasomotor Symptoms Vasomotor instability in the form of a hot flash (flush) is one of the most consistent and bothersome symptoms that women face as they enter the menopausal transition and subsequent menopause. Hot flashes result from estrogen deficiency and a resetting of the hypothalamic thermoregulatory set point. They occur in 65% to 76% of women who undergo spontaneous menopause or surgical oophorectomy. Symptoms may begin during the menopausal transition, when estrogen levels may fluctuate dramatically from cycle to cycle and even day to day. A hot flash usually is characterized by intense warmth, described as “heat or burning” that usually begins in the head, neck, and thorax and can spread in waves over the entire body. It may be preceded by pressure in the head and may be accompanied by heart palpitations. The hot flash usually is followed by an outbreak of sweating, followed by chills as the body's thermostat resets. The length of the episode varies from seconds to approximately 5 minutes, although episodes as long as 30 minutes have been described. The event frequency varies from a few per year to 30 per day. For most women, the hot flashes commence prior to the FMP, although initially this may be perceived only as a sleep disturbance. In general, the episodes are noted more frequently at night, and the dysfunctional sleep pattern that follows may result in fatigue, irritability, loss of concentration, and depression, symptoms that often are elicited from patients in the menopausal transition. More than 80% of women who experience hot flashes will experience them for longer than 1 year. Twenty-five percent of women complain of severe hot flashes. Without treatment, the symptoms usually subside slowly over 3 to 5 years. Investigators in 25-year longitudinal study from Gothenburg, Sweden, with 1,462 participants, found the prevalence of hot flashes a maximum of 60% at 52 to 54 years of age. Interestingly, 9% of subjects still reported hot flashes at age 72. The etiology of hot flashes seems to be the withdrawal of estrogen rather than the state of hypoestrogenism. For example, women with Turner syndrome who have not been treated with exogenous estrogen do not experience hot flashes. Those who are treated with estrogen, which is later withdrawn, will experience symptoms of vasomotor instability. Obese women seem to be less symptomatic than matched controls with lower BMI. The explanation may be that obese women are less hypoestrogenic secondary to peripheral conversion of adrenal androgens into estrone or that obesity lowers sex hormone-binding globulin levels, allowing a greater proportion of their estrogen to remain unbound and able to act on target tissues. Studies of hot flashes with external monitoring of skin temperature and resistance have shown a frequency of approximately 54 plus or minus 10 minutes. This frequency has been shown to interrupt random eye movement sleep and potentially contributes to some of the psychosocial complaints. Hot flashes are correlated temporally with pulses of LH; however, exogenous LH does not induce a flash, suggesting there is some central mediator leading simultaneously to hot flashes and LH pulses. Several biochemical alterations are associated with the hot flash. During the actual episode, there is evidence of a rise in plasma LH, epinephrine, corticotropin, cortisol, androstenedione, DHEA, ß-lipotropin, ß-endorphin, and growth hormone. Levels of estradiol, estrone, prolactin, thyroid stimulating hormone, FSH, and norepinephrine are unchanged. Treatment Vasomotor symptoms are the most common indication for use of estrogen treatment in menopause and are also a U.S. Food and Drug Administration (FDA)-approved indication. Estrogen therapy, in either oral or transdermal form, has a greater than 95% efficacy for the treatment of hot flashes. Hot flash frequency is first notably reduced after 2 weeks of treatment, and the full effect of a certain dosage can be determined reliably after 4 weeks. Due to concerns over the risk-benefit analysis of HRT use, it is recommended that women be treated for vasomotor symptoms for short periods (1-4 years) and then gradually tapered, because symptoms often recur when treatment is discontinued abruptly. For patients with contraindications to estrogen use, vasomotor symptoms can be treated, albeit less efficiently, with progestins, a2-adrenergic agonists (clonidine, methyldopa, lofexidine) and, possibly, antidepressants (selective serotonin reuptake inhibitors [SSRIs], venlafaxine hydrochloride [Effexor]). The role of serotonin (5-hydroxytryptamine [5-HT]) in symptoms of menopause is being investigated increasingly. Serotonin levels fall with menopause, either naturally or surgically induced, and replacement estrogen is known to increase serotonergic tone. The 5-HT 2A receptor subtype is thought to underlie changes in thermogenesis. Stimulation of this receptor may lead to changes in the set-point temperature, leading to autonomic changes which cool the body. An increased skin temperature and sweating may result. Thus, an involvement for the 5-HT 2A receptor in the etiology of hot flashes has been suggested. A theoretical model, illustrating a role for 5-HT in the mechanism of the hot flash, is shown in Figure 41.3.

FIG. 41.3. Possible mechanism by which a hot flush is induced. (From Berendsen HH. The role of serotonin in hot flushes. Maturitas 2000;36:155–164, with permission.) Most studies using SSRIs for the treatment of vasomotor symptoms have been in patients with breast cancer. The differential between the antidepressant effects of these agents versus a direct effect on vasomotor symptoms may be more difficult to detect in this population. Genitourinary Atrophy Vagina A decrease in circulating estrogen levels has deleterious effects on urogenital epithelium. Up to 50% of postmenopausal women experience symptoms of vaginal atrophy. The most common symptoms include dryness, irritation, itching, burning, and dyspareunia. Atrophic vaginitis is associated with a rise in vaginal pH, which can lead to more frequent infections and worsening, irritative symptoms. A concurrent decrease in vaginal lubrication can lead to bleeding and decreased sexual comfort and pleasure. Estrogen replacement therapy (ERT) is an effective treatment for vaginal atrophy. The systemic dosage necessary for vaginal protection is somewhat higher than needed for bone protection (see below) and, thus, topical therapy by means of creams or vaginal rings may be advisable to limit systemic absorption. Unless systemic HRT is required for vasomotor instability, local estrogen therapy can be used effectively to treat urogenital atrophy. Vaginal estrogen cream or tablets can be used daily for approximately 2 to 3 weeks, and then twice weekly after initial symptoms have improved and vaginal vascularization (hence, hormone uptake) have increased. Treatment is usually long term, because symptoms tend to recur when estrogen is discontinued. The twice-weekly estrogen regimen can be used without supplemental progestin without an increase in endometrial thickness. The dosage should be kept low, however, because the well-vascularized vagina is extremely efficient in the absorption of steroids. The new low-dose vaginal ring may also be used without progestin protection of the endometrium. Vaginal estrogen frequently will improve symptoms of urinary frequency, dysuria, urgency, and post-void dribbling. A direct effect to improve urinary incontinence is less clear. Alternatives to estrogen include vaginal moisturizers and lubricants. There is no evidence to support the use of agrimony, black cohosh, chaste tree, dong quai, witch hazel, or phytoestrogens for the treatment of atrophic vaginitis. Urinary Tract Infections Urinary tract infection (UTI) is common among women of all ages. Worldwide, an estimated 150 million UTIs occur annually. In the United States, UTIs account for more than $6 billion in health care costs. The incidence is highest in 18- to 24-year-old women at 17.5% and is 9% for women older than 50 years. In younger women, the major risk factors for recurrent UTI are sexual intercourse and spermicide exposure. In older, institutionalized women, the most important risk factors include urinary catheterization and functional status. Healthy, postmenopausal women have different risk factors for recurrent UTI than those mentioned above. Recurrent UTIs in healthy postmenopausal women are associated with urinary incontinence, cystocele, and increased post-void residual volumes. Other significant risk factors include at least one episode of UTI prior to menopause, urogenital surgery, and reduced urinary flow. From the Heart and Estrogen/progestin Replacement Study (HERS) of postmenopausal women with coronary heart disease (CHD), additional risk factors included diabetes, vaginal itching, and vaginal dryness. Changes in the vaginal environment after menopause may also predispose to UTI. These alterations include the absence of lactobacilli, elevated vaginal pH, and increased rate of vaginal colonization with Enterobacteriaceae. The intravaginal administration of estrogen has been shown to reduce the rate of recurrent UTI by normalizing the vaginal environment. Low-dose oral hormone therapy, with conjugated estrogen plus medroxyprogesterone acetate (MPA), does not reduce the frequency of UTIs in older women. Urinary Incontinence The prevalence of urinary incontinence in menopausal women is estimated to be in the range of 17% to 56%. Urinary incontinence is the eighth most prevalent chronic medical condition among U.S. women. Anatomic and physiologic alterations associated with aging and incontinence include thinning of the urethral mucosa, reversal of the proteoglycan-to-collagen ratio in the paraurethral connective tissue, decrease in urethral closure pressure, and changes in the normal urethrovesical angle. Many risk factors have been associated with incontinence. Menopause often is considered to be one of these risk factors, especially because a prevalence peak in midlife has been reported by many authors. Epidemiologic studies generally have not found an increase in the prevalence of urinary incontinence in the menopausal transition. A risk factor for incontinence is an increased BMI. This is an especially important factor, because it is modifiable. Vaginal delivery is associated with transient postpartum incontinence, as well as an increased risk of incontinence later in life. Interestingly, a study of nulliparous nuns, with mean age of 68, found that 50% of the nuns had urinary incontinence. Meta-analyses have found an association between hysterectomy and urinary incontinence, with an increase in incontinence of 60%. Considering that more than 600,000 hysterectomies are performed yearly in the United States and that approximately 40% of women have undergone hysterectomy by age 60, these results are quite relevant. Women should be counseled about this relationship prior to undergoing hysterectomy. Other significant risk factors include history of UTI and depression. Treatment Oral ERT has been shown to restore the genitourinary connective tissues to that of premenopausal women, but it seems to have little short-term clinical benefit in regard to urinary incontinence. Interestingly, oral estrogen replacement is associated consistently with an increased risk of incontinence in women aged 60 years and older in epidemiologic studies. This increase may reflect that women with more severe symptoms seek medical care and HRT more often than asymptomatic women. It is also possible that the local levels of estrogen in these studies was too low to benefit fully the urogenital system, given the data suggesting higher systemic dosages may be needed for a vaginal effect. Osteoporosis Osteoporosis is a condition in which bone loss has been sufficient to allow mechanical fracture with limited stress. The likelihood of developing osteoporosis is dependent on the combination of peak bone density (stressing the importance of bone building in the young) and the rate of loss (accelerated with estrogen deficiency). Primary, or “senile,” osteoporosis usually affects women between the ages of 55 and 70 years. The most common sites include the vertebrae and the long bones of the arms and legs. Secondary osteoporosis is caused by a specific disease (such as hyperparathyroidism) or medication usage (glucocorticoids, thyroid hormone excess, anticonvulsants). Menopausal bone loss begins before the FMP during stage -1. Postmenopausal osteoporosis causes over 1.3 million fractures annually. Most of the 250,000-plus hip fractures are due to primary osteoporosis. Excess mortality may exceed 20% within a year of a hip fracture and, because 75% of patients lose their independence, the social costs, not to mention the financial costs, are great. Bone loss following natural menopause is approximately 1% to 2% per year, compared with 3.9% per year following oophorectomy. A woman's genetic background, lifestyle, dietary habits, and coexisting disease will impact the development of osteoporosis. Cigarette smoking, caffeine use, and alcohol consumption are associated with increased bone loss, while weight-bearing activity appears to slow it. Approximately 30% of postmenopausal women have osteoporosis. The World Health Organization has defined osteoporosis as a hip BMD value, as measured by dual x-ray absorptiometry (DEXA), that is greater than 2.5 standard deviations below the adult peak (mean level for young, white women: t-score). Women with existing fractures, regardless of BMD, are also classified as osteoporotic. Both groups are at increased risk for fractures. Those patients with a low z-score (age-matched comparison) should be investigated for secondary causes of osteoporosis. Peak bone mass in women is achieved by the end of the third decade and is an important contributor to bone strength in later life. Adolescence is a critical period of rapid skeletal growth during which almost one half of the adult bone mass is accrued. Many factors contribute to a woman's peak skeletal mass including heredity, diet, physical activity, and endocrine milieu. Hormones which may be a factor in peak bone mass attainment include insulinlike growth factor (IGF)-1, which regulates skeletal growth, and gonadotropins which stimulate sex steroid production and epiphyseal maturation. Estrogen deficiency and amenorrhea can decrease peak bone mass, whereas weight-bearing exercise leads to an increase. Early influences, including birth weight and poor childhood growth, are linked directly to risk of hip fracture. Independent predictors of low peak bone mass include low body weight, menarche at over 15 years of age, and physical inactivity as an adolescent. Early intervention during childhood and adolescence may reduce a woman's risk for osteoporosis in later life—this includes adequate calcium intake and education regarding diet, ideal weight, and physical activity. Treatment Hormone Replacement Therapy. HRT has been used widely for the prevention of osteoporosis and is FDA approved for this indication. It is clear that HRT helps prevent bone loss, as indicated by increased BMD. Whether these benefits translate into decreased fracture risk has been an important topic of research for the past decade. Observational studies have shown lower vertebral and nonvertebral fracture rates in women receiving estrogen compared with those not receiving this therapy. The addition of a progestin does not alter these results. However, observational studies may be biased, because women using HRT have better access to medical care and maintain healthier lifestyles in general. Estrogen therapy acts via an inhibition of bone resorption. Although both BMD and fracture rate are improved with estrogen therapy, there is a rapid and progressive loss of bone mineral content after cessation of estrogen therapy. By 4 years after therapy, bone density is no different from that of patients who were never treated with estrogen. Although estrogen is approved for prevention of osteoporosis, there is some evidence to support its usage in treatment. Dosages of 0.625 mg of conjugated estrogen and, more recently, as low as 0.3 mg have been shown to slow bone loss and provide adequate protection against the development of osteoporosis. Higher dosages may be required to treat existing disease. Meta-analysis of randomized trials shows an overall 27% reduction in nonvertebral fractures in a pooled analysis, with the effect being greater in women under 60 years of age. The HERS trial showed no evidence of reduction in incidence of fractures or rate of height loss in older women with CHD not selected for osteoporosis. The Women's Health Initiative (WHI), the first randomized primary prevention trial studying the effects of postmenopausal HRT, showed a significant reduction in hip fracture (hazard ratio [HR] = 0.66; CI = 0.45–0.98). It seems clear that HRT must be taken indefinitely to preserve bone mass. Calcitonin. Calcitonin is a hormone normally secreted by the thyroid gland and responsible for calcium homeostasis. Calcitonin is now available as a nasal spray, specifically developed to decrease local side effects caused by the earlier subcutaneous injection. Intranasal calcitonin has been shown to improve spinal bone density and decrease the vertebral fracture rate in women with established osteoporosis. The increase in bone density appears to peak in as little as 12 to 18 months. This may be due to down-regulation of the calcitonin receptors and the development of neutralizing antibodies. Although few studies have been performed and no data are available regarding calcitonin-related reductions in hip fracture, calcitonin does seem to be especially beneficial for women with a recent and still painful vertebral fracture. Unfortunately, it appears some patients do not respond to this therapy and those nonresponders cannot be identified prospectively. Bisphosphonates. These nonhormonal compounds are analogs of pyrophosphates which have an affinity for hydroxyapatite in bone. The basic molecular structure of bisphosphonates allows a large number of manipulations, producing different types of bisphosphonates which vary considerably in their potency. The first one used clinically was etidronate, the least potent. In succeeding order of development were pamidronate, alendronate, and risedronate, which is the most potent of these compounds. Etidronate, if given continuously for more than 6 months, impairs

mineralization of bone and may cause osteomalacia. There have been occasional reports of pamidronate also causing impairment of mineralization of bone. However, continuous administration of either alendronate or risedronate has not caused osteomalacia. Alendronate has been evaluated more extensively than calcitonin and has a proven track record, reducing the risk of all major fracture types (vertebral and nonvertebral) in women with osteoporosis. In one study, risk of hip fracture was reduced by 53%, clinical vertebral fracture by 45%, and wrist fracture by 30%. For all fracture types, some reduction in fracture risk was evident within the first year of treatment. Bisphosphonates have only a modest effect on BMD early in the treatment course. Because alendronate is effective so quickly in reducing fracture risk, mechanisms other than increased BMD, such as changes in bone remodeling rates, may play a role in fracture reduction. Alendronate has been shown to inhibit markers of bone remodeling and to increase BMD at the lumbar spine, hip, and in all other bones. Bone density increases with alendronate are greater than with calcitonin and are equal to that with HRT. The escape phenomenon seen with calcitonin is not seen with alendronate, which has the advantage of oral administration. The recommended daily dosage of 10 mg, however, must be taken according to a very strict dosing schedule (in the morning on an empty stomach, with the patient required to remain upright for 30 minutes thereafter). The medication has very poor bioavailability (approximately 1%) and, thus, these restrictions must be followed precisely. Alendronate also has a propensity for irritation of the esophagus and stomach, especially in women with preexisting esophageal reflux or gastric or duodenal disease. A newer formulation allows for once-weekly administration, and development of a once-yearly intravenous formulation may further decrease side effects and administration difficulties. Risedronate has been shown to reduce substantially the risk of both vertebral and nonvertebral fractures, also. Bone density is increased in both early postmenopausal women and those with established osteoporosis. The final remaining question concerning bisphosphonates has to do with the near-permanent changes in bone with the incorporation of this agent into the bone matrix. Although short-term fracture data appear favorable, the long-term effects of these agents and the ability of bone treated with bisphosphonates to heal (e.g., following hip fracture) are not known. Raloxifene. Raloxifene is the first of a new generation of compounds known as selective estrogen receptor modulators (SERMs) to have a treatment indication for osteoporosis. It is likely that there will be an explosion in the development of SERMs in the years to come. They may represent a new alternative for our patients with breast cancer or for long-term use in all patients. These new agents act as selective estrogen receptor agonists on the bone and possibly the heart and antagonists of estrogen action on the breast and uterus. Effects on the brain are not well characterized and likely will vary among compounds (see below). Data on raloxifene suggests good preservation of bone density, albeit less than seen with alendronate or HRT, and fracture data further supports a protective effect. It is believed that the differential effect of estrogen agonists (estrogens) and estrogen antagonists (“anti”-estrogens) is related to the transcriptional activation of specific estrogen-response elements. There appear to be two different domains of the estrogen receptor (AF-1 and AF-2) responsible for this transcriptional activation. Estrogen agonists and estrogen antagonists appear to act via different domains, resulting in their differential effects. Both appear to act to maintain bone density, at least partially, via regulation of the gene for transforming growth factor-ß. It has further been suggested that there may be a raloxifene-response element, which is distinctly different from the estrogen-response element. In women with osteoporosis, the 60-mg or 120-mg daily dose of raloxifene increases bone mass density in the spine and femoral neck and reduces the risk of vertebral fracture by 30% to 50% over no treatment. However, as with estrogen use, women receiving raloxifene have an increased risk of deep vein thrombosis compared with women receiving placebo. Calcium and Vitamin D. Calcium and vitamin D are important components of all three antiresorptive agents (calcitonin, bisphosphonates, SERMs). Decreased ability to absorb calcium among older women is due, in part, to impaired vitamin D activation and effect. In addition, older women may have limited exposure to sunlight, and their dietary vitamin D intake may be lower than that of their younger counterparts. Daily calcium intake of 1,500 mg and 400 to 800 IU of vitamin D per day is, in and of itself, probably sufficient to reduce the risk of fragility fractures by about 10%. Summary-Osteoporosis Most current strategies regarding osteoporosis treatment have focused on identifying postmenopausal women who have low BMD and are already at increased risk for fracture. An evidence-based approach is to recommend appropriate calcium and vitamin D intake, smoking cessation, weight-bearing exercise, moderation in alcohol consumption, and fall prevention. If pharmacologic therapy is indicated, one of the above FDA-approved regimens should be instituted ( Table 41.1 and Table 41.2). An alternative (complementary) approach to prevention is to focus on intervention beginning in childhood and adolescence, with attention to achieving maximal peak bone mass and minimizing premenopausal and postmenopausal bone loss.

TABLE 41.1. Magnitude of Effect on Vertebral Fractures

TABLE 41.2. Magnitude of Effect on Nonvertebral Fractures Cardiovascular Disease CVD is the number one killer of both men and women in Western societies and is attributed primarily to age and lifestyle. Lifestyle modifications are well known to decrease the incidence of CVD. For women, CVD is largely a disease of the after menopause. Women will now spend more than a third of their lives beyond menopause and, thus, preventive measures are paramount. A large body of observational evidence supports a protective effect of ERT on CVD. Observational data, however, are limited by the confounding variables of patient self-selection. Animal and in vitro studies, as well as assessment of surrogate markers in women, have shown a positive effect of estrogen and (less so) HRT against CVD development. Approximately 2.5 million women in the United States are hospitalized each year for cardiovascular illness. CVD claims the lives of 500,000 women annually. One half of these deaths are due to CHD, making this the most frequent cause of death among U.S. women. CVD can be separated into two categories, (a) CHD and (b) noncoronary CVD, such as stroke, valvular heart disease, peripheral vascular disease, congestive heart failure, and sudden death from cardiac causes. A woman's risk of heart disease is far lower than a man's risk until after menopause. This change in incidence of heart disease may be related to advancing age, changes in hormonal milieu, or other unknown factors. CHD entails a worse prognosis for women than for men following either medical or surgical therapies. These sex discrepancies may reflect the older age, smaller body size, more frequent and severe coexisting illnesses of women and, perhaps, a higher incidence of delayed or suboptimal care. Chest pain or tachycardia may be overlooked as benign problems in women, potentially caused by depression, anxiety, or panic disorders. These misperceptions may lead to bias in evaluating women with chest pain. Chest pain in women may also have atypical clinical pictures compared with the more “classic” symptomatologty in men. Chest pain, regardless of symptoms compatible with angina pectoris, warrants evaluation for CHD, regardless of sex. If a woman seeks treatment for typical, or atypical, signs and symptoms of myocardial ischemia, a careful clinical history, including assessment of cardiac risk factors, should be taken. Electrocardiographic exercise testing is recommended for women who give a history typical of angina pectoris if the resting electrocardiogram findings are normal. When the resting electrocardiogram results are abnormal, the patient should be referred for either perfusion imaging studies or coronary arteriography. No screening test is of value for asymptomatic patients, even when risk factors are present. Fewer symptomatic women than men undergo diagnostic coronary arteriography and therapeutic angioplasty or bypass surgery. Women who do undergo bypass surgery more often require emergency surgery and typically are sicker than men at the time of intervention. They have increased operative mortality rates and postoperative complications. It seems that women are referred for revascularization procedures at a later, more symptomatic, stage of illness. Both coronary angioplasty and coronary bypass surgery have a comparable long-term survival for men and women who survive the initial hospital stay. Beta-blockers and aspirin are equally efficacious in both sexes in preventing reinfarction after myocardial infarction. Women should be encouraged to seek medical attention if they have any symptoms suggestive of myocardial ischemia. Physicians should emphasize the importance of modifiable risk factors for CHD, most notably weight management, fat restriction, increased physical activity, treatment of hypertension, and smoking cessation. Randomized, controlled studies have failed to support a protective role for HRT. This is in conflict with earlier observational studies, studies utilizing surrogate markers, and animal studies. The differences may be due to the unbiased sample of subjects taking HRT in these newer studies and, hence, may be a more realistic reflection of the impact of HRT on CVD. However, other potential confounders may be present, even in these studies, such as the menopausal stage when treatment was begun, the particular HRT regimen selected, and the individual risk factors of each subject (although the results may be more reasonably generalized, this may not accurately reflect clinician practice). The first reported trial evaluated HRT for secondary prevention. The HERS trial evaluated daily HRT (0.625 mg conjugated estrogen + 2.5 mg MPA) in 2,763 postmenopausal women with a mean age of 66.7 years and documented preexisting vascular disease. The study failed to demonstrate any overall difference in subsequent vascular events. This occurred despite improvements in lipid parameters in those patients receiving HRT. Although some have questioned the negative impact of the progestin, in this trial, the Estrogen Replacement and Atherosclerosis (ERA) trial published in 2000, compared 3.2 years of treatment with estrogen, combined estrogen and progestin, and placebo. Study participants were postmenopausal women aged 42 to 80 years. This, again, was a secondary prevention trial and also failed to demonstrate a significant difference in the rate of progression of coronary atherosclerosis among the three groups. The importance of this study was the inclusion of an estrogen-only arm. The WHI is the first large, randomized study to look at primary prevention. The same combined HRT regimen used in the HERS trial was evaluated. This study was stopped when interim analyses demonstrated an unacceptable risk profile for a drug in a prevention trial. There was an increase in the incidence in breast cancer (an increase of 8 cases per 10,000 women), with no cardiovascular protection (and potentially increased cardiovascular risk). There was an increase in the absolute number of blood clots, strokes, and CHD. The risk of stroke and clot continued for the 5 years of study, while most of the CHD was limited to the first year of treatment. Whether these finding were related to the intrinsic properties of HRT, the oral administration route, or the unique regimen tested is not clear. There were, however, documented decreases in the risk of fracture and colon cancer. Management for cardiovascular risk for women should parallel that for men. In other words, lifestyle modifications should be made and antihypertensive therapy given as needed. If hyperlipidemia persists, statin therapy should be instituted. And, as above, ß-blockers and aspirin should be given to prevent recurrent myocardial infarction. There is no evidence from well-designed prospective trials supporting a role for either ERT or HRT for the primary indication of cardiovascular protection.

HORMONE REPLACEMENT

Introduction Approximately 38% of U.S. women aged 50 to 74 use HRT of some description. Of those surveyed, 59% of users had undergone a hysterectomy and 19.6% had a uterus. In 2000, Premarin (conjugated equine estrogen [CEE]) was the second most widely prescribed drug in the United States, accounting for 46 million prescriptions and over $1 billion in sales. Historically, patients desiring hormone treatment for menopausal symptoms were supplemented initially with unopposed estrogen, regardless of the presence of a uterus. This therapy was proven clearly to increase a woman's risk of endometrial cancer, which was eliminated with the addition of either cyclic or continuous low-dose progestin. Women who have undergone hysterectomy can, and should, be treated with estrogen alone. According to the FDA, approved indications for ERT-HRT include treatment of menopausal symptoms (e.g., hot flashes and genital tract atrophy) and the prevention of osteoporosis. During the past 30 years, it has become popular to prescribe ERT-HRT to prevent a range of chronic diseases, most notably heart disease. As above, evidence has been accumulating which suggests that estrogen-progestin therapy for prevention of chronic diseases is not evidence based. Benefits With the publication of the WHI data, HRT risks and benefits have been examined critically ( Fig. 41.4). HRT benefits include reduction in hot flash frequency and severity, improvement of atrophic vaginitis and UTIs, and prevention of osteoporosis and fractures. Neither vasomotor symptoms nor vaginal atrophy were evaluated in the WHI, although likely occurrence was low, because the participants were asymptomatic and not taking HRT when enrolled. The WHI results did support an additional benefit of decreased risk of colorectal cancer.

FIG. 41.4. Estimated hazard ratios for major clinical outcomes in the Women's Health Initiative trial of hormone replacement therapy. CHD, coronary heart disease; PE, pulmonary embolism. (Source: Writing Group for the Women's Health Initiative investigators. From Grimes DA, Lobo RA. Perspectives on the Women's Health Initiative trial of hormone replacement therapy. Obstet Gynecol 2002;100:1344–1353, with permission.)

A variety of areas have been studied in postmenopausal women, including estrogen effects on depression and Alzheimer disease (AD). The cognitive benefits of HRT remain more controversial than benefits in other areas. A meta-analysis from 1998 incorporating all studies published between 1966 and 1997 suggests several mechanisms by which estrogen may affect cognition, including maintenance of neural circuits, favorable lipoprotein changes, prevention of cerebral ischemia, and promotion of serotonergic and cholinergic activity in the brain. It is also possible that cognition improves in estrogen users during the menopausal transition because vasomotor symptoms improve, because there does not appear to be a clear benefit in asymptomatic women receiving HRT. The HERS data also support a cognitive benefit, concluding that the effects of HRT on emotional measures, of quality of life, depend on the presence of menopausal symptoms. The Seattle Midlife Women's Health Study showed that perceived memory functioning is more closely related to depressed mood, perceived stress, and perceived poor health than to age or reproductive stage. AD is an enormous public health concern that intensifies as the population ages. In 1997, AD cases in the United States numbered 2.32 million, with 68% of individuals being female. In 1998, the annual number of new cases was 360,000. Interventions that could prevent this disease, or delay its onset, would have a major public health impact. Women with high serum concentration of unbound (bioavailable) estradiol are less likely to develop cognitive impairment than women with low concentrations. Randomized trials studying estrogen use and AD are limited but suggest that estrogen users have a reduced risk of developing AD. However, there is no therapeutic benefit of estrogen usage in women with preexisting AD. Larger randomized trials are needed to evaluate the true significance of estrogen use and dementia. The WHI includes a memory study, and these results should be available in 2005. Risks As more randomized trials of HRT are conducted, the risk profile has broadened. CHD was once a main off-label indication for HRT, and former guidelines from the American College of Physicians even suggested that HRT be considered for all women. The HERS trial and, more importantly, the WHI have shown that nonfatal cardiac events are increased in HRT users in the regimen tested. These studies have also reemphasized known risks of HRT including pulmonary embolus, stroke, deep vein thrombosis, and gallbladder disease.

CONTROVERSIES Cardiovascular Disease During the 1990s, HRT was prescribed increasingly to postmenopausal women for CHD prevention. Much of this enthusiasm was based on a meta-analysis published in 1992 that concluded that “there is extensive and consistent observational evidence that estrogen use reduces risks for CHD about 35%.” This same year, the American College of Physicians published guidelines recommending that all postmenopausal women should consider HRT. They emphasized that those who are at increased risk of CHD are especially likely to benefit. Further observational studies have supported this recommendation. For example, the Nurse's Health Study of 1996 confirmed a 40% to 60% reduction in cardiovascular events in women taking HRT. The HERS was a randomized, blinded, secondary prevention, clinical trial with results published in 1998. The study population was 2,763 postmenopausal women with documented CHD. The average age was 67 and all women had a uterus. Subjects were randomly assigned to receive either 0.625 mg CEE with 2.5 mg MPA daily or placebo. The average follow-up was 4.1 years. The investigators found that the women assigned to receive CEE plus MPA had a 50% increased risk of coronary events during the first year of the trial compared with the placebo group. The risk was greatest during the first 4 months. The risk returned to baseline over the next 2 years and seemed to be lower in the hormone-treated group beginning in the third year of the study. The changes could have occurred by chance or may be due to detrimental effects of the hormone replacement, the specific regimen studied, or the procoagulant effect associated with oral administration. An initial prothrombogenic effect was thought to be the cause of early morbidity, and there was some expectation that longer treatment would document an improved outcome. However, HERS II, which continued to follow this group (6.8 years of follow-up), failed to see a developing protective effect of HRT. The WHI is the first randomized, primary prevention trial studying the effects of a specific postmenopausal HRT regimen (the daily regimen of 0.625 mg CEE plus 2.5 mg MPA). In contrast to the HERS trial, these 50- to 79-year-old women were considered healthy, with only a small proportion (7.7%) of subjects having clinical signs or symptoms of CHD. The subjects were randomized to receive the HRT regimen or placebo if they had a uterus. Women without a uterus received either the estrogen alone or placebo. The planned duration of the trial was 8.5 years, but the estrogen-plus-progestin arm (16,608 women) was stopped after 5.2 years because of concerns regarding cardiovascular events and breast cancer. “The test statistic for invasive breast cancer exceeded the stopping boundary for this adverse effect and the global index statistic supported risks exceeding benefits.” Estimated HRs for diseases other than breast cancer, including CHD, were also significant. The HR results were as follows: CHD, 1.29 (95% CI = 1.02–1.63); breast cancer, 1.26 (1.00–1.59); stroke, 1.41 (1.07–1.85); pulmonary embolus (PE), 2.13 (1.39–3.25); endometrial cancer, 0.83 (0.47–1.47); colorectal cancer, 0.63 (0.43–0.92); hip fractures, 0.66 (0.45–0.98). This important study suggests that breast cancer, nonfatal CHD events, stroke, and PE are all significantly increased in the overall cohort, of women 50 to 79 years old using combined estrogen and progestin therapy. It showed that colorectal cancer and hip fracture risk are reduced significantly in the same group. Ultimately, the data suggest that the use of this combined HRT regimen for primary prevention of CHD is not justified. In 2001, the American Heart Association published guidelines regarding HRT and CVD. They have taken a cautious stance. These recommendations were made prior

to the publication of the WHI results. American Heart Association Recommendations Secondary Prevention HRT should not be initiated for the secondary prevention of CVD. The decision to continue or stop HRT in women with CVD who have been undergoing long-term HRT should be based on established noncoronary benefits and risks and patient preference. If a woman develops an acute CVD event or is immobilized while undergoing HRT, it is prudent to consider thromboembolic prophylaxis while she is hospitalized to minimize risk of a venous thromoembolism associated with immobilization. Reinstitution of HRT should be based on established noncoronary benefits and risks, as well as patient preference. Primary Prevention Firm clinical recommendations for primary prevention await the results of ongoing randomized clinical trials. There are insufficient data to suggest that HRT should be initiated for the sole purpose of primary prevention of CVD. Initiation and continuation of HRT should be based on established noncoronary benefits and risks, possible coronary benefits and risks, and patient preference. There is still some question about whether WHI was really assessing “primary” prevention based on concern about the preexistence of atherosclerotic disease. Enrollees were asymptomatic and older (mean age, 63 years) than the typical HRT-ERT user. Some vascular biologists have suggested the early institution of HRT may inhibit the development of atherosclerotic plaque and that a delay in treatment for several years after menopause, after plaque formation is well established, would not only not offer benefit but might increase risk by destabilizing existing plaque, leading to plaque rupture and thrombosis. Although this thesis is consistent with animal studies, little human data are available to support or repute it. It is yet another consideration for patients and physicians in making this difficult decision. Breast Cancer A woman's lifetime risk of breast cancer is 1 in 8. There are 192,000 new cases of breast cancer per year, accounting for 41,000 deaths per year. It is the second leading cause of cancer death after lung cancer and the leading cause of death among women ages 40 to 55. The most compelling reason to believe that long-term use of postmenopausal estrogen increases the risk of breast cancer is the inherent biologic plausibility. Many of the risk factors associated with breast cancer are thought to be linked to increased duration of exposure to estrogen over a woman's lifetime. These risk factors include early menarche, late menopause, nulliparity, and older age at the birth of her first child. Oophorectomy can induce breast tumor regression and early oophorectomy is protective against breast cancer, which seem to support further the notion that estrogen is involved, but this remains controversial. All of these risk factors take ovarian function into account, but not necessarily estrogen exposure, because the ovary is responsible for the formation of many other compounds which may influence the risk, such as androgens. The association of estrogen therapy and breast cancer remains controversial, despite the publication of over 50 epidemiologic studies during the past 25 years. This subject is of great public health concern because of women's understandable fear of breast cancer and the complexity of the decision-making process regarding hormone therapy. A reanalysis of the world's data on HRT-ERT and breast cancer was performed in 1997. A team of epidemiologists invited all investigators who had previously studied the association of postmenopausal hormone use and the risk of breast cancer (51 studies) to submit their original data for a collaborative combined reanalysis, an undertaking more rigorous than a standard meta-analysis (Collaborative Group of Hormonal Factors in Breast Cancer). This analysis reached the following conclusions: Ever users of postmenopausal hormones had an overall increased relative risk of breast cancer of 1.14. Current users for 5 or more years had a relative risk of 1.35 (CI = 1.21–1.49), and the risk increased with increasing duration of use. Current and recent users had evidence of having only localized disease (no metastatic disease) and ever users had less metastatic disease. There was no effect of a family history of breast cancer. There was no increase in relative risk in past users. The increase in relative risk in current and recent users was greatest in women with lower body weights. The WHI has brought even greater attention to any relationship between breast cancer and HRT use. The estrogen-progestin versus placebo arm of the study was stopped early, due primarily to an increase in breast cancer risk with a HR of 1.2. No similar effect was noted in the estrogen-only arm of the study, which is ongoing. Whether or not women in this category will also have an increased risk of breast cancer is unknown at this time. There is other epidemiologic evidence to suggest combined, continuous HRT may confer a higher risk of breast cancer than either cyclic estrogen-progestin regimens or those with estrogen only. The comparison between the collaborative epidemiologic analysis described above and the WHI is interesting (relative risk, 1.35 vs. 1.26). Another relevant point the authors from the collaborative reanalysis make is that the quantitative effect of their conclusion is similar to the impact of raising the age of menopause. According to their calculations, current and recent hormone use was associated with a 2.3% increase in breast cancer risk per year, and the effect of age of menopause was equivalent to a 2.8% increase in risk per year of delay. The presence of only localized disease, in the collaborative study, raises concern regarding screening bias and whether ERT-HRT treatment simply accelerated the detection of tumors already present. All of these studies that have examined the mortality rates of women who were taking estrogen at the time of breast cancer diagnosis have documented improved survival rates. This reflects earlier diagnosis in users, because the greater survival rate in current users is associated with a lower frequency of late-stage disease. There has been a suggestion that estrogen users develop better differentiated tumors and the surveillance-detection bias is not the only explanation for better survival. This suggests that hormone treatment accelerated the growth of a malignant locus already in place, which appears clinically at a less virulent and aggressive stage. Selective Estrogen Receptor Modulators The controversy regarding HRT-ERT has heightened the call for alternatives. The term hormone-estrogen replacement implies the replacement of the beneficial aspects of premenopausal estrogen. SERMs make up a class of compounds that act like estrogen agonists on some tissues and estrogen antagonists on others. SERMs act by causing dimerization and conformational changes in the estrogen receptor, altering the interaction with the promoter regions on DNA. These conformational changes further impact function through their interactions with co-regulators and co-repressors. This is illustrated in cartoon form in Figure 41.5.

FIG. 41.5. Differential selective estrogen receptor modulator activity (stimulation and inhibition) is determined by the specific tissue, the endogenous milieu, and the impact of conformational changes on the interaction with regulatory sequences of DNA and co-regulatory proteins (co-activators and co-repressors). (From Katzenellenbogen BS, Katzenellenbogen JA. Defining the “S” in SERMs. Science 2002;295:2380–2381, with permission.)

The search for the “ideal” SERM, with positive (estrogenic) effects on the heart, bone, and brain and negative (antiestrogenic) effects on the breast and uterus, continues. Improved scientific knowledge regarding estrogen receptor function (more like a rheostat than an “on-off” switch), two receptor types (a and ß), more than 50 transcription factors, over 20 modifying proteins, and different response elements should enhance the development of therapeutic agents for selective clinical applications. The beneficial aspects of raloxifene, the most studied SERM, with respect to osteoporosis, already have been discussed. There are evolving data with respect to CVD. Both raloxifene and tamoxifen decrease total cholesterol and low-density lipoprotein levels. Neither has an effect on high-density lipoprotein or triglyceride levels. Interestingly, raloxifene is either neutral or decreases two inflammatory markers (homocysteine and C-reactive protein), which might participate in the increase in cardiovascular events seen with the institution of HRT-ERT. The available data about SERMs and CVD are, again, based on surrogate markers. A prospective study, Raloxifene Use in The Heart (RUTH), is ongoing. This study will evaluate 10,000 at-risk women and will add information to guide treatment strategies and, hopefully, increase our treatment options. Limited data have suggested SERMs may have a beneficial effect on cognition. Although raloxifene is known not to improve (and potentially worsen) vasomotor symptoms, there is some preliminary evidence to suggest a neuroprotective effect. The extent and mechanism of this effect is not well characterized and awaits further study. There is no doubt that the available SERMs are antiestrogenic at the breast. Tamoxifen is used primarily in the United States for cancer chemoprevention and treatment. Raloxifene also appears to decrease the incidence of estrogen receptor-positive cancer development. The Study of Tamoxifen and Raloxifene (STAR) should yield direct information regarding the benefit of raloxifene in women at risk for breast cancer as the available data are derived from osteoporosis trials. Lastly, it should be remembered that both tamoxifen and raloxifene increase the risk of thromboembolic phenomena, and this may have an impact on cardiovascular risk.

SUMMARY The WHI did not address the effect of hormone treatment on hot flashes and vaginal atrophy. Clearly, there are alternatives for the treatment of osteoporosis and CVD that have less risk if CVD prevention is the sole reason for using HRT. Each woman should discuss with her health care provider the optimal treatment management to address her individual goals. This obviously would need to take into account medical and family history, as well as symptomatology. If HRT-ERT is selected by the patient, progestin should be used only for those women with a uterus. Furthermore, in developing an overall treatment strategy, physicians should utilize the full armamentarium of medications (conjugated estrogens, estradiol, MPA, norethindrone, micronized progesterone, and androgens), routes of administration (oral, vaginal, transdermal), and cyclic patterns (continuous, intermittent, monthly, quarterly) available, including low-dose aspirin and statins. Duration of hormone therapy would also be an individual decision based on the original indication for treatment and its persistence. It can be recommended uniformly that menopausal women maintain appropriate nutrition, a healthy body weight, and regular exercise, including both weight-bearing aerobic exercise and muscle-strengthening exercise. This should be associated with moderation in alcohol intake and cessation of smoking.

SUMMARY POINTS Estrogen production in women is related to follicular maturation and the number of ovarian follicles is fixed, with declining numbers of follicles associated with hypoestrogenism. Declining ovarian estrogen production is associated with vasomotor instability, hot flashes, urogenital tract atrophy, and accelerated loss of BMD. Treatment of the symptoms associated with declining ovarian estrogen production is optimal with ERT in women without a uterus and estrogen with progestin regimens in women with a uterus. Although primary or secondary prevention of CVD has been suggested in observational studies, it has not been demonstrated in randomized treatment trials (HERS and WHI). A slight increase in the risk of breast cancer has been associated with HRT, albeit with a lower mortality likely secondary to earlier detection at a less aggressive stage of disease. Decisions regarding the use or ERT-HRT should be individualized, based on the woman's goals of therapy, and reevaluated periodically to ensure that the risk-benefit analysis continues to favor hormone use. SUGGESTED READINGS General Chetkowski RJ, Meldrum DR, Steingold KA, et al. Biologic effects of transdermal estradiol. N Engl J Med 1986;314:1615–1620. Faddy MJ, Gosden RG, Gougeon A, et al. Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 1992;7:1342–1346. Soules MR, Sherman S, Parrott E, et al. Stages of Reproductive Aging Workshop (STRAW). J Womens Health Gend Based Med 2001;10:843–848.

Premature Ovarian Failure Cramer DW, Xu H, Harlow BL. Does “incessant” ovulation increase risk for early menopause? Am J Obstet Gynecol 1995;172:568–573. Mehta AE, Matwijiw I, Lyons EA, et al. Noninvasive diagnosis of resistant ovary syndrome by ultrasonography. Fertil Steril 1992;57:56–61. Tibilietto MG, Testa G, Vegetti W, et al. The idiopathic forms of premature menopause and early menopause show the same genetic pattern. Hum Reprod 1999;14:2731–2734. Van Kasteren YM, Hundscheid RD, Smits AP, et al. Familial idiopathic premature ovarian failure: an overrated and underestimated genetic disease. Hum Reprod 1999;14:2455–2459.

Vasomotor symptoms Berendsen HH. The role of serotonin in hot flushes. Maturitas 2000;36:155–164. Shanafelt RD, Barton DL, Adjei AA, et al. Pathophysiology and treatment of hot flashes. Mayo Clin Proc 2002;77:1207–1218. Weitzner MA, Moncello J, Jacobsen PB, et al. A pilot trial of paroxetine for the treatment of hot flashes and associated symptoms in women with breast cancer. J Pain Symptom Manage 2002;23:337–345.

Genitourinary Atrophy Brown JS, Sawaya G, Thorn DH, et al. Hysterectomy and urinary incontinence: a systematic review. Lancet 2000;356:535–539. Keil K. Urogenital atrophy: diagnosis, sequelae, and management. Curr Womens Health Rep 2002;2:305–311. Sherburn M, Guthrie JR, Dudley EC, et al. Is incontinence associated with menopause? Obstet Gynecol 2001;98:628–633. Willhite LA, O'Connell MB. Urogenital atrophy: prevention and treatment. Pharmacotherapy 2001;21:464–480.

Osteoporosis Multiple authors. Meta-analyses of therapies for postmenopausal osteoporosis. Endocrine Rev 2002;23:495–578.

Cardiovascular Disease

Barrett-Conner E, Grady D, Sashegyi A. Raloxifene and cardiovascular events in osteoporotic postmenopausal women. JAMA 2002;287:847–857. Grady D, Herrington D, Bittner V, et al. Cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/progestin Replacement Study follow-up (HERS II). JAMA 2002;288:49–57. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N Engl J Med 2000;343:522–529. Hulley S, Grady D, Bush T. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605–613. Mikkola TS, Clarkson TB. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovasc Res 2002;53:605–619. Postmenopausal hormone replacement therapy for primary prevention of chronic conditions: recommendations and rationale. Ann Intern Med 2002;137:834–839. Teede HJ. Hormone replacement therapy and the prevention of cardiovascular disease. Hum Reprod Update 2002;8:201–215.

Breast Cancer Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet 1997;350:1047–1059. Cui Y, Whiteman MK, Flaws JA, et al. Body mass and stage of breast cancer at diagnosis. Int J Cancer 2002;98:279–283. Cummings SR, Duong T, Kenyon E, et al. Serum estradiol level and risk of breast cancer during treatment with raloxifene. JAMA 2002;287:216–220. Cummings SR, Eckert S, Krueger KA, et al. The effect of raloxifene on risk of breast cancer in postmenopausal women: results from the MORE randomized trial. JAMA 1999;281:2189–2197. Ursin G, Tseng CC, Paganini-Hill A, et al. Does menopausal hormone replacement therapy interact with known factors to increase risk of breast cancer? J Clin Oncol 2002;20:699–706.

Cognitive Function Fillit JM. The role of hormone replacement therapy in prevention of Alzheimer disease. Arch Intern Med 2002;162:1934–1942. Yaffe K, Lui LY, Grady D, et al. Estrogen receptor 1 polymorphisms and risk of cognitive impairment in older women. Biol Psychiatry 2002;51:677–682. Yaffe K, Sawaya G, Lieberburg I, et al. Estrogen therapy in postmenopausal women: effects on cognitive function and dementia. JAMA 1998;279:688–695.

Selective Estrogen Receptor Modulators American College of Obstetricians and Gynecologists. Selective estrogen receptor modulators. ACOG Practice Bulletin No. 29, Oct 2002. Katzenellenbogen BS, Katzenellenbogen JA. Defining the “S” in SERMs. Science 2002;295:2380–2381. Morello KC, Wurz GT, DeGregorio MW. SERMs: current status and future trends. Crit Rev Oncol Hematol 2002;43:63–76.

Controversies Grimes DA, Lobo RA. Perspectives on the Women's Health Initiative trial of hormone replacement therapy. Obstet Gynecol 2002;100:1344–1353. McDonough PG. The randomized world is not without its imperfections: reflections on the Women's Health Initiative study. Fertil Steril 2002;78:951–956. Writing group for the Women's Health Initiative investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAMA 2002;288:321–333.

Treatment Regimens Samsioe G. The role of ERT/HRT. Best Pract Res Clin Obstet Gynaecol 2002;16:371–381.

Chapter 42 Human Sexuality and Female Sexual Dysfunction Danforth’s Obstetrics and Gynecology

Chapter 42 Laurie Jane McKenzie and Sandra Ann Carson

Human Sexuality and Female Sexual Dysfunction

ANATOMY Erectile Tissues Role of the Cervix/Uterus Graefenberg Spot SEXUAL RESPONSE CYCLE The Masters and Johnson Hypothesis The Kaplan Hypothesis The Basson Hypothesis NORMAL SEXUAL PHYSIOLOGY SEXUAL DYSFUNCTION Risk Factors Evaluation Classification ETIOLOGY OF SEXUAL DYSFUNCTION TREATMENT Desire Disorders Arousal Disorders Orgasmic Disorders Pain Disorders SUGGESTED READINGS Anatomy Normal Female Sexual Response Cycle Physiology Sexual Dysfunction Evaluation Classification Etiology Treatment Appendix A—Female Sexual Function Index (FSFI)a Footnote

“In view of the pervicacious gonadal urge in human beings, it is not a little curious that science develops its sole timidity about the pivotal point of the physiology of sex.” W. H. Masters and V.E. Johnson, Human Sexual Response, 1966 Sexuality and satisfying sexual functioning are important aspects of a woman's general health and feelings of well-being. However, the reluctance to engage in open discussions about sex extends even to physicians. Surveys of primary care physicians reveal that less than 50% ask new patients about sexual practices and concerns. Nevertheless, a thorough understanding of the normal sexual response and evaluation and treatment of common sexual dysfunctions is imperative to provide comprehensive gynecologic care.

ANATOMY Female pelvic anatomy demonstrates pronounced changes during sexual arousal ( Fig. 42.1). During sexual arousal there is a pronounced alteration of these anatomic structures. The vaginal walls assume a darker coloration due to vascular engorgement, the inner two-thirds of the vagina lengthen and distend, the labia minora increase in size, and the clitoris swells. Pelvic vasocongestion, swelling of external genitalia, and lubrication occurs. Contrary to findings by Masters and Johnson, magnetic resonance imaging (MRI) studies during either coitus or masturbation, revealed that the size of the uterus does not increase in size, but does elevate in the pelvis. The vagina lengthens by 1 cm, and after further arousal, the vaginal opening constricts secondary to vascular engorgement, called the “orgasmic platform”. The “orgasmic platform” undergoes a series of clonic contractions with the first three to six being strong and having an intercontractile interval of 0.8 seconds with subsequent periods lengthening before the strength of the contractions recede. However, the actual presence of a physical “orgasmic platform” has been questioned.

FIG. 42.1. Female external anatomy.

These anatomic changes noted in response to sexual arousal involve four functional groups of muscles. The first group consists of the ischiocavernous and bulbospongiosus. These muscles insert anteriorly on the shaft and crura of the clitoris which pulls the clitoris down to compress venous drainage and facilitate engorgement. The second group is the transverse perinea (superficial and transverse), which supports the perineum. The levator ani (the medial pubococcygeus and the lateral iliococcygeus) comprises the third set of musculature and forms the anterior pelvic diaphragm. The levator ani fixes the vesicular neck, anorectal junction, and vaginal fornices to the side wall of the pelvis. During sexual intercourse, vaginal distention from the penis evokes the vaginolevator and vaginal puborectalis reflex causing contraction of the levator ani muscles. This contraction leads to upper vaginal ballooning which acts as a receptacle for semen, elevates the uterus, and narrows and elongates the vagina. The development of the pubococcygeus muscle has been advocated to increase sexual pleasure and orgasm during coitus. Kegel proposed that underdevelopment of the pubococcygeus resulted in less pleasurable vaginal sensation during coitus and “every woman with a sexual complaint should be evaluated for possible pubococcygeal dysfunction.” The fourth set of muscles is the smooth muscle fibers of the upper two thirds of the vagina and the striated lower third. These four muscle groups vary considerably in size, strength, and volitional control among women. Although these muscle groups are well-defined anatomically, their exact impact on sexual functioning remains poorly understood. Erectile Tissues Erectile tissues, or cavernous bodies, are present in the clitoris and the vestibular bulbs. These areas become engorged with blood when successful sexual arousal occurs. The clitoris is comprised of three parts: the outermost glans, the midline corpus, and the innermost crura. The paired crura are called corpora cavernosa and are separated by a connective tissue septum. The clitoris is suspended to the anterior abdominal wall by a suspensory ligament, and is exquisitely sensitive to touch, temperature, and pressure. The autonomic innervation of the clitoris arises from the pelvic and hypogastric plexuses, employing both sympathetic and parasympathetic fibers. The clitoris has a dense concentration of pacinian corpuscles, as well as Merckles tactile discs, free nerve endings, and Meissner corpuscles. Sensory innervation to the clitoris travels from the skin via the dorsal nerve of the clitoris to the pudendal nerve, and subsequently to the spinal cord ( Fig. 42.2.) Other erectile tissues are the paired vestibular bulbs. Located laterally to the vaginal orifice and directly beneath the skin of the labia, the vestibular bulbs are homologous to the corpus spongiosum of the penis.

FIG. 42.2. Pelvic innervation.

Role of the Cervix/Uterus The role of the cervix in sexual functioning has been widely debated. The cervix has a rich vascular and nerve supply, and it appears plausible that loss of the cervix may have an adverse effect on sexual arousal. In the widely reported Finnish cohort study from 1979, total abdominal hysterectomy was performed on 105 women and supracervical hysterectomy on 107. The only statistically significant finding was a decrease in orgasm frequency in the total hysterectomy group. The proportion of women that could not achieve orgasm in more than 75% of their coital encounters increased from 30% prior to the total abdominal hysterectomy to 47% at 1 year following the operation. Both groups had a statistically significant decrease in dyspareunia and no change in coital frequency. It is difficult to draw a definitive conclusion from this study as the treatment groups were not blinded and there were demographic differences between the two study groups. A subsequent study comparing 532 women after total abdominal hysterectomy with 146 women after supracervical hysterectomy found no differences. Attempting to define the role of the cervix and uterus in sexual functioning is a challenge, and the existing studies are methodologically flawed. A randomized, controlled trial will be necessary to further elucidate this issue. Graefenberg Spot The existence of a G-spot or Graefenberg spot refers to an allegedly highly erogenous region on the anterior vaginal wall, approximately one-third of the distance from the introitus. The concept of a G-spot is widely accepted among the public, however there is little objective evidence to support its existence. The anterior vaginal wall has more innervation than the posterior wall, however this innervation is subepithelial; intraepithelial innervation would be expected if a sensitive region existed. Thus, sketchy clinical findings and no anatomic evidence exist to document the existence of a G-spot. Nonetheless, the reader is reminded that each woman is an individual and sexually unique and universal anatomic axioms should not dictate clinical care.

SEXUAL RESPONSE CYCLE The Masters and Johnson Hypothesis The classic approach to the female sexual response was proposed by William Masters and Virginia Johnson in 1966 when they arbitrarily divided the sexual response into four stages. These stages of excitement, plateau, orgasm, and resolution were presented as a continuum, with each stage progressing into the next. The excitement phase was defined as sexual arousal or tension resulting from psychogenic or somatic stimuli. Pelvic vasocongestion and increased muscle tension is evident. Vaginal lubrication occurs, the vaginal walls assume a darker coloration due to engorgement, and the uterus elevates slightly. The inner two thirds of the vagina lengthen and distend, the labia minora increase in size, and the clitoris swells. Masters and Johnson noted a two- to three-fold increase in the labia minora diameter. Often the nipples become erect and a sex flush may occur. During the plateau phase, the vaginal opening constricts due to vaginal engorgement. The clitoris often retracts, and if stimuli are adequate, orgasm occurs. The orgasmic phase is described as the few seconds at climax in which sexual tension is relieved in a wave of intense pleasure, often accompanied by myotonia. This is an involuntary reflex that demonstrates uterine and vaginal contractions, and often varies in duration and intensity. Rhythmic pelvic thrusting may occur. Following the release of sexual tension, there is often a feeling of well-being and satisfaction. This resolution stage follows orgasm and the body returns to its baseline state. Relaxation occurs, the sex flush disappears, and vasocongestion reverses. The clitoris and vagina resume their original size and shape. The Kaplan Hypothesis In 1974, Helen Kaplan became unsatisfied with the four stages of sexual response described by Masters and Johnson as she concluded that the sexual responses of patients did not appear to consist of an orderly sequence of separate events but consisted of two fundamentally separate components. The two distinct and relatively independent components are a genital vasoconstrictive reaction causing vaginal lubrication followed by the reflexive, clonic striated and nonstriated muscular contractions at orgasm. Kaplan also proposed a three-phase integrated model to include desire, arousal, and orgasm. She describes desire as an urge to seek out, initiate, or respond to sexual stimulation. Kaplan defines sexual desire as “a motivational or drive state that is generated by specific neurological processes in the brain—similar to other drives or appetites that subserve individual and species survival, not simply a subjective sensation or merely a mental event.” Kaplan's definitions of arousal and orgasm are similar to those proposed by Masters and Johnson; however, Kaplan focuses on the integration of these stages as opposed to occurring in a strict chronologic sequence. The Basson Hypothesis More recently, Basson has outlined an alternative sexual response cycle that emphasizes the role of sexual interaction with a partner in enhancing emotional closeness. The cycle begins with a state of “sexual neutrality” and if the woman seeks or is receptive to sexual stimuli, she will experience arousal and subsequent desire. Once desire is accessed, receptivity to sexual stimuli increases, which leads to further and more intense arousal. Emotional and physical satisfaction can then be attained, which facilitates emotional intimacy. The mental feedback of emotions, physical and genital arousal, mutual excitement, and cognition provide stimuli that modulate the ongoing sexual response. The focal point of this paradigm is “enhanced emotional intimacy,” and orgasm is not imperative for the experience to be satisfying.

NORMAL SEXUAL PHYSIOLOGY Objective evidence of sexual arousal begins with increased clitoral length and diameter and vasocongestion of the vagina, vulva, clitoris, and uterus. The corpora cavernosa of the clitoris consists of a fibroelastic network and bundles of trabecular smooth muscle. Pelvic nerve stimulation results in clitoral smooth muscle relaxation and arterial smooth muscle dilation. With arousal there is an increase in the clitoral cavernosal artery inflow and an increase in the clitoral intercavernous pressure that leads to tumescence and extrusion of the clitoris. Engorgement of the genital vascular network increases pressure inside the vaginal capillaries and results in lubrication of the epithelial surface of the vaginal wall by a transudate of serum. The vaginal fluid demonstrates significant changes in its ionic content before and after sexual arousal. Sodium and chloride levels increase significantly. Although most lubrication is from the transudate, which is passively transported through intraepithelial spaces, and appears on the surface of the vagina, a small amount of additional lubrication arises from the mucus secretions of the Bartholin glands. These objective physiologic changes that occur with arousal are initiated by the medial preoptic, anterior hypothalamic region and related limbic-hippocampal structures within the central nervous system (CNS). Once activated, these centers transmit signals via the parasympathetic and sympathetic nervous systems to the clitoral and vaginal tissues. Serum concentrations of epinephrine and the norepinephrine metabolite, vanillylmandelic acid, increase prior to intercourse and remain elevated over baseline for up to 23 hours following sexual activity. Peripheral adrenergic stimuli may selectively “prepare” the body for genital responsiveness. Interestingly, anorgasmic women demonstrate an inhibition in physiologic sexual arousal under conditions of CNS activation. This may indicate that CNS activation may have a different influence on women with and without anorgasmia, and may have important implications for cardiac patients who take antihypertensive medications with high affinities for a-adrenergic receptors.

Another neuropeptide that impacts sexual function is serotonin ( Table 42.1). Activation of the serotonin-2 receptor appears to impair sexual functioning and stimulation of the serotonin-1a receptor facilitates sexual functioning. Nefazodone (Serzone) is a selective serotonin reuptake inhibitor (SSRI) as well as a serotonin-2 receptor antagonist, and reportedly causes fewer sexual side effects than traditional SSRIs. The mechanism of action may be that nefazadone produces both a reduction in the number and activity of serotonin-2 receptors as well as an up-regulation in serotonin-1a receptors. The sexual side effects of SSRIs however, may be through a peripheral effect. Ninety-five percent of serotonin receptors are located in the periphery of the body, and peripheral serotonin acts on vascular smooth muscle to produce vasodilation and vasocongestion, acts on the smooth muscles of the genitals, and is active in peripheral nerve function.

TABLE 42.1. Selective serotonin reuptake inhibitors a and incidence of sexual dysfunction

Dopamine is probably the most frequently studied of the neurochemicals that collectively regulate sexual activity, and while serotonin is generally inhibitory to sexual behavior, dopamine is regarded as facilitative. Dopamine generally enhances sensorimotor integration by removing tonic inhibition. Although steroid hormones increase the responsiveness of certain hormones, those neurons cannot fully respond to stimuli unless the tonic inhibition is removed. Therefore, dopamine does not directly affect sexual behavior; rather it allows hormonally primed output pathways to have access to sexually receptive stimuli. Dopamine agonists have been shown to increase mounting behavior and increase sexual behavior in sexually satiated male rats. Antiparkinsonian medications act as dopamine agonists and have been reported to increase sexual desire in human men. Bromocriptine, which decreases prolactin levels and is also a long-acting dopamine agonist, facilitates erectile functioning. Delayed or inhibited orgasm in women has been associated with antipsychotic medications that decrease dopamine activity. Cocaine enhances dopamine activity by blocking the presynaptic autoreceptor and, in low doses, is believed to enhance sexual pleasure. At higher doses, cocaine impairs sexual functioning and often leads to anorgasmia, perhaps a result of the vasoconstrictive effects of the drug. Women with elevated prolactin levels frequently report a decrease in sexual interest that is often reversed with bromocriptine treatment. A significant two-fold increase in prolactin levels occurs in women following orgasm and prolactin remained elevated when measured 60 minutes after sexual arousal. Circulating oxytocin increases during sexual arousal and orgasm. In female animals, oxytocin injected either centrally or peripherally facilitates sexual behavior as measured by increases in a lordosis response. Blood cortisol levels appear unchanged during female sexual arousal or orgasm; however, hypercortisolism has been associated with decreased libido or erectile dysfunction in men. Sexual stimuli involve a complex interplay between neurotransmitters and steroid hormones. The steroid hormones facilitate a sexual response by biasing sensorimotor integration so that a sexual stimulus is more likely to produce a sexual response. Testosterone administration to female-to-male transsexuals and androgen deprivation in male-to-female transsexuals support the theory that androgenic hormones play an important role in sexual desire. The androgens influence sexual desire, but are not sufficient alone as oral contraceptives and androgen antagonists do not consistently suppress libido, and patients with hypoactive sexual desire do not have lower androgen levels than women with normal sexual function. However, women report that their sexual desire diminishes following surgical menopause, and this desire can be restored with testosterone administration. Further evidence of androgen's effect on libido is illustrated when administration of dehydroepiandrosterone sulfate (DHEA-S) to postmenopausal women increases subjective ratings of sexual arousal when compared to placebo, although vaginal blood flow measures are unaffected. In premenopausal women, the effect of androgens is conflicting. There is a positive relationship between midcycle testosterone levels and coital frequency as well as testosterone levels and masturbation. In contrast to the role of androgens, estrogens appear to have little direct effect on sexual desire and functioning. Exogenous estrogen, particularly when administered orally, increases sex hormone binding globulin which in turn reduces free testosterone and estradiol. No differences in estradiol levels have been found in women with or without hypoactive sexual desire, or as a function of coital frequency. Estradiol levels influence central and peripheral nerve transmission and exert a vasoprotective effect on pelvic vasculature. Estrogen is necessary to maintain the function of the vaginal epithelium, stromal cells, and smooth muscles of the muscularis as well as thickness of the vaginal rugae and vaginal lubrication. Estrogen deficiency may negatively impact sexual functioning by resulting in atrophy of the vaginal epithelium causing decreased lubrication and dyspareunia. The role of progesterone on female sexual response still remains to be elucidated. Progesterone treatment does not appear to have a significant impact on sexual response, although there are some reports of diminished sexual desire with oral contraceptives or progestin implants. Other chemical mediators have been proposed to augment and initiate sexual motivation. Pheromones are chemical substances released by one member of a species to communicate with another member, to their mutual benefit, and are secreted from the glands at the anus, breasts, urinary outlet, and mouth. Studies of these chemicals secreted by insects and animals have led to speculation that perhaps these substances would affect human sexual behavior. The main olfactory system in humans may respond to odor cues with behavior significance. There is a strong olfactory input to the prefrontal cortex and decision making in this region of the brain appears to be influenced by emotional signals. Investigation of the role of pheromones in humans has found no evidence of any change in sexual behavior. It is apparent that the female sexual response is a complex integration of endocrine, neural, and neurochemical mediators. The steroid hormones appear to allow a permissive state to exist so that a sexually relevant stimulus can more likely elicit a sexual response. The long-term steroid effects may then induce a rapid physiologic change by altering levels of neurotransmitters. These chemical mediators are then communicated via multiple neural pathways which induce genital responses and sexual motivation. These complex interactions must be considered when attempting to define sexual functioning and to elucidate causes of sexual dysfunction.

SEXUAL DYSFUNCTION Clearly, the complexity of the normal sexual response eschews a simple cause and effect, search and find, approach to human sexual dysfunction. Indeed, sexual dysfunction is a multicausal and multidimensional problem combining biological, interpersonal, and psychological components. It is not surprising that the prevalence is high. A 1992 study of adult sexual behavior in the United States (National Health and Social Life Survey) revealed 43% of women exhibit sexual dysfunction. Approximately 20% of women reported sex was not pleasurable, nearly one-fourth were consistently unable to achieve orgasm, and 8% to 21% experienced pain during intercourse. Often women will experience more than one sexual dysfunction, as marked overlap exists. Risk Factors Several risk factors are known to lead to sexual dysfunction. These include emotional or stress-related problems, deterioration in economic position, history of sexual trauma, and lower educational levels. Postmarital (divorced, separated, or widowed) status is associated with an elevated risk of orgasmic difficulties when compared to married women, and married women are one and a half times more likely to achieve orgasm than unmarried women. Black women tend to exhibit higher rates of hypoactive sexual desire compared to white women (44% vs. 29%) who report more sexual pain disorders (16% vs. 13%). Hispanic women, in contrast, consistently report lower rates of sexual dysfunction. Evaluation History Obtaining a thorough history for the evaluation of sexual dysfunction is imperative. The physician should ensure that the patient is comfortable and is assured of strict confidentiality. The interview can begin with open-ended questions and gender inclusive terminology should be used, so as not to assume heterosexuality. An introductory question such as “Most people experience sexual concerns or problems at times during their lives. What problems, if any, have you experienced?” can be helpful. The expectations and goals of the patient and her partner, the patient's degree of commitment to the relationship, and the presence or extent of extramarital relationships should be ascertained. Inquiring as to what the patient believes is the etiology of her sexual dysfunction may provide some insight, as the patient may simply admit that she is not attracted to her partner or has an underlying fear of pregnancy. The patient should be asked to define specifically what she perceives as the sexual difficulty. When she noted the onset of the problem, precipitating circumstances, the frequency and intensity, and whether she has experienced difficulty with other partners should be determined. Early sexual development and education should be discussed. Severely negative family attitudes toward sex (often associated with religious orthodoxy), traumatic prior sexual experience, or gender identity confusion are relevant. Relationship issues should be explored, such as whether the

couple has divergent sexual preferences or concerns of partner fidelity, and if there is any underlying hostility or abuse. Cognitive factors such as ignorance of sexual anatomy or sexual response and acceptance of cultural myths impact greatly upon therapy and should be addressed from the outset. The physician should inquire about ongoing illnesses, prior injuries, surgical and psychiatric history, and alcohol and recreational drug use. A thorough medication history is crucial, specifically the use of ß- adrenergic blockers, CNS depressants, anticholinergics, or antidepressants ( Table 42.2). It should also be noted if the patient breast-feeding, or if she has recently undergone menopause.

TABLE 42.2. Medications and female sexual dysfunction Assessment Scales Several self-report questionnaires are available for assessing sexual dysfunction, all with some advantages and limitations. The Sexual Evaluation Scale is a 16-item scale that pertains to global factors of sexual interest, arousal, and performance, but does not address the factors of sexual drive and satisfaction. The Brief Index of Sexual Functioning is lengthier and includes weekly assessment of sexual function. However, it contains sexually explicit questions that may contribute to patient noncompliance. The Arizona Sexual Experience Scale is a brief five-question scale designed to evaluate psychotropic drug-induced changes in sexual functioning, and is helpful in patients who have recently initiated antidepressant therapy. The Female Sexual Function Index (FSFI) is a brief, self-reported measure of sexual function with a high degree of internal consistency and construct validity. The FSFI evaluates desire, orgasm, arousal, sexual pain, lubrication, and sexual satisfaction ( Appendix A). Examination A detailed examination should be performed in the presence of a chaperone. The internal/external genitalia are examined closely and evidence of vaginal atrophy should be noted. A cotton swab can be used to elicit painful areas, particularly in patients reporting entry dyspareunia, and the vestibular areas, Bartholin ducts, Skene ducts, and the urethral meatus should be carefully checked. Areas of erythema or tenderness should be noted. Some women have minute papillae of the vestibular skin which is a normal variant that does not represent viral or other disease entities. A vaginal exam is performed using one finger prior to the bimanual exam to minimize confusion arising from abdominal tenderness. Vaginismus may be apparent during the exam, however approximately 25% of women who tolerate vaginal exams or tampon insertion experience involuntary spasms during coitus. The lateral walls of the vagina are palpated along the bladder anteriorly, the posterior wall, and vaginal fornices. The vagina is visually evaluated using a narrow, warmed, and well-lubricated speculum. The presence or absence of vaginal rugae, fissures, or friable tissue should be noted. Characteristics such as skin elasticity, pubic hair, labial fullness, and evaluation of the introital and vaginal depth offer an assessment of vaginal atrophy. Inspecting the cervix may detect dysplastic lesions or evidence of infection, which warrants further evaluation, including a Pap smear, cultures, or wet mounts. The vaginal fornices around the cervix are palpated for nodularity which is suggestive of endometriosis or may represent fixed adnexa from pelvic inflammatory disease. Muscle tone and Kegel strength are ascertained in addition to the bulbocavernosus reflex and genital/perineal sensation is assessed. With two examining fingers in the vagina, locate the levator muscles which are at the 5- and 7-o'clock positions just superior to the hymeneal ring. The patient is then instructed to contract these muscles. To assist in proper muscle group isolation, various verbal cues can be tried, such as “squeeze as if you are trying to stop your urine stream” or “squeeze as if you are trying to prevent flatus” or “pull my examining fingers up into the vagina.” A correctly performed pelvic floor contraction is demonstrated by cephalad retraction of the perineum and anus, posterior rotation of the clitoris, and anterior displacement of the examining fingers. Once the patient is able to contract the muscles properly, ask her to contract as strongly as possible for as long as possible in order to assess baseline pelvic floor function. This maximal effort is rated according to a pelvic muscle rating scale ( Table 42.3).

TABLE 42.3. Pelvic muscle rating scale The bulbocavernosus reflex is evaluated by squeezing the clitoris and feeling (or seeing) the anal sphincter and perineal muscles contract. Alternatively, the reflex may be initiated by suddenly pulling the balloon of a Foley catheter against the vesicle neck. The absence of this reflex in a man is almost always associated with a neurologic lesion, but the reflex is not detectable in up to 30% of otherwise normal women ( Table 42.4).

TABLE 42.4. Gynecologic causes of female sexual dysfunction and method of gynecologic examination More involved evaluation includes determination of vaginal pH, vaginal wall compliance, genital vibratory perception thresholds, and genital hemodynamics. The use of Semmes-Weinstein monofilaments to assess vulvar sensitivity to pressure and touch has been proposed. These 20 nylon monofilaments are all of equal length but varying diameters. The filament is pressed against the skin until it bends, providing a consistent reproducible application of force. The filaments are inexpensive, convenient, easy to use, and associated with a high degree of patient compliance. This technique reveals that women with sexual dysfunction have significant diminished vulvar sensitivity to pressure/touch compared to women without sexual dysfunction. Blood flow to the clitoral, labial, urethral, vaginal, and uterine arteries can be recorded and both blood velocity and venous pooling may be compared pre- and poststimulation. While it may be possible to demonstrate these differences objectively, in order for these assessments to be clinically relevant, a patient must be able to recognize these changes as sexual arousal, and their absence must be significant enough to interfere with sexual functioning ( Fig. 42.3).

FIG. 42.3. Semmes-Weinstein monofilaments for detection of tactile neuropathy. (Photograph provided by Semmes-Weinstein Corp.) Laboratory Evaluation Measurements of estradiol, follicle-stimulating hormone, prolactin, and testosterone are often obtained in patients with sexual dysfunction; however, there is little evidence to support a direct correlation between sexual dysfunction and hormonal parameters. Most researchers agree that estrogen plays only a minimal role in female sexual desire, yet estrogen deficiency does affect genital vasocongestion and lubrication and may induce atrophy of vaginal epithelium. Androgens and their role in sexual desire have yielded conflicting results (see Normal Sexual Physiology). With natural menopause, androgen levels (testosterone, DHEA-S) have been positively correlated with sexual interest; however, no significant differences in circulating testosterone levels have been demonstrated in normal women compared to women without hypoactive sexual desire. Hemodynamic Testing Vaginal blood flow may be measured indirectly via vaginal photoplethysmography, pulsed wave Doppler ultrasonography, and measures of heat dissipation. Most of these techniques are reserved for the research setting. Vaginal photoplethysmography is the most studied technique in which a clear acrylic tampon is inserted into the vagina. This tampon contains either an incandescent light source or an infrared light-emitting diode as a light source and photosensitive light detector. The light source illuminates the capillary bed of the vaginal wall and the phototransistor detects the light that is reflected back from the vaginal wall and the

blood circulating within it. Disadvantages of this method are an inability to detect subtle changes in vaginal blood flow and often there is motion artifact. Measurement of tissue perfusion by measuring the cooling of a heated metal disk is another noninvasive technique that has been adapted for vaginal studies. A silver disk electrode is applied against the vaginal mucosa and maintained electrically at a temperature that exceeds body temperature. The electrical power needed to maintain this elevated temperature is then recorded. Increased blood perfusion under the disk will increase heat loss from the disk and require more power to maintain temperature. This power output therefore becomes an index of local blood perfusion. This technique is not affected by movement to the same degree as photoplethysmography. Classification The two primary classification systems for sexual dysfunction are the World Health Organization's International Statistical Classification of Diseases, 10th Revision ( ICD-10) and the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition DSM-IV ( DSM-IV). The ICD-10 definition of sexual dysfunction includes “the various ways in which an individual is unable to participate in a sexual relationship as he or she would wish.” Categories are lack or loss of sexual desire (F52.0), sexual aversion disorder (F52.1), failure of genital response (F52.2), orgasmic dysfunction (F52.3), nonorganic vaginismus (F52.5), nonorganic dyspareunia (F52.6), and excessive sexual drive (F52.7). The DSM-IV defines sexual dysfunction as a “disturbance in sexual desire and in the psychophysiologic changes that characterize the sexual response cycle and cause marked distress and interpersonal difficulty.” The described disorders include hypoactive sexual desire (302.71), sexual aversion (302.79), female arousal disorder (302.72), dyspareunia (302.76), vaginismus (306.51), and female orgasmic disorder (302.73). An interdisciplinary consensus, known as the International Consensus Development Conference on Female Sexual Dysfunction: Definitions and Classification, was convened to address the shortcomings and problems associated with the DSM-IV and ICD-10 classifications of sexual dysfunction. The panel consisted of representatives from a wide range of disciplinary backgrounds—endocrinology, family medicine, gynecology, pharmacology, psychiatry, psychology, physiology, urology, and rehabilitative medicine. The panel preserved the general structure of the DSM-IV and International Classification of Diseases, Ninth Revision ( ICD-9) classifications and expanded/altered the traditional definitions of sexual disorders. A new category of sexual pain disorder was added to include noncoital sexual pain disorder. Diagnosis of these disorders requires the component of “personal distress”. The diagnoses were also subtyped as (a) lifelong versus acquired, (b) generalized versus situational, and (c) for etiology—organic, psychogenic, mixed, or unknown ( Table 42.5).

TABLE 42.5. Classification of sexual dysfunction Desire Disorders Hypoactive sexual desire is the persistent or recurrent deficiency/absence of sexual fantasies/thoughts or desire for/receptivity to sexual activity which causes personal distress. Hypoactive sexual desire is the most commonly encountered dysfunction, seen in approximately 40% of patients in sexual counseling. Sexual desire is affected by many conditions such as medication use, life stages with accompanying hormonal changes, medical conditions, and interpersonal issues. Individuals with hypoactive sexual desire will also frequently exhibit difficulty with arousal and achieving orgasm. While obtaining a history, the physician should ask the patient how often she has felt sexual desire or interest, and inquire what she would rate her level of sexual interest on a scale from 1 to 5. The average population reports sexual desire approximately 3.4 times per month and a desire level of 3.5, as based on the FSFI control group. Hypoactive sexual desire is common in the female partners of men suffering from erectile dysfunction, and it is important to inquire regarding the partner's sexual functioning. Inquire regarding the presence of depression in the patient or partner, as decreased libido disproportionately affects patients with depression. In patients with major depression being evaluated for sexual dysfunction prior to the initiation of antidepressant therapy, only 50% of women and 75% of men reported sexual activity during the preceding month. Over 40% of men and 50% of women report decreased sexual interest. Reduced levels of arousal were more common in both depressed men and women (40% to 50%) than ejaculatory or orgasm difficulties. On physical exam the estrogen status of the external genitalia can carefully be evaluated by assessing the presence or absence of vaginal rugae, fissures, or friable tissue. Characteristics of skin elasticity, pubic hair, labial fullness, and evaluation of the introital and vaginal depth offer a further assessment of estrogen status. An extreme form of hypoactive sexual desire is sexual aversion disorder, which is the persistent or recurrent phobic aversion to and avoidance of sexual contact with a partner that causes personal distress. The etiology of sexual aversion disorder is often related to a history of traumatic sexual assault, such as a rape or child abuse. Arousal Disorders Sexual arousal disorder is the persistent or recurrent inability to attain or maintain sufficient sexual excitement causing personal distress which may be expressed as a lack of subjective excitement or genital (lubrication, swelling) or other somatic responses. Arousal difficulties may be physiologic or psychological. Insufficient foreplay, emotional distraction, lack of vaginal lubrication, decreased clitoral or labial sensation or engorgement, lack of vaginal smooth muscle relaxation, pelvic trauma, medication use, and postsurgical changes may all contribute to diminished sexual arousal. Arousal disorder can be elucidated by inquiring how often the patient felt sexually aroused over the preceding 4 weeks, how she would rate her sexual arousal on a scale from 0 to 5, and how satisfied she is with her level of arousal during sexual intercourse. The physician should ascertain how often the patient was able to become lubricated during sexual activity (0 = never to 5 = always), and how difficult it was to maintain lubrication (0 = very difficult to 5 = not difficult). Most women without sexual difficulty will report an average score of approximately 4 on these questions. On physical exam, careful attention should be paid to the estrogen status of the external genitalia and the degree of vulvar sensation and muscle tone. Orgasmic Disorders Orgasmic disorder is the persistent or recurrent difficulty, delay in, or absence of attaining orgasm following sufficient stimulation and arousal, which causes personal distress. Fifty percent of women report intermittent or situational difficulties in achieving orgasm and nearly 10% of sexually active women are incapable of attaining orgasm. There appears to be both physical and psychological components; as compared to orgasmic women, anorgasmic women report (a) greater discomfort in communicating with a partner regarding sexual activities involving direct clitoral stimulation, (b) more negative attitudes toward masturbation, (c) greater endorsement of sex myths, and (d) greater sexual guilt. Orgasmic disorders may be primary (never experiencing an orgasm) or secondary. The physician should inquire as to how often the patient was able to attain orgasm during the preceding 4 weeks (0 = never to 5 = always), and how difficult it was to achieve orgasm (0 = impossible to 5 = not difficult). Women without sexual difficulty report a score assessment of approximately 4 for each question. Inquire as to the ability of the patient's partner to maintain an erection, or if there is difficulty with premature ejaculation. Assess if there is a history of alcohol or drug abuse, and what prescription medications the patient is taking. On physical exam specifically evaluate muscle tone and review clitoral anatomy with the patient. Pain Disorders Sexual pain disorders encompass dyspareunia, vaginismus, and nonsexual pain categories. Dyspareunia is recurrent or persistent genital pain associated with sexual intercourse, and may result from genital stimulation, attempted entry, deep thrusting, or occur immediately after intercourse. Endometriosis, vestibulitis, hypoestrogenism, ovaries adherent in the cul-de-sac, and pelvic or urinary infections may be etiologic factors. Often women with chronic pelvic pain that were scheduled for laparoscopy are evaluated for interstitial cystitis by cystoscopy with hydrodistention and biopsy. Interstitial cystitis is diagnosed in approximately one third of women with a combination of urgency, frequency or nocturia, and positive cystoscopic findings. If sensation is impaired, neurologic referral may be indicated as disorders such as multiple sclerosis, spinal cord, or peripheral nerve injury may lead to sexual dysfunction. There is also a psychological component associated with these disorders as the anticipation of pain leads to decreased arousal and subsequent lack of desire. Vaginismus, the recurrent or persistent involuntary spasm of the musculature of the outer third of the vagina that interferes with vaginal penetration, is another sexual pain disorder leading to personal distress. The diagnosis of vaginismus however, does not require the presence of pain. Spasm of the pubococcygeus and levator ani muscle render vaginal penetration difficult, if not impossible. Frequently, this disorder is situational in that the patient may tolerate speculum or digital exams, but is unable to tolerate vaginal penetration during intercourse. This may be the function of either a psychological component or a conditioned response to painful stimuli. The final subgroup under the pain disorders is noncoital pain which includes such diagnoses as noncoital sexual pain that is recurrent or persistent genital pain induced by nonpenetrating sexual stimuli. To determine the degree of pain being experienced, the physician should inquire as to how often the patient has pain with and immediately following intercourse (1 = always to 5 = never). Most women on the FSFI report an average score of 4.6. Inquire if the pain is noted with entry or deep penetration, and if it varies with position. Ask if the patient experiences difficulty with lubrication or experiences discomfort with tampon insertion? Determine if the pain is localized specifically to the introitus or more pronounced during micturation? In addition to the general pelvic exam, the anterior vaginal wall should be carefully palpated to examine for the presence of a urethral diverticulum. The pathognomonic finding is a tender cystic swelling on the anterior vaginal wall. Patients in whom urethral diverticulum is strongly suspected may undergo voiding cytourethrography, followed by possible positive pressure urethrography, ultrasound, or MRI. Determine if the pain is superficial, deep, or vaginal, and attempt to reproduce the pain. Look for evidence of a poorly healed episiotomy scar or evidence of friable tissue at the vaginal cuff in a hysterectomy patient. A prolapsed fallopian tube into the cuff often results in significant dyspareunia. A biopsy and pathologic analysis of the tissue would be diagnostic. Aggressively seek the diagnosis of an underlying etiology, even if laparoscopy is required. Davis and Telinde state that “the most important single diagnostic instrument for the discovery of a suburethral diverticulum is a high index of clinical suspicion.”

ETIOLOGY OF SEXUAL DYSFUNCTION The interplay of vasogenic, musculogenic, neurogenic, hormonal, and psychogenic factors influences the female sexual response, and a disruption in any of these components may cause sexual dysfunction.

Neurogenic sexual dysfunction may result from spinal cord trauma or disease. In a complete transection of the spinal cord, no afferent impulses from the genitals can occur, however sexual fantasy and erotica have induced orgasm in some patients. Complete upper motor neuron injuries preclude vaginal lubrication, however women with incomplete injuries still maintain lubricating ability. Peripheral nervous system disease such as diabetes may impact nerve endings in the external genitalia precluding adequate arousal and the ability to achieve orgasm. Hormonal alterations have a significant impact upon sexual functioning. Surgical or medical menopause, premature ovarian failure, hormonal contraceptive use, hypothalamic–pituitary–axis dysfunction, and postpartum states may induce marked hormonal fluctuations and result in vaginal dryness, altered arousal, and decreased desire. Hypoestrogenism, from any cause, results in atrophy of the vaginal epithelium and a decrease in genital vasocongestion and lubrication. Administration of both androgen and estrogen to medically or surgically menopausal women has been shown to restore sexual desire; however estrogen treatment alone has not been shown to be successful. The role of exogenous androgens is still debated, as several studies have rendered somewhat mixed results. With natural menopause, androgen levels have been positively correlated with sexual interest, and there is a small but positive correlation between midcycle testosterone levels and intercourse frequency. These results are contrary to the lack of difference in testosterone levels between women with and without clinically diagnosed hypoactive sexual desire, and from studies that show oral contraceptives and androgen antagonists do not consistently suppress libido. There is a subset of women with hypoactive sexual desire in which testosterone treatment appears to be useful. Elevated levels of prolactin appear to suppress libido in women, and restoration of libido is noted with dopamine agonist therapy. Vascular-mediated effects are noted on sexual arousal in patients with high blood pressure, smoking, elevated cholesterol, and cardiac disease. Aortoiliac or atherosclerotic disease diminish pelvic blood flow and may reduce lubrication and clitoral engorgement. Cardiovascular disease may interfere with intercourse secondary to dyspnea. Traumatic injury to the arterial bed from fractures, surgical disruption, blunt trauma, or chronic perineal pressure from bicycle riding can result in diminished blood flow and subsequently sexual dysfunction. Impairments in pelvic floor musculature contribute to sexual dysfunction. When the levator ani muscles are hypertonic, vaginismus may result leading to dyspareunia and other sexual pain disorders. If hypotonic, urinary incontinence or anorgasmia can result. A poorly healed episiotomy site or vaginal delivery of a macrosomic infant may contribute to muscular dysfunction. The psychogenic component of sexual response is a critical one. Body image, self-esteem, and relational issues impact sexual arousal, as does emotional conflict such as anxiety, fear, anger, or guilt. An existing mood disorder is further compounded by the effects of psychotropic medications, which tend to inhibit sexual functioning in up to 75% of patients. SSRIs have been associated with decreased arousal, decreased genital sensation, and anorgasmia. Tranquilizers such as diazepam inhibit sexual function secondary to antidopaminergic action. Previous unresolved negative experiences such as sexual assault or domestic abuse will impact desire and arousal, and may manifest as a sexual aversion disorder or vaginismus.

TREATMENT A multidisciplinary approach is imperative in thorough evaluation and treatment of sexual dysfunction. Psychiatric consultation is necessary for patients with a suspected assault history or evidence of depressive symptomatology. Urologic referral, particularly in patients with pelvic pain may reveal the presence of interstitial cystitis or a urethral disorder such as a urethral diverticulum. After arriving at a working diagnosis, diagnosis-specific treatment is most efficacious. Desire Disorders Pharmacologic Treatment Hormone Replacement Therapy Indicated for use in postmenopausal women, estrogen replacement therapy results in improved clitoral sensitivity, increased libido, and decreased pain during sexual intercourse. Estrogen creams or the vaginal estradiol ring (Estring) may reduce complaints of vaginal irritation, pain, or dryness secondary to atrophy. Long-term use of estrogen vaginal creams is considered an unopposed estrogen treatment in women with an intact uterus and requires progesterone opposition. An oral progesterone, such as medroxyprogesterone 5 mg daily for 10 days a month (or equivalent), may be used initially, with frequency or dose adjustment if breakthrough bleeding occurs. The estradiol ring has little systemic absorption and does not require the addition of progesterone. Patients who are uncomfortable wearing the ring during the day often achieve relief with night use only. Testosterone Testosterone replacement therapy has been prescribed for menopausal woman with symptoms of inhibited sexual desire, dyspareunia, and decreased lubricating ability. In women in whom menopause had been surgically induced, supraphysiologic doses of testosterone enanthate given by injection alone or in conjunction with estrogen increase arousal, desire, sexual activity, orgasm frequency, and fantasies more than estrogen alone. Transdermal testosterone (300 µg) given to women following hysterectomy and oophorectomy results in increased sexual frequency and orgasm despite an appreciable placebo response. The percentage of women with sexual fantasies, masturbation, or coitus at least once per week increases two to three times above baseline. Transdermal testosterone has no significant effects on the serum concentration of high-density lipoprotein, total cholesterol, low-density lipoprotein, triglycerides, liver function tests, or fasting insulin/glucose levels. Serum free and bioavailable testosterone increase to normal values. Dihydrotestosterone and total testosterone increase to above normal with treatment. This is not a treatment approved by the Food and Drug Administration (FDA), and no treatment guidelines for testosterone replacement therapy exist. Many physicians are concerned about the lack of safety data on the role of testosterone in breast cancer and hepatic side effects; however hepatocellular damage or carcinoma is rare at prescribed dosages, and the development of breast cancer has not been reported clinically. Potential side effects of testosterone, which may occur in 5% to 35% of patients, include lipoprotein alterations, acne, hirsutism, clitoromegaly, and voice deepening. Before initiating testosterone treatment, physicians should discuss the potential and theoretic risks as well as the individual risk and benefit assessments with the patient. In general, patients with current or previous breast cancer, uncontrolled hyperlipidemia, liver disease, acne, or hirsutism should not receive testosterone therapy ( Table 42.6).

TABLE 42.6. Dosages for testosterone administration a

Arousal Disorders Pharmacologic Treatment Sildenafil Citrate Sildenafil citrate (Viagra) is a selective inhibitor of phosphodiesterase-5 and therefore enhances and prolongs the cellular effects of the second messenger substrate cyclic guanosine monophosphate (cGMP), the specific substrate for this enzyme. The guanylyl cyclase/cGMP pathway is the signal transduction mechanism activated by nitric oxide (NO). NO is released by the endothelial cells and the postsynaptic parasympathetic cavernosa neurons in males and leads to a cascade of chemical reactions that ultimately relaxes smooth muscle, dilates arteries, and engorges the corpora cavernosa. Therefore, once the psychological and physical arousal set in motion these chemical events in the corpora cavernosa, sildenafil, when taken 1 hour prior to sexual activity, helps maintain erections in men. Research has also identified phosphodiesterase-5 in human clitoral tissue. In women, there is the potential that sildenafil may be useful in treating orgasm delay and anorgasmia by facilitating lubrication, clitoral perfusion and engorgement, and the swelling and engorgement of the female genitalia. The role of sildenafil in treating sexual dysfunction in women, however, remains to be determined. In women experiencing psychotropic medication-induced sexual dysfunction, there may be a role for sildenafil. There may be some improvement in sexual functioning in women who reported sexual dysfunction since initiating antidepressant or antipsychotic medication therapy. Further clinical studies evaluating the safety and efficacy of sildenafil for treatment of female sexual dysfunction are needed. Nonpharmacologic Treatment The EROS-CTD (Clitoral Therapy Device, Urometrics, Inc. St Paul, MN) is the first FDA-approved therapy for female sexual dysfunction. This small, battery-powered vacuum device is designed to enhance clitoral enlargement, increase blood flow to the clitoris, and improve sexual arousal. Women with sexual arousal disorder noted improvement in all symptoms of female sexual arousal disorder, and patients without such disorder also report similar changes in sensation, lubrication, ability to orgasm, and overall satisfaction. Orgasmic Disorders Anorgasmia treatment relies on maximizing stimulation and minimizing inhibition. Stimulation may include masturbation with prolonged stimulation (initially up to 1 hour) with the use of a vibrator as needed. Surveys of young women reveal that the most effective techniques for triggering orgasm are manual and oral stimulation of the clitoris, and that joining the male's thrusting pattern without additional clitoral stimulation was one of the least effective methods. Muscular control of sexual tension such as alternating contraction and relaxation of the pelvic musculature during high sexual arousal has proven helpful. The volitional muscle control is similar to Kegel exercises ( Table 42.7). Methods to minimize inhibition include distraction by fantasy or listening to music. Past failure to achieve orgasm can elicit self-defeating and distracting thoughts about whether the patient will be able to achieve orgasm. The patient mentally monitors her own and her partner's physical response, unable to relax and enjoy the sexual stimulation for its own sake. Such behavior makes the patient more a spectator than a participant in her own sexual pleasure. Masters and Johnson coined the term “spectatoring,” which they describe as a major deterrent to effective sexual functioning. Kaplan suggests this obsessive self-observation

arising out of fear of failure to be the single most immediate cause of female orgasmic dysfunction.

TABLE 42.7. Kegel exercises

Masters and Johnson's approach to anorgasmia used a temporary ban on orgasm and sexual intercourse until permitted by their sensate focus program. Their exercises focused on “non-demand pleasuring” and attempted to improve sexual skills while avoiding goal-oriented behavior. Their program also included assertiveness training, modeling, behavioral reversal, and education. Although widely accepted among sex therapists, their “success” rates of 80% could not be reproduced by other therapists. Group therapy and erotic fantasy have also been proposed. Pain Disorders Vestibulitis Several treatment modalities have been employed in the treatment of pain disorders. The most common type of premenopausal dyspareunia is vulvar vestibulitis. The most commonly used treatments are cognitive-behavioral therapy, biofeedback, and vestibulectomy. Typical cognitive–behavioral interventions aim at reducing pain and improving sexual functioning and include Kegel exercises, vaginal dilation, and relaxation techniques. Surface electromyographic biofeedback has been applied to dyspareunia. Vestibulectomy, in which the painful tissue of the vulvar vestibule is excised, has variable success rates, ranging from 43% to 100%. Vaginismus Treatment of vaginismus consists of smooth muscle relaxation and vaginal dilation. Muscle relaxation can be taught during an exam by having the patient isolate and alternatively contract and relax the pelvic musculature around the examiner's finger. Commercial dilators or tampons of increasing diameter may be placed in the vagina for 15 minutes twice daily to facilitate muscle relaxation and dilation ( Fig. 42.4).

FIG. 42.4. Vaginal dilators. (Photograph provided by Milex Products Inc.) Interstitial Cystitis Mild symptoms of interstitial cystitis may respond to dietary modifications, stress reduction, biofeedback, and bladder retraining. Avoiding acidic, alcoholic, or carbonated beverages or spicy foods and artificial sweeteners have been reported by some patients to reduce their interstitial cystitis pain. The only oral medication currently FDA-approved to treat interstitial cystitis is pentosan polysulfate sodium (Elmiron), and other medications such as tricyclic antidepressants that have analgesic properties, antiinflammatory drugs, antihistamines, antispasmodics, and muscle relaxants may offer relief. More invasive treatments include hydrodistention of the bladder done at time of cystoscopy, bladder instillation with dimethyl sulfoxide (DSMO), or an implantable sacral nerve stimulator. Surgical procedures such as bladder augmentation or urinary diversion are used as a last resort. SUGGESTED READINGS Anatomy Berman JR, Berman LA, Lin H, et al. Effect of sildenafil on subjective and physiologic parameters of the female sexual response in women with sexual arousal disorder. J Sex Marital Ther 2001;Oct–Dec:27. Kegel AH. Sexual functions of the pubococcygeus muscle. West J Surg Obstet Gynecol 1952;60:521. Masters WH, Johnson VE. Human sexual response. Boston: Little, Brown and Co, 1966. Schultz WW, van Andel P, Sabelis I, et al. Magnetic resonance imaging of male and female genitals during coitus and female sexual arousal. BMJ 1999;319:18–25. Shafik A. The role of the levator ani muscle in evacuation, sexual performance and pelvic floor disorders. Int Urogynecol Pelvic Floor Dysfunct 2000;11:361–376.

Normal Female Sexual Response Cycle Basson R. Human sexual response cycles. J Sex Marital Ther 2001;27:33–43. Kaplan HS. The sexual desire disorders. New York: Bruner/Mazel, 1995.

Physiology Carter CS. Oxytocin and sexual behavior. Neurosci Biobehav Rev 1992;16:131–144. Chevalier G, Deniau JM. Disinhibition as a basic process in the expression of striatal functions. Trends Neurosci 1990;13:277–280. Exton MS, Bindert A, Krugert T, et al. Cardiovascular and endocrine alterations after masturbation induced orgasm in women. Psychosom Med 1999;61:280–289. Hull EM, Lorrain DS, Jianfang DU, et al. Hormone neurotransmitter interactions in the control of sexual behavior. Behav Brain Res 1999;105:105–116. Levin RJ. The physiology of sexual function in woman. Clin Obstet Gynecol 1980;7:213–229. Meston CM, Frohlich PF. Neurobiology of sexual function. Arch Gen Psychiatry 2000;57:1012–1030. Schriener-Engel P, Schiavi RC, White D, et al. Low sexual desire in women: the role of reproductive hormones. Horm Behav 1989;23:221–234. Van Goozen SM, Frija NH, Wiegant VM, et al. The premenstrual phase and reactions to adversive stimuli and emotionality. Psychneuroendocrinology 1996;21:479–497.

Sexual Dysfunction Lauman EO, Gagnon JH, Michael RT, et al. The social organization of sexuality: sexual practices in the United States. Chicago: University of Chicago Press, 1994. Lauman EO, Paik A, Rosen Raymond. Sexual dysfunction in the United States: prevalence and predictors. JAMA 1999;281:537–544.

Evaluation Blaivas JG. The bulbocavernosus reflex in urology: a prospective study of 299 patients. J Urol 1981;126:197. McGahuey CA, Gelenberg AJ, Laukes CA, et al. The Arizona sexual experience scale (ASEX): reliability and validity. J Sex Marital Ther 2000;25:25–40. Othmer E, Othmer SC. Evaluation of sexual dysfunction. J Clin Psychiatry 1987;48:191–193. Romanzi LJ, Groutz A, Feroz F, et al. Evaluation of female external genitalia sensitivity to pressure/touch: a preliminary prospective study using Semmes-Weinstein monofilaments. Urology 2001; 57:1145–1150.

Rosen R, Brown C, Heiman J, et al. The female sexual function index: a multidimensional self-report instrument for the assessment of female sexual function. J Sex Marital Ther 2000;26;191–208. Taylor JF, Rosen RC, Leiblum SR. Self-report assessment of female sexual function: psychometric evaluation of the Brief Index of Sexual Functioning for Women. Arch Sex Behav 1994;23:627–643.

Classification Clemons J, Arya LA, Myers DL. Diagnosing interstitial cystitis in women with chronic pelvic pain. Obstet Gynecol 2001;97[4 Supp1]:s7. Kelly MP, Straussberg DS, Kircher JR. Attitudinal and experiential correlates of anorgasmia. Arch Sex Behav 1990;19:165–177. Kennedy SH, Dickens SE, Eisfeld BS, et al. Sexual dysfunction before antidepressant therapy in major depression. J Affect Disorders 1999;56:201–208.

Etiology Kohn II, Kaplan SA. Female sexual dysfunction: what is known and what can be done? Contemp Obstet Gynecol 2000;45:25–46. Krane R, Goldstein I, Traish A, et al. New perspectives in the management of female sexual dysfunction. Burlington, MA: Boston University School of Medicine, 1998.

Treatment Davis S, McCloud P, Strauss B, et al. Testosterone enhances estradiol's effect on postmenopausal bone density and sexuality. Maturitas 1996;21:227–236. De Bruijn G. From masturbation to orgasm with a partner: how some women bridge the gap and others don't. J Sex Marital Ther 1982;8:151–167. Kaplan HS. The evaluation of sexual disorders: psychological and medical aspects. New York: Bruner/Mazel, 1983. Kaplan SA, Reis RB, Kohn IJ, et al. Safety and efficacy of sildenafil in postmenopausal women with sexual dysfunction. Urology 1999;53:481–486. Phillips NA. Female sexual dysfunction. Am Fam Physician 2000;July:1–14. Salerian AJ, Deibler WE, Vittone BJ, et al. Sildenafil for psychotropic-induced sexual dysfunction in 31 women and 61 men. J Sex Marital Ther 2000;26:133–140. Sherwin BB, Gelfand MM, Brender W. Androgen enhances sexual motivation in female, a prospective, crossover study of sex steroid administration in the surgical menopause. Psychosom Med 1985;47:339–351. Utiger RD. A pill for impotence. New Engl J Med 1998;338:1458–1459. Wilson SK, Delk JR II, Billups KL. Treating symptoms of female sexual arousal disorder with the EROS-Clitoral Therapy Device. J Gend Specif Med 2001;4:54–58.

Appendix A—Female Sexual Function Index (FSFI) a

Questions Q1: Over the past 4 weeks, how often did you feel sexual desire or interest? 5 = Almost always or always 4 = Most times (more than half the time) 3 = Sometimes (about half the time) 2 = A few times (less than half the time) 1 = Almost never or never Q2: Over the past 4 weeks, how would you rate your level (degree) of sexual desire or interest? 5 = Very high 4 = High 3 = Moderate 2 = Low 1 = Very low or none at all 0 = No sexual activity Q3: Over the past 4 weeks, how often did you feel sexually aroused (“turned on”) during sexual activity or intercourse? 5 = Almost always or always 4 = Most times (more than half the time) 3 = Sometimes (about half the time) 2 = A few times (less than half the time) 1 = Almost never or never 0 = No sexual activity Q4: Over the past 4 weeks, how would you rate your level of sexual arousal (“turn on”) during sexual activity or intercourse? 5 = Very high 4 = High 3 = Moderate 2 = Low

1 = Very low or none at all 0 = No sexual activity Q5: Over the past 4 weeks, how confident were you about becoming sexually aroused during sexual activity or intercourse? 5 = Very high confidence 4 = High confidence 3 = Moderate confidence 2 = Low confidence 1 = Very low or no confidence 0 = No sexual activity Q6: Over the past 4 weeks, how often have you been satisfied with your arousal (excitement) during sexual activity or intercourse? 5 = Almost always or always 4 = Most times (more than half the time) 3 = Sometimes (about half the time) 2 = A few times (less than half the time) 1 = Almost never or never 0 = No sexual activity Q7: Over the past 4 weeks, how often did you become lubricated (“wet”) during sexual activity or intercourse? 5 = Almost always or always 4 = Most times (more than half the time) 3 = Sometimes (about half the time) 2 = A few times (less than half the time) 1 = Almost never or never 0 = No sexual activity Q8: Over the past 4 weeks, how difficult was it to become lubricated (“wet”) during sexual activity or intercourse? 0 = No sexual activity 1 = Extremely difficult or impossible 2 = Very difficult 3 = Difficult 4 = Slightly difficult 5 = Not difficult Q9: Over the past 4 weeks, how often did you maintain your lubrication (“wetness”) until completion of sexual activity or intercourse? 5 = Almost always or always 4 = Most times (more than half the time) 3 = Sometimes (about half the time) 2 = A few times (less than half the time) 1 = Almost never or never 0 = No sexual activity Q10: Over the past 4 weeks, how difficult was it to maintain your lubrication (“wetness”) until completion of sexual activity or intercourse? 0 = No sexual activity 1 = Extremely difficult or impossible 2 = Very difficult 3 = Difficult 4 = Slightly difficult 5 = Not difficult

Q11: Over the past 4 weeks, when you had sexual stimulation or intercourse, how often did you reach orgasm (climax)? 5 = Almost always or always 4 = Most times (more than half the time) 3 = Sometimes (about half the time) 2 = A few times (less than half the time) 1 = Almost never or never 0 = No sexual activity Q12: Over the past 4 weeks, when you had sexual stimulation or intercourse, how difficult was it for you to reach orgasm (climax)? 0 = No sexual activity 1 = Extremely difficult or impossible 2 = Very difficult 3 = Difficult 4 = Slightly difficult 5 = Not difficult Q13: Over the past 4 weeks, how satisfied were you with your ability to reach orgasm (climax) during sexual activity or intercourse? 5 = Very satisfied 4 = Moderately satisfied 3 = About equally satisfied and dissatisfied 2 = Moderately dissatisfied 1= Very dissatisfied 0 = No sexual activity Q14: Over the past 4 weeks, how satisfied have you been with the amount of emotional closeness during sexual activity between you and your partner? 5 = Very satisfied 4 = Moderately satisfied 3 = About equally satisfied and dissatisfied 2 = Moderately dissatisfied 1= Very dissatisfied 0 = No sexual activity Q15: Over the past 4 weeks, how satisfied have you been with your sexual relationship with your partner? 5 = Very satisfied 4 = Moderately satisfied 3 = About equally satisfied and dissatisfied 2 = Moderately dissatisfied 1= Very dissatisfied Q16: Over the past 4 weeks, how satisfied have you been with your overall sexual life? 5 = Very satisfied 4 = Moderately satisfied 3 = About equally satisfied and dissatisfied 2 = Moderately dissatisfied 1= Very dissatisfied Q17: Over the past 4 weeks, how often did you experience discomfort or pain during vaginal penetration? 0 = Did not attempt intercourse 1 = Almost always or always 2 = Most times (more than half the time)

3 = Sometimes (about half the time) 4 = A few times (less than half the time) 5 = Almost never or never Q18: Over the past 4 weeks, how often did you experience discomfort or pain following vaginal penetration? 0 = Did not attempt intercourse 1 = Almost always or always 2 = Most times (more than half the time) 3 = Sometimes (about half the time) 4 = A few times (less than half the time) 5 = Almost never or never Q19: Over the past 4 weeks, how would you rate your level (degree) of discomfort or pain during or following vaginal penetration? 0 = Did not attempt intercourse 1 = Very high 2 = High 3 = Moderate 4 = Low 5 = Very low or none at all

Scoring System The individual domain scores and full scale score of the FSFI are derived by the computational formula outlined in the table below. Individual domain scores are obtained by adding the scores of the individual items that comprise the domain and multiplying the sum by the domain factor (see below). The full scale score is obtained by adding the six domain scores. It should be noted that within the individual domains, a domain score of zero indicates that no sexual activity was reported during the past month.

Domain

Questions Range

Desire 1,2 Arousal 3,4,5,6 Lubrication 7,8,9,10 Orgasm 11,12,13 Satisfaction 14,15,16 Pain 17,18,19

Factor Min Score Max Score

1–5 0.6 0–5 0.3 0–5 0.3 0–5 0.4 0 (or 1)–5 0.4 0–5 0.4

1.2 0 0 0 0 0

6.0 6.0 6.0 6.0 6.0 6.0

Footnote aFor

the complete FSFI questionnaire, instructions, and scoring algorithm, please go to http://www.fsfiquestionnaire.com/, or contact Raymond Rosen, Ph.D., Department of Psychiatry, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, 675 Hoes Lane, Piscataway, NJ 08854.

Chapter 43 Chronic Pelvic Pain Danforth’s Obstetrics and Gynecology

Chapter 43 Howard T. Sharp

Chronic Pelvic Pain

GLOSSARY CONTEMPORARY PAIN THEORIES PUTATIVE PELVIC PAIN STATES Adhesions Endometriosis Pelvic Inflammatory Disease Myofascial Pain Pelvic Congestion Irritable Bowel Syndrome Ovarian Remnant Syndrome Residual Ovary Syndrome Pain of Uterine Origin PSYCHOLOGICAL ASPECTS OF CHRONIC PELVIC PAIN PATIENT HISTORY PHYSICAL EXAMINATION PAIN MANAGEMENT NSAIDs Neurolytic Agents Narcotics Narcotic Agreements SUMMARY POINTS SUGGESTED READINGS Glossary and Contemporary Pain Theories Putative Pain States Endometriosis Pelvic Inflammatory Disease and Myofascial Pain Pelvic Congestion Syndrome OVARIAN REMNANT SYNDROME Pain of Uterine Origin

Science has produced many drugs, therapies, and surgeries in attempts to relieve the suffering associated with chronic pelvic pain (CPP). Unfortunately, only modest success has been achieved. This likely is due, in part, to the heterogeneity that exists in this ill-defined population of patients, as well as our limited understanding of pain modulation. Often, when standard therapies fail, or when no visible pathology can be identified, other specialties such as gastroenterology, urology, neurology, or psychiatry are consulted. Patients eventually may be referred to centers specializing in CPP or empiric pain management, wherein neurolytic or opioid therapy may be commenced without finding a specific cause for pain. Along the course of therapy and referral, frustration is often a byproduct of treating a supposedly remediable pain state only to have pain persist. Furthermore, a confusing aspect of CPP is that its definition implies that no etiology can be found; however, most textbooks list several causes for CPP (adhesions, endometriosis, pelvic inflammatory disease [PID], etc.). Therefore, the clinical challenge is to decipher which of the following is most likely: 1. A defined pain state exists but has been improperly diagnosed and treated. 2. A defined pain state exists, has been correctly identified, but happens to be present in an incidental form. 3. A neuropathic pain processing state is present. GLOSSARY Allodynia: pain due to a stimulus which does not normally provoke pain. Central pain: pain initiated or caused by a primary lesion of dysfunction in the central nervous system. CPP: nonmenstrual pain of 6 months' duration or greater, localized to the pelvis, anterior abdominal wall below the pelvis, or lower back, severe enough to result in functional disability or require medical or surgical treatment. Dysesthesia: an unpleasant abnormal sensation, whether spontaneous or evoked. Endometriosis: the existence of two or more of the following outside of the endometrium: (a) endometrial epithelium, (b) endometrial stroma, (c) endometrial glands, (d) hemosiderin-laden macrophages. Hyperalgesia: an increased response to a stimulus which is normally painful. Myofascial pain syndrome: a heterogeneous pain-producing disorder characterized by localized, reproducible, hyperirritable trigger points within a muscle or its investing fascia. Neuropathic pain: pain initiated or caused by a primary lesion, dysfunction, or transitory perturbation in the peripheral or central nervous system. Neuralgia: pain in the distribution of a nerve or nerves. Pain: an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of damage.

CONTEMPORARY PAIN THEORIES Until the 1960s, pain was considered a sensory response to tissue damage. It is now widely accepted that pain has a reactive or emotional as well as sensory component, which is influenced by genetic differences, experiences, gender, anxiety, or expectation. The perception of and response to pain are believed to be determined by four simultaneous processes which include: transduction (depolarization of a peripheral sensory nerve ending to generate an impulse), transmission (neural events to carry the impulse), modulation (neural events that control transmission neurons), and perception (event processing influenced by behavioral and emotional factors). The theory most widely accepted to explain the mechanism whereby pain transmission occurs was described by Melzak and Wall as the gate-control theory. This theory suggests that the modulation of nociception at the spinal cord is mediated by descending signals for high centers within the brain through neurotransmitters. It describes a bidirectional gate at the level of the spinal cord, rather than merely unidirectional pain transmission as previously held by the Cartesian theory.

PUTATIVE PELVIC PAIN STATES Adhesions Pelvic inflammatory disease, endometriosis, inflammatory bowel disease, or surgery may cause adhesions; yet, in up to 50% of cases there may be no significant antecedent event. Although adhesions commonly are found in patients with CPP, it may be difficult to assess whether they are contributing to pelvic pain or are merely incidental findings. Several studies have raised questions about the surgical treatment of adhesions. First, in a prospective study which involved second-look laparoscopy following laparotomy for reproductive surgery, 51% of patients developed de novo adhesions. Therefore, the possibility exists that treating adhesions may lead to more adhesions. Secondly, there is evidence to suggest that the lysis of adhesions may not result in significant pain relief. A randomized, prospective study of patients with CPP and known stage II to IV (moderate to severe) pelvic adhesions assessed the efficacy of surgical adhesiolysis on patients with CPP. Patients either underwent minilaparotomy adhesiolysis or no treatment. At 9- to 12-month follow-up, using the McGill Pain Scale to compare outcomes, there was no difference in the treatment group (n = 24) compared with controls (n = 24). After stratification, the authors acknowledged a trend toward a benefit in patients with well-vascularized or thick adhesions involving the intestinal tract, the pain hypothesis being symptoms of intermittent small bowel subileus. This small study lacks the power to answer definitively

the question as to whether treatment of adhesions is beneficial. In a case-control study of 100 consecutive laparoscopies in patients with CPP and 88 laparoscopies for infertility, pelvic adhesions were found more often in the control group (26% versus 39%). Trimbos and others studied 200 asymptomatic women undergoing laparoscopic sterilization and found a 14% incidence of pelvic adhesions. In a prospective cohort study of 102 women undergoing laparoscopy, wherein the surgeons were blinded as to the indications, which were pain (64%), infertility (35%), previous abnormal findings (19%), and sterilization (15%), adhesion scores were no different in the CPP group compared with controls. In an uncontrolled, prospective study of 30 women undergoing laparoscopic adhesiolysis for CPP with a mean follow-up interval of 8.2 months, there was an overall improvement of pain in 63%. There was a trend toward greater improvement in the group with CPP, compared with the group exhibiting CPP syndrome. CPP syndrome was defined as having at least four of the following (a) pain longer than 6 months' duration, (b) previous treatments unsuccessful in relieving pain, (c) diminished physical activity (work, exercise, sex), (d) at least one vegetative sign of depression (sleep dysfunction, decreased appetite, psychomotor retardation), (e) altered family role. Laparoscopic pain mapping under conscious sedation provides additional information to better define whether a patient's adhesions are associated with pain. This technique employs local anesthesia at trocar sites in combination with intravenous opioids and benzodiazepines to enable the conscious patient to tolerate laparoscopic surgery. Pelvic structures can be mapped using a verbal analog pain scale (VAPS) from 0 to 10 to determine whether pain can be generalized to the pelvic viscera or to a focal point or adhesion. It is important to distinguish between surgically induced pain and pain which truly mimics the patient's chronic pain. Taking these studies into consideration we can learn some valuable lessons. (a) The more surgery that is performed, the more likely adhesions are to occur. (b) Adhesions may be present in asymptomatic patients. (c) Adhesions may be incidental in symptomatic patients. (d) Although some case series have shown some benefit to adhesiolysis, the only randomized clinical trial of adhesiolysis in patients with CPP showed no treatment benefit. (e) Conscious sedation pain mapping surgery may better enable us to determine which adhesions are painful. (f) Judgment is critical as to whether or not to perform surgery in patients who have had multiple laparoscopic procedures, weighing the risks and benefits. Endometriosis Endometriosis is one of the most enigmatic of gynecologic disorders. Not only is there little correlation between the extent of disease and the degree of pain, but it often is found in asymptomatic women. The noted exception is deep, infiltrating endometriosis in the rectovaginal septum, which has been shown to have a direct correlation to pain. Endometriosis may have several appearances, ranging from the more typical “powder burn,” blue-gray lesions, to atypical lesions which may be clear, red, or white. Various explanations have been proposed to account for endometriosis-related pain including inflammation, prostaglandin production, neuronal involvement, and adhesions. Data from conscious sedation laparoscopic pain mapping have demonstrated areas of pain well beyond the visible endometriosis. Symptoms associated with endometriosis include cyclic pelvic pain or dysmenorrhea. The pain associated with endometriosis may precede the menses, occur with menses, and continue after menses. Tenesmus may be associated with involvement of the rectosigmoid colon. Other clinical manifestations may include dyspareunia or ovarian mass (endometrioma). Conservative medical treatment is recommended as the initial therapy for endometriosis or presumed endometriosis. This may include nonsteroidal antiinflammatory drugs (NSAIDs), oral contraceptive pills (OCPs), danazol, progestins, or gonadotropin-releasing hormone (GnRH) agonists. These medications have been shown to reduce the size of endometriotic lesions and, thus, the stage; however, they have not been shown to eliminate disease. There are no convincing data to suggest that one form of medical suppressive therapy is superior to another. Usually, the agent with the lowest side effect profile, such as an NSAID or OCP, is selected as first-line therapy. Oral contraceptive pills may be used in a cyclic (standard) or continuous fashion. Continuous use refers to the ingestion of active pills every day for 3 to 6 months, rather than allowing menses to occur during placebo administration. This usually allows the patient to avoid the dysmenorrhea which may occur with the cyclic regimen. If OCPs are not effective after 3 months, danazol (600–800 mg/d) or GnRH agonists may be used. The use of danazol for endometriosis-associated pain was evaluated by Cochrane Database reviewers, wherein four trials were found to have study design adequate for inclusion. The reviewers concluded that danazol is effective in treating the symptoms and signs of endometriosis; however, its use is limited by the occurrence of androgenic side effects. GnRH agonists such as leuprolide or goserelin generally are used for 6 months if a favorable response is noted. The main concern about prolonged use of these agents is the loss of bone mineral density. The treatment window may be extended to 12 months with the use of add-back therapy (addition of norethindrone acetate with or without conjugated equine estrogens). Surgery generally is reserved for refractory cases. If surgery is to be performed, it should be tailored toward the patient's reproductive wishes. If the patient desires to preserve her childbearing capacity, endometriotic lesions may be destroyed or removed by vaporization or excision laparoscopically or by laparotomy. It is argued by some that vaporization may treat only the “tip of the iceberg,” with the potential to leave deep, infiltrating endometriosis behind. To date, there are no convincing data to recommend one laparoscopic therapy over the other. Unfortunately, a significant number of patients treated conservatively (without hysterectomy and bilateral salpingo-oophorectomy) will develop recurrent symptoms 12 months after surgery. Laparoscopic surgery for the treatment of endometriosis-associated pain was reviewed in the Cochrane Database. Only one study was deemed adequate for evaluation; therefore, most of our understanding stems from level 3 data (case series and opinions of experts). The review concluded that laparoscopic laser treatment of endometriosis was more effective than expectant treatment of endometriosis, but the reviewers included a caution about the interpretation of results due to the lack of any corroborating studies. Hysterectomy with or without adnexectomy may be appropriate in cases in which childbearing is no longer an issue. Pelvic Inflammatory Disease PID is a significant health problem (approximately 1 million cases per year), resulting in an expense of $3.5 billion annually in the United States alone. It clearly can be a cause of acute pain, yet it also may be asymptomatic. The most likely mechanism for pain is from inflammation and distension of the fallopian tubes. A distended, fluid-filled fallopian tube or hydrosalpinx will sometimes persist for months or years and may cause CPP. It is less clear why pain persists in patients with treated PID, who subsequently have normal-appearing reproductive organs and whose culture results are negative for causative microorganisms. It is theorized that the initial inflammatory insult may have started a cascade of signals within the pelvis, spinal cord, and brain, resulting in visceral neuropathic pain. One study estimated that as many as 15% to 25% of patients with PID will go on to have CPP (relative risk = 3.0–5.0). There are no good studies addressing how to treat patients with laparoscopy-negative, culture-negative presumed persistent PID. They, therefore, often are treated with NSAIDs, neurolytic agents, or opioids. The patient's partner should be treated, also, to avoid reinfection. Persistent hydrosalpinx usually is treated surgically by salpingectomy. Myofascial Pain Myofascial pain is common in patients with a history of abdominal trauma or multiple surgeries and often is overlooked as a cause for CPP. Myofascial pain syndrome is a heterogeneous pain-producing disorder characterized by localized, reproducible, hyperirritable trigger points within a muscle or its investing fascia. The clinical sign of the trigger point is that it is tender when compressed, and has characteristic referred pain patterns, referred tenderness, motor dysfunction, or autonomic dysfunction. Abdominal wall myofascial pain is detected best by isolating the rectus abdominus muscles, by having the patient flex her abdomen by lifting her feet or head and shoulders off the examination table while in the supine position. A one-finger search along the anterior abdominal wall is performed to identify painful trigger points. When localized, trigger points can be treated successfully with icing, stretching exercises, and with local anesthetic injection. One to two milliliters of a 50:50 mixture of 1% lidocaine and 0.25% bupivacaine may be injected into the muscle and fascia with a 22- or 25-gauge needle to achieve a diagnostic and therapeutic block. Slocumb reported on the successful treatment of 89% of 131 patients with CPP who had trigger point injections. Most patients obtained relief within five injections. Physical therapy also has been successful in treating myofascial pain syndrome. Pelvic Congestion

Patients with pelvic congestion syndrome typically complain of pelvic pain and aching which becomes progressively worse throughout the day, and they also may complain of dyspareunia or postcoital aching. Pelvic congestion syndrome as a cause for CPP has been a controversial entity since it was first described by Taylor in 1949. One reason for skepticism is the observation that some women who demonstrate dilated vessels at the time of surgery or during pregnancy are asymptomatic. Beard and co-workers proposed a more objective method of diagnosing pelvic varicosities using transcervical pelvic venography to measure vessel diameter, vessel tortuosity, and dye transit time, suggesting that vein diameter alone is not the only significant finding of pelvic congestion syndrome. Subsequent studies by these authors propose diagnostic criteria for pelvic congestion syndrome and pathophysiologic mechanisms for pain production. It has been postulated that pain from vascular congestion is caused by vasoactive nociceptive peptides, such as substance P and calcitonin gene-related peptide. In a study by Reginald and colleagues of patients with venographically diagnosed pelvic congestion syndrome injected with a potent vasoconstrictor (dihydroergotamine), a 35% reduction in vein diameter, decreased dye transit time, and up to 4 days of pain relief was demonstrated when compared with patients injected with placebo. They hypothesized that vasoconstriction allows clearance of nociceptive vasoactive peptides. Conservative treatment of pelvic congestion includes medroxyprogesterone acetate in 30- to 50-mg daily doses. Ultimately, surgery may be necessary. In Beard's series of 36 patients treated with hysterectomy and bilateral salpingo-oophorectomy, 67% obtained complete relief of pelvic pain. Only 1 of the remaining 12 patients had significant pain. Irritable Bowel Syndrome Irritable bowel syndrome (IBS) is a common gastrointestinal disorder affecting approximately 15% of adults in the United States, yet only 25% of persons with questionnaire-detected IBS actually seek health care. It has been defined as lower abdominal pain, which is present for at least 12 weeks (not necessarily consecutive) out of the past year, that cannot be explained by structural or biochemical abnormalities. At least two of the following clinical features are needed for diagnosis: (a) pain relief with defecation, and onset associated with a change in (b) frequency or (c) consistency of stools. Theories to explain a putative mechanism for IBS include visceral hyperalgesia, infection, an imbalance of neurotransmitters, and psychological factors. More recently, growing evidence suggests that these patients have a gastrointestinal sensory-reflex dysfunction as a common pathophysiologic mechanism, which may be manifested by different forms depending upon specific pathways involved. This may explain why some individuals are more pain prone, while others may be more prone to have diarrhea or constipation. Rectal and sigmoid balloon distension studies have shown a lower threshold for pain, referred to as visceral hypersensitivity, in patients with IBS compared with controls. However, when balloon testing is performed, only 40% to 60% of patients with IBS report pain at levels of distension below the range of normal values; therefore, gut wall distension my not be the only stimulus to cause IBS. Therapy for IBS involves treating symptoms, whether one is diarrhea prone or constipation prone. Diarrhea-prone patients respond to loperamide, hyoscyamine, and desipramine, whereas constipation-prone patients respond to fiber therapy or cisapride. Patients with a pain predominance may be treated with tricyclic antidepressants, NSAIDs, anticholinergic agents, calcium channel blockers and, in some cases, opioid analgesics. Newer agents are under development, such as visceral antinociceptive agents, ?-opioid agonists, and 5HT3 and 5HT4 receptor antagonists. Patients who have rectal bleeding, persistent occult fecal blood, or failure of medical therapy should be referred to a gastroenterologist to rule out malignancy or inflammatory bowel disease. Ovarian Remnant Syndrome Ovarian remnant syndrome may result from incomplete removal of ovarian tissue at the time of oophorectomy and can be associated with CPP in premenopausal women. Patients with this syndrome usually have a history of extensive endometriosis or pelvic inflammatory processes, resulting in a technically difficult oophorectomy. Cyclic pelvic pain caused by ovarian remnant syndrome most commonly is associated with the development of ovarian follicles within hormonally active ovarian tissue. This diagnosis is suspected when serum levels of follicle-stimulating hormone and luteinizing hormone are normal. Hormone replacement therapy should be discontinued at least 10 days before assessing serum gonadotropin levels. Clomiphene citrate and GnRH agonists have been used to stimulate ovarian tissue, which assists in making an ultrasonographic diagnosis of ovarian remnant syndrome. Surgical treatment usually requires extensive intraperitoneal adhesiolysis and retroperitoneal dissection to remove all ovarian tissue. In a series of eight patients treated surgically for ovarian remnant syndrome, three required large bowel resection, cystotomy was performed in three and a ureteroneocystostomy in one, and one required a small bowel resection. Outcome data regarding postoperative pain relief after surgical resection are limited to small case series, but cure rates as high as 90% are reported. Residual Ovary Syndrome Residual ovary syndrome originally was described by Grogen in 1958. It is characterized by the development of pain in one or both ovaries conserved at the time of hysterectomy, theoretically caused by perioophoritis with a thickened ovarian capsule. It has been postulated that pain is produced by the cyclic expansion of the ovary encased in adhesions. The most common complaint in women with residual ovary syndrome is chronic lower abdominal pain, dyspareunia, and radiation of pain to the back or anterior thigh. A tender mass may be palpated on bimanual examination. This syndrome may occur in as many as 3% of women who have undergone hysterectomy with ovarian conservation. Treatment typically has been oophorectomy rather than lysis of adhesions, to avoid recurrent adhesion formation. Care should be taken to avoid ureteral damage, because retained ovaries often are near the ureter. Outcome data are limited regarding pain relief after surgery for retained ovaries (level 3 data). Postoperative pain relief has been reported in over 80% of patients after oophorectomy. Pain of Uterine Origin In the United States, approximately 18% of hysterectomies are performed for CPP, with endometriosis cited as the most common indication. Carlson and co-workers demonstrated significant reduction in pain following hysterectomy for CPP of all causes. In patients with pelvic pain who underwent hysterectomy, 85% complained of frequently occurring pain preoperatively, compared with 13% at 12 months postoperatively. In a prospective cohort study comparing surgical to nonsurgical treatment, 49% of medically managed patients with CPP had continued symptoms compared with 3% treated by hysterectomy. Moreover, 25% of patients in the nonsurgical group underwent hysterectomy within 1 year. Stovall and others reported a 78% success rate in patients undergoing hysterectomy for pelvic pain of presumed uterine origin. In this retrospective study, patients were excluded for nonuterine pain such as endometriosis. A prospective multicenter cohort of 308 women undergoing hysterectomy for CPP reported resolution of pain in 74% and improvement in 21%. Risk factors for continued pain included age 30 or younger, history of pelvic inflammatory disease, and use of public assistance. These studies are helpful because they underscore the importance of investigating other, nongynecologic causes for CPP, and they provide guidelines for preoperative counseling. Hysterectomy may be indicated in the absence of significant pathology, as long as a diagnostic laparoscopy has been performed, and the patient's pain has persisted for longer than 6 months with a serious negative impact on the quality of life. Prior to undergoing a hysterectomy the patient should be treated medically (oral contraceptives, NSAIDs, and induced amenorrhea) and should be evaluated for urinary, gastrointestinal, musculoskeletal, and psychological causes for pain. Possible causes of pain of uterine origin are listed in Table 43.1.

TABLE 43.1. Possible causes of pain of uterine origin

PSYCHOLOGICAL ASPECTS OF CHRONIC PELVIC PAIN Patients sometimes get the message “your pain is all in your head.” Whether patients are actually told this or whether they hear this is irrelevant. This concept reflects what physicians were taught historically as the somatic model of pain, wherein there was a direct relationship between tissue damage and pain intensity. This model does not account adequately for the discrepancy between objective physical findings and the perception of pain severity. A more contemporary approach recognizes the central models of pain, which include the effect of peripheral nociception and central pathways that determine disability and distress. The psychological aspects of CPP can be significant. It is, therefore, worthwhile to distinguish among co-morbid states such as depression, anxiety, and panic attacks. It often is helpful to obtain consultation from an interested psychiatric health care professional. However, some screening tests can be done by primary care physicians, such as the Beck Depression Inventory or Zung Depression Index. Clinical depression associated with CPP usually is manifested by sleep disturbance (insomnia or hypersomnia), loss of interest in pleasurable activities, guilt feelings, loss of energy, diminished concentration, appetite changes (decreased or increased), psychomotor changes, and possible suicidal ideation. The mainstays of therapy include the use of serotonin reuptake inhibitors, tricyclic antidepressants, and psychotherapy. Clinicians who prescribe these medications should be aware of diagnostic criteria for depression and be familiar with the side effects and contraindications of these medications.

PATIENT HISTORY A thorough patient history is one of the most helpful tools in the evaluation of patients with CPP. The COLDERR acronym may be used to gain a general understanding of the patient's history of present illness. Character—What does the pain feel like? (sharp, dull, crampy) Onset—Does the pain come on suddenly or gradually? Is it cyclic or constant? Location—Is the pain localized or diffuse? Duration—How long has the pain been present and how has it changed over time? Exacerbation—What activities or movements make it worse? Relief—What medication, activities, and positions make it better? Radiation—Does the pain radiate anywhere (back, groin, flank)? There are several important aspects of taking a detailed pain history and reviewing organ systems. Comprehensive history and physical examination forms are available through the International Pelvic Pain Society Web site and can be downloaded at http://www.pelvicpain.org/. Establishing whether the pain is cyclic or not is helpful is narrowing the list of possible causes ( Table 43.2). If the patient has poor insight into the nature of her pain, a pain calendar may be used prospectively to chart her symptoms during the month.

TABLE 43.2. Cyclic causes for chronic pelvic pain a

Reviewing past surgeries is helpful for gaining information about which organs can be eliminated from consideration and what effect surgery had on their pain. It may help establish a causal relationship between surgery and pain, which is common among patients with myofascial pain syndrome. A thorough review of systems is important to exclude nongynecologic causes for CPP. Gastrointestinal complaints should be explored with specific questions about bowel frequency, consistency, associated pain, and to exclude a history of blood in the stool, which would alert the clinician to rule out neoplasm ( Table 43.3). Likewise, urinary complaints such as dysuria, hematuria, nocturia, enuresis, and increased frequency should be probed for urologic causes of CPP ( Table 43.4).

TABLE 43.3. Gastrointestinal causes for chronic pelvic pain

TABLE 43.4. Urologic causes for chronic pelvic pain

PHYSICAL EXAMINATION Observing a patient's gait, body language, and facial expression can provide valuable information. Myofascial pain syndrome can result in or be a result of musculoskeletal dysfunction as seen with abnormalities in gait, body posture, leg lengths, or sacroiliac joint mobility. Such abnormalities warrant evaluation by a physiatrist or physical therapist. A one-finger abdominal examination is performed on the anterior abdominal wall to detect trigger points, as described in the myofascial pain section. In addition, abdominal wall hernias may be diagnosed using this technique. Trigger points should be marked on the patient's skin with a soft pen tip and then mapped in the patient's chart for future comparison. Trigger points also may be found within the muscles of the back and buttocks. A cotton-tipped applicator is useful in the examination of the vulvar vestibule to elicit painful areas or sites of vulvar vestibulitis. Special attention should be given to Bartholin ducts and Skene glands. On speculum examination, the vaginal fornices, or vaginal cuff in the posthysterectomy patient, can also be examined for tenderness using a cotton-tipped applicator. Any cervical discharge should be noted, examined by microscopy, and cultured. The posterior vaginal fornix should also be inspected visually to identify endometriosis. This sometimes is possible only with the aid of a tenaculum. It is helpful to perform a “unimanual” examination prior to the bimanual examination to avoid “cross-contamination” from anterior abdominal wall pain signals. This consists of examining the vagina, cervix, and pelvic floor muscles with the hand in the vagina, without using the hand on the abdomen. The pelvic floor muscles should be examined individually to include the levator ani group, coccygeus, obturator internus, and the piriformis muscles. The uterus, tubes, and ovaries should be examined for localized pain. The pelvis should then be examined using the bimanual technique. The rectovaginal examination is particularly useful to evaluate nodular or infiltrating endometriosis. If there is no tenderness in the cul-de-sac or along the uterosacral ligaments, yet endometriosis is strongly suspected, the patient should be reexamined during the menses.

PAIN MANAGEMENT Pain management may include the use of NSAIDs, neurolytic agents, and narcotics. Knowledge of drug interactions, contraindications, and side effects are important aspects of successful pain management.

NSAIDs NSAIDs are usually a safe category to start with, realizing there are several contraindications to these drugs, especially in patients with gastrointestinal disorders. The newer COX-2 (cyclooxygenase inhibitor-2) drugs have the advantage of fewer adverse effects on the gastrointestinal tract ( Table 43.5).

TABLE 43.5. Oral NSAIDs

Neurolytic Agents Neurolytic agents encompass a number of drug categories used in the treatment of neuropathic pain, which include tricyclic antidepressants, serotonin reuptake inhibitors, and ion channel blockers. Table 43.6 lists some of these drugs which have been studied in patients with chronic pain.

TABLE 43.6. Oral Neurolytic Agents

Narcotics Narcotics often are thought of as appropriate for acute pain only; however, in some instances under careful supervision, patients with chronic pain may be candidates for narcotic therapy ( Table 43.7). When narcotics are used, narcotic contracts are often helpful. Patients who demonstrate drug abuse behavior are not candidates for narcotic therapy. A patient who repetitively loses her prescription, uses multiple physicians for obtaining narcotic prescriptions, or routinely runs out of narcotics early is not considered a narcotic candidate. Patients who use narcotics over the long term should regularly report the symptoms on a VAPS and should report their degree of functionality. They should be encouraged regularly to wean from these agents.

TABLE 43.7. Oral narcotics

Narcotic Agreements When long-term narcotic therapy is used for pain management, it is useful to have the patient enter into a pain agreement to prevent misunderstandings and to help both the physician and patient comply with the federal laws regarding controlled substances. When the agreement is entered, it is understood that if the patient terminates the agreement, the physician will discontinue narcotic therapy but will taper the medicine over a period of several days, as necessary, to avoid withdrawal symptoms. Also, a drug-dependence treatment program should be recommended. The patient agrees to describe the pain character and intensity, the effect of the pain on her daily life, and how well the narcotic is helping to relieve the pain. The patient also agrees not to use any illegally obtained controlled substances, and not share, sell, or trade medication with anyone. Opioids, controlled stimulants, or antianxiety medication must not be obtained from any other doctor. The patient must agree to safeguard her pain medicine from loss or theft with knowledge that lost or stolen medicines will not be replaced. She should understand that prescriptions will be made only at the time of an office visit or during regular office hours and that no refills will be available during evenings or on weekends. This is helpful for covering physicians who receive calls from patients asking for narcotics. The pharmacy name, location, and telephone number are recorded and a log of the dosage, quantity, and prescription date is recorded in the patient's chart.

SUMMARY POINTS CPP is a complex entity, often without obvious, visible pathology. A distinction must be made between putative causes of CPP, which will respond to specific treatment, versus CPP associated with incidental pathology. Treatment of CPP with incidental pathology may result in failure and, in some cases, removal of otherwise normal organs. In such cases, treatment may take a more general and empiric course, resulting in the use of neurolytic medications and possibly the judicious use of narcotics. SUGGESTED READINGS Glossary and Contemporary Pain Theories IASP Task Force on Taxonomy. Classification of chronic pain. Seattle: IASP Press, 1994;209–214. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965:150:971.

Putative Pain States Diamond MP, Daniell JF, Feste J, et al. Adhesion reformation and de novo adhesion formation after reproductive pelvic surgery. Fertil Steril 1987:47:864–866. Peters AA, Trimbos-Kemper GC, Admiraal C, et al. A randomized clinical trial on the benefit of adhesiolysis in patients with intraperitoneal adhesions and chronic pelvic pain. Br J Obstet Gynaecol 1992;99:59–62. Rapkin AJ. Adhesions and pelvic pain: a retrospective study. Obstet Gynecol 1986;68:13–15. Steege JF, Stout AL. Resolution of chronic pelvic pain after laparoscopic lysis of adhesions. Am J Obstet Gynecol 1991;165:278–281, discussion 281–283. Stout A, Steege JF, Dodson WC, et al. Relationship of laparoscopic findings to self report of pelvic pain. Am J Obstet Gynecol 1991;164:73–79. Stovall TG, Elder RF, Ling FW. Predictors of pelvic adhesions. J Reprod Med 1989:34:345. Trimbos JB, Trimbos-Kemper GC, Peters AA, et al. Findings in 200 consecutive asymptomatic, healthy women having a sterilization. Arch Gynecol Obstet 1990;247:121–124.

Endometriosis Demco L. Mapping the source and character of pain due to endometriosis by patient-assisted laparoscopy. J Am Assoc Gynecol Laparosc 1998;5:241–245. Hornstein MD, Surrey ES, Weisbery GW, et al. Leuprolide acetate depot and add-back in endometriosis: a 12 month study. Lupron add-back study group. Obstet Gynecol 1998;91:16–24. Jacobson TZ, Barlow DH, Garry R, et al. Laparoscopic surgery for pelvic pain associated with endometriosis (Cochrane Review). Cochrane Database Syst Rev 2001;4:CD001300. Koninckx P, Meuleman C, Demeye S, et al. Suggestive evidence that pelvic endometriosis is a progressive disease, whereas deeply infiltrating endometriosis is associated with pelvic pain. Fertil Steril 1991;55:759–765. Martin D, Hubert G, Vander Zwaag R, et al. Depth of infiltrating endometriosis. J Gynecol Surg 1989;5:55. Selak V, Farquuhar C, Prentice A, et al. Danazol for pelvic pain associated with endometriosis (Cochrane Review). Cochrane Database Syst Rev 2001;4:CD000068. Sutton CJ, Pooley AS, Ewen SP, et al. Follow-up report on a randomized controlled trial of laser laparoscopy in the treatment of pelvic pain associated with minimal to moderate endometriosis. Fertil Steril 1997;68:1070–1074.

Pelvic Inflammatory Disease and Myofascial Pain Slocumb JC. Neurological factors in chronic pelvic pain: trigger points and the abdominal pelvic pain syndrome. Am J Obstet Gynecol 1984;149:536–543. Travell JG, Simons DG. Myofascial pain and dysfunction: the trigger point manual. Baltimore, Williams & Wilkins, 1983.

Pelvic Congestion Syndrome Beard RW, Belsey EM, Lieberman MB, et al. Pelvic pain in women. Am J Obstet Gynecol 1977;128:566–570. Beard RW, Kennedy RG, Gangar KF, et al. Bilateral oophorectomy and hysterectomy in treatment of intractable pelvic pain associated with pelvic congestion. Br J Obstet Gynaecol 1991;98:988–992. Beard RW, Pearse S, Highman JH, et al. Diagnosis of pelvic varicosities in women with chronic pelvic pain. Lancet 1984;8409:946–949. Reginald PW, Beard RW, Kooner JS, et al. Intravenous dihydroergotamine to relieve pelvic congestion with pain in young women. Lancet 1987;8555:351–353. Stones RW, Loesch A, Beard RW, et al. Substance P: endothelial localization and pharmacology in the human ovarian vein. Obstet Gynecol 1995;85:273–278. Stones RW, Rae T, Rogers V, et al. Pelvic congestion in women: evaluation with transvaginal ultrasound and observations of venous pharmacology. Br J Radiol 1990;63:710–711. Stones RW, Thomas DC, Beard RW. Suprasensitivity to calcitonin gene-related peptide but not vasoactive intestinal peptide in women with chronic pelvic pain. Clin Autonomic Res 1992;2:343–348. Taylor HC. Vascular congestion and hyperemia. Am J Obstet Gynecol 1949;57:211–230.

OVARIAN REMNANT SYNDROME Grogen RH. Residual ovaries. Obstet Gynecol 1958;12:329–332. Lafferty HW, Angioli R, Rudolph J, et al. Ovarian remnant syndrome: experience at Jackson Memorial Hospital. Am J Obstet Gynecol 1996;174:641–645. Scott RT, Beatse SN, Illions EH, et al. Use of the GnRH agonist stimulation test in the diagnosis of ovarian remnant syndrome. J Reprod Med 1995;40:143–146. Siddall-Allum J, Rae T, Rogers V, et al. Chronic pelvic pain caused by residual ovaries and ovarian remnants. Br J Obstet Gynaecol 1994;101:979–985.

Pain of Uterine Origin Carlson KJ, Miller BA, Fowler FJ. The Maine women's health study: I. Outcomes of hysterectomy. Obstet Gynecol 1994;83:556–565. Carlson KJ, Miller BA, Fowler FJ. The Maine women's health study: II. Outcomes of nonsurgical management of leiomyomas, abnormal bleeding, and chronic pelvic pain. Obstet Gynecol 1994;83:566–572. Hillis SD, Marchbanks PA, Peterson HB. The effectiveness of hysterectomy for chronic pelvic pain. Obstet Gynecol 1995;86:941–945. Stovall TG, Ling FW, Crawford DA. Hysterectomy for chronic pelvic pain of presumed uterine etiology. Obstet Gynecol 1990;75:676–679. Wilcox LS, Koonin LM, Pokras R, et al. Hysterectomy in the United States, 1988–1990. Obstet Gynecol 1994;83;549–555.

Chapter 44 Perioperative Evaluation Danforth’s Obstetrics and Gynecology

Chapter 44 F. Joseph Kelly, Phillip Y. Roland, and James W. Orr, Jr.

Perioperative Evaluation

INTRODUCTION SURGICAL INDICATIONS AND CONSENT FOR SURGERY HISTORY AND PHYSICAL EXAMINATION LABORATORY INVESTIGATION RADIOLOGIC INVESTIGATION ENDOSCOPIC EVALUATION CARDIAC DISEASE General Considerations Cardiac Disease–Specific Approach Risk Stratification According to Type of Surgery Risk Assessment and Treatment PERIOPERATIVE PULMONARY EVALUATION Patient-related Risk Factors Procedure-related Pulmonary Risk Factors Preoperative Evaluation Risk Reduction Strategies RENAL DISEASE Diagnosis, Morbidity, and Mortality Management WOUNDS AND INCISIONS Wound Preparation Surgical Technique GASTROINTESTINAL CARE Early Oral Feeding and Nasogastric Suction Liver Disease INFECTION Treatment for Existing Infection THROMBOEMBOLISM NEUROLOGIC COMPLICATIONS Operative Neurologic Deficits SUMMARY POINTS SUGGESTED READINGS General Considerations Surgical Indications and Consent Preoperative Evaluation and Testing Perioperative Pulmonary Evaluation Renal Disease Wounds and Incisions Gastrointestinal Care Infection Thromboembolism Neurologic

INTRODUCTION “One important key to success is self-confidence. An important key to self-confidence is preparation.” Arthur Ashe In this country, obstetrician-gynecologists are the primary surgeons in over 3.6 million pelvic operative procedures completed on an annual basis ( Table 44.1). These procedural volumes, when combined with the many thousand minor office and obstetric procedures, are testimony to the enormous potential physical, psychological, and economic impact of surgery and potentially preventable adverse events on women's health care. Importantly, nearly 50% of all adverse perioperative events are preventable. The sheer numbers undeniably suggest that risk prevention should be the ultimate goal of every surgeon. Assuming the role of primary care physician and surgical subspecialist for women, obstetrician-gynecologists should accept responsibility for developing, instituting, and completing all aspects of perioperative management. As an important step to improve quality of care, we should be cognizant of the potential role and benefits that allow objective discussion of nonsurgical options, as well as understand appropriate surgical indications for pelvic disease processes.

TABLE 44.1. Surgical procedures in obstetrics and gynecology

Our primary goals should consist of prospective preoperative recognition, evaluation, and management of the significant clinical aspects of existing medical comorbidities and development of operative techniques to form a flexible multifaceted surgical skill set that increases the opportunity for safe completion of the procedure, regardless of inherent technical difficulty. Finally, pelvic surgeons should be capable of devising a postoperative care plan that further minimizes the risk of an adverse perioperative outcome, enhances the early recognition of acute perioperative problems, and lessens the risk and the intensity of potentially catastrophic perioperative complications. This care plan should be formalized prior to entering the preanesthetic area or the operating suite, and its execution should contain enough clinical alternatives to ensure a successful result, regardless of the intraoperative findings or postoperative complications encountered. The volumes of existing published literature on surgical alternatives and indications establish a clear message: “optimal” perioperative care is a moving target. Every treatment plan and intervention requires careful aim if the goals of improved quality and optimal patient outcome are to be attained. Mastery of these perioperative planning processes only serves to improve the individual and collective quality of patient care. Proper implementation should result in decreased individual, regional, and national health care expenditures.

SURGICAL INDICATIONS AND CONSENT FOR SURGERY It is the ultimate responsibility of every obstetrician-gynecologist to evaluate thoroughly the patient's reproductive tract complaints prior to entertaining the possibility of surgical management. In addition to evaluating the pelvic condition, it is vital to conduct a preoperative search for physical or psychological comorbidities. Only when armed with this global information can the surgeon objectively present the risks and realistically relate the expected outcome of the entire spectrum of accepted medical and surgical options. This information should be processed and presented to the patient (and other support persons when appropriate) in a manner that is not demeaning, using understandable language that allows and facilitates an informed decision and consent process. Although the preoperative diagnostic period may be deemed routine by the surgeon and staff, many women consider the prediagnostic interval to be the most stressful time of treatment. This psychological stress contributes to difficulty in understanding the importance and results of diagnostic studies or results and potentially clouds the information discussed during preoperative discussions. Decisions, options, and alternatives, as they relate to the potential need for postoperative therapy, can be lost. The preoperative use of illustrative drawings, pamphlets, videos, or other visual educational materials often can be of assistance in improving patient (and family) comprehension of medical alternatives, operative indications, associated risks, and expected treatment outcomes. Typically, this preoperative encounter (or encounters) to discuss surgical options and obtain consent is best undertaken in a private setting, completed in a manner that is conducive of a two-way flow of communication. Eye contact, a caring touch, an unhurried approach, and other “soft” physician behaviors lower communication barriers and facilitate patient and “significant other” comprehension. The complexity of preoperative discussion and necessary time expenditure may vary dramatically with the surgical indications, as well as with the medical alternatives to the proposed

procedure. Regardless, the algorithm of care, surgical risks and outcome expectations should be detailed, recognizing that modification may be required secondary to important or imperative patient desires (e.g., a patient desiring retention of ovaries or fertility). Regardless of the extent of preparation and best surgical technique, any unexpected or adverse intraoperative or postoperative event can alter dramatically the final surgical procedure and ultimate result. The apparently “simple” laparoscopic salpingo-oophorectomy may be complicated by uncontrolled abdominal wall, intraperitoneal, or retroperitoneal bleeding, intestinal injury, or the finding of an ovarian malignancy. Postoperative myocardial infarction (MI), infection, or thromboembolism may occur. A good rule of thumb is to discuss during the explanation of surgical risks adverse events that occur at a frequency of 1% or more. Any preoperative consent discussion should include the risk of death or permanent disability. Failure to address any or all of these risks can result in misunderstanding, fear, and anxiety and fosters mistrust. As important as the consent process, the resultant preoperative patient education decreases the need for postoperative analgesia and is associated with less anxiety and earlier ambulation. Unless the clinical situation is emergent or life-threatening, these discussions are best undertaken at a time remote from the day of surgery. Timely entered, legible chart documentation should follow any and all of these discussions. Following the mutual decision to bypass medical management, the surgeon is faced with a number of other important procedural questions including the route (i.e., vaginal versus abdominal versus laparoscopic approach), as well as who should perform the procedure. It is intuitive and reported data confirm that, even among subspecialists, individual surgical skills and outcomes of specific procedures differ dramatically. These differences may relate to the surgeon's innate technical ability, the specifics of the individual's previous training, patient selection, or factors unrelated to the physician. Without placing blame, the training demands of our subspecialty have altered the overall extent of residency surgical experience, resulting in fewer major operative procedures being completed per trainee. These educational changes have resulted in a noticeable decrease in resident surgical skill acquisition. Importantly, these relative deficiencies noted by senior observers may not be perceived or recognized by the individual trainees. Coupled with the increasing ratio of postresidency physicians to annual operative procedures and the increasing technical procedural subspecialization, it is easily understood how difficult it is for the pelvic surgeon to obtain, maintain, or refine new surgical skills, even in our restricted subspecialty where the surgical focus is entirely on reproductive tract procedures. These factors lend credence to the adage, “no one can be all things to all people,” and each physician must individualize the decision to operate, consult, or refer. Specific referral guidelines citing evidence-based data have been created to assist practicing obstetrician-gynecologists in the care of women with selective diagnoses. Additionally, a favorable outcome of many procedures is related to increasing surgical experience, surgical volume, and expertise. Although disease-specific guidelines may not relate to “routine procedures,” it is logical and scientific that improved perioperative outcomes should follow increased surgical case volume. Although these effects may be surgeon or program related, arguments against consultation or referral have little scientific merit and may not be in the patient's best interest. Although circumstances may prohibit actual patient referral, the ready availability of curbside, personal, telephone, and internet consultation should encourage information dissemination, and utilization of available resources should improve patient care.

HISTORY AND PHYSICAL EXAMINATION “Listen to women and they will tell you what's wrong with them.” Charles E. Flowers Every operative procedure requiring anesthesia should be viewed as a physiologic stress test. The intensity of the stress response varies in direct proportion to the extent of the actual surgical procedure and is associated with a clinically and biochemically measurable adverse effect. Nearly every organ system is affected adversely by general anesthesia. The associated 20% reduction in resting heat production and increased surface heat loss predispose patients to hypothermia, increasing the risk of an adverse cardiac event, altered pulmonary response to hypercarbia and hypoxemia, impaired coagulation, and poor wound healing. The untoward anesthesia-related pulmonary effects of impaired oxygenation and altered lung mechanics are likely well tolerated by the 40 year old, but they may have catastrophic consequences in the elderly or in those with coexisting pulmonary disease. Most inhalational drugs directly create increased cardiac risk by altering myocardial oxygen supply and demand kinetics. Myocardial depression, increased arrhythmogenicity (as high as 27% incidence), and altered neural tone form the triad for predisposing women to adverse cardiovascular effects. Importantly, these risks are not lessened with regional anesthesia use. Reduction in renal blood supply (decreased by 30 to 70%), suppression of the immune system, ileus, and stress-related gastric ulceration represent but a few of the other quantitative adverse effects of anesthesia. Many of these effects can be correlated with the duration of operation, and every effort should be made to increase operative efficiency in an attempt to safely shorten the procedure and minimize risks. Despite these anesthesia-enhanced risks, only a small percentage of operative deaths are attributable solely to anesthesia. Less than 18% of surgical mortality is directly attributable to the surgical procedure. The vast majority of surgical mortality (79%) can be related directly to problems created in great part from the patient's coexisting medical disease. This mortality risk further illustrates the need to undertake a diligent preoperative search to identify and potentially modify the adverse effects of any significant coexisting medical condition. The literature is inundated with articles espousing the proposed benefits of preoperative radiologic testing and laboratory evaluation; however, the diagnostic power contained in a carefully obtained history and a thorough physical examination is enormous. The reproducible validity of this simple, easily obtainable evaluation is underscored when recognizing that as many as 98% of abnormalities found with preoperative laboratory and radiologic screening can be predicted by historical or physical findings. Although studies relating to the pelvic process (i.e., urodynamics) frequently are necessary to delineate the need for surgery, attempts to obtain important historical information regarding diagnosed medical comorbidities and clinically silent medical disease are vital to direct additional preoperative evaluation. Review of pertinent medical records combined with initial direct questioning about previous hospitalizations, current treating physicians, concurrent diagnoses, use of prescribed and over-the-counter medications, and allergies will assist in the detection of coexisting medical disease. This information may be vital to the preoperative plan regarding the continuation (e.g., cardiac, hypertensive) or discontinuation (e.g., aspirin, oral contraceptives, anticoagulant) of medications. Over-the-counter and herbal medication usage can create the potential for additional untoward complications and should be evaluated ( Table 44.2). Conducting a pointed review of systems allows the determination of performance status and aids in the diagnosis of significant but previously undetected medical disease. Although intended to cover multiple systems, simple questions as to the patient's ability to walk a mile, climb two flights of stairs, or blow out a match from 12 inches away may suffice as a screen for significant cardiopulmonary disease. Historical information obtained by questionnaire can be considered valid; however, the presence or absence of important symptoms should be confirmed during the office visit.

TABLE 44.2. Herbal medicines: possible adverse effects

A thorough multisystem physical examination is an essential part of every preoperative evaluation. Specific attention to the cardiopulmonary system assists in the detection of important coexisting disease that may alter outcome adversely and directs additional investigation. Every system carries some import, and abnormal findings allow for appropriate morbidity-reducing, cost-effective adjustments in the perioperative plan.

LABORATORY INVESTIGATION The primary goal of preoperative laboratory preparation is to obtain results that allow reduction of those inherent risks related to the proposed procedural component, as well as those risks associated with occult or recognized coexisting medical morbidities. It has become apparent that a “broadly cast net” preoperative laboratory screening strategy confers little patient benefit for the following reasons. (a) The majority of laboratory abnormalities can be predicted by findings noted in the history and physical examination. It is rare to detect an unexpected abnormality. (b) The physiologic, psychological, and economic costs associated with the evaluation of abnormal laboratory (including false-positive) studies brings little value, and it rarely influences clinical care. (c) As many as 60% of abnormal preoperative test results are not known or evaluated preoperatively, creating potential liability. Multiple reports suggest that nearly 70% of ordered preoperative laboratory tests are not indicated by facts obtained in the history and physical examination. This information reiterates the importance of the history and physical examination and suggests that a directed preoperative laboratory testing strategy ( Table 44.3) for routine procedures is both safe and cost effective. Obviously, special studies such as tumor markers (e.g., CA-125) are appropriate during evaluation and management

of women with pelvic malignancies, because they offer diagnostic assistance, facilitate decisions regarding patient triage, are potentially prognostic, and are of significant value during postoperative management. Although care should be individualized, it has become apparent that directed preoperative laboratory testing forms a firm foundation for quality preoperative care for patients undergoing elective procedures.

TABLE 44.3. Preoperative testing strategy

Although advances in blood banking technology have lessened the risk of transfusion and favorably affected the outcomes of many surgical procedures, it is apparent that women do not need to have a type and crossmatch performed prior to the majority of obstetric or gynecologic surgical procedures. Although maximal surgical blood order schedules have been established, they should be validated at each institution. When deemed necessary, a type and screen allows for identification of specific antibodies and assures rapid availability (=20 min) of red blood cell products. Individual decisions regarding blood bank strategies should be related to the patient's preoperative status (i.e., hemoglobin and hematocrit, blood volume), anticipated losses, and existing comorbidities that might carry an early transfusion trigger. Anemia is not uncommon among women undergoing pelvic surgery. Although it may constitute an indication (menorrhagia, associated with leiomyomas) or result from an indication (cervical cancer), its presence should prompt an evaluation. As deemed appropriate, a preoperative search for other causes or losses should be undertaken. Additionally, preoperative discussions regarding procedure-associated blood loss and its attendant risks represent an important aspect of informed consent. Critically ill patients may benefit from having their serum hemoglobin levels maintained at about 10 g/dL, but data suggest little effect of transfusion on survival in patients whose hemoglobin levels are between 8 and 10 g/dL. The majority of women can safely undergo elective procedures without a type and screen.

RADIOLOGIC INVESTIGATION Routine imaging studies can be obtained safely as directed by findings of the history and physical examination. The indications for a preoperative chest radiograph are relatively straightforward, and abnormalities are associated with an increased risk of perioperative pulmonary complications. Unfortunately, unnecessary chest radiographs and many additional, sometimes unnecessary, undirected diagnostic preoperative imaging studies are obtained frequently. Diagnostic ultrasonographic examination, pelvic and abdominal computed tomography (CT), and magnetic resonance imaging studies are performed in patients with suspected or known gynecologic malignancies. These studies may suggest the benefit of preoperative triage to a subspecialist but rarely influence the surgical approach. Their results should not be considered binding, typically contribute little to clinical care, and often add significantly to health care costs. Although the radiologist's suggestions for additional studies are noted frequently, the astute pelvic surgeon should combine physical findings, patient symptoms, and laboratory results to develop the perioperative care plan, based on necessity and benefit of histologic confirmation of a radiologic abnormality. Although proponents of preoperative “radiologic staging studies” suggest relative accuracy, the pelvic surgeon must recognize the subjective aspects of interpretation. In general, their routine use has not been associated with an alteration in perioperative clinical care. Although findings may alter approach, negative results do not exclude the finding of significant pathology. Not to condone or condemn, the value and use of other noncardiac imaging studies have not been evaluated adequately. However, it appears that in specific situations CT scans have a significant false-negative result rate (e.g., for the evaluation of extraovarian disease). Radiologic investigation adds little to clinical care in patients with endometrial cancer in the absence of physical findings. There is little clinical value for intravenous pyelography (IVP) to evaluate ureteral location or displacement, because it is not a substitute for intraoperative ureteral identification. In the absence of malignancy, hydronephrosis is likely related to displacement and is easily corrected or managed during the operative procedure. The exception is an abnormal finding on a clinically indicated routine cancer screening study (e.g., mammography), in which test results may have a significant impact on clinical care, particularly in those women undergoing “elective pelvic procedures.”

ENDOSCOPIC EVALUATION There is little evidence to suggest the value of multiple endoscopic procedures prior to most routine gynecologic procedures. However, it is certainly appropriate to consider obtaining history-directed or disease-specific (e.g., inflammatory bowel disease, previous colon cancer) or recommended cancer screening endoscopic procedures prior to elective procedures. For example, endoscopy performed prior to rectovaginal or vesicovaginal fistula repair may render significant and important information and cause the physician to alter the surgical approach or initiate referral.

CARDIAC DISEASE Preoperative assessment to detect undiagnosed heart disease and direct appropriate perioperative treatment of women with longstanding or newly diagnosed cardiac disease is of vital importance. The majority of inhalational anesthetics are myocardial depressants, modify neural tone, and are arrhythmogenic, creating risk even in the healthy patient. Nearly 50,000 perioperative MIs occur annually. Approximately 20,000 are fatal, and hundreds of thousands of related serious extracardiac complications occur, resulting in poor outcome. The prevalence of cardiovascular disease increases directly with increasing age, and within the aging population serous cardiac events occur regularly. Although no age group is risk free, gynecologic procedures in the aging population are estimated to result in a 25% increase in the incidence of necessary major intraabdominal procedures, potentially increasing the absolute risk of cardiac morbidity. Although heart disease may not be readily apparent, the answers to carefully framed historical questions can assist in unmasking occult cardiac risk factors. Gynecologists should feel compelled to become familiar with the important aspects of perioperative cardiac care, to improve individual patient outcome as well as the overall health of the entire female population. Proper preoperative cardiac assessment requires a systematic approach, typically undertaken in close coordination with qualified consultants. Attention to history, physical findings, and the effects of comorbid diseases forms the foundation for initial evaluation. These findings assist in the direction of ordering ancillary cardiac studies, which can clarify risk and improve predictive power for perioperative morbidity associated with the proposed gynecologic surgery. They also allow institution of a sound plan for perioperative monitoring and management A disease-specific approach is important, specifically addressing signs, current symptoms, and management of coronary artery disease (CAD), hypertension, heart failure, valvular heart disease, arrhythmias, pacemakers, pulmonary vascular disease, and type and extent of surgery. General Considerations Particular energy should be directed preoperatively to elicit historical or physical evidence of cardiac ischemia. Important elements include angina, a history of MI, past or current signs or symptoms suggesting congestive heart failure, unexplained palpitations, evidence of previous cardiac intervention, and renal impairment. Important aspects of noncardiac diseases that increase the incidence of cardiac risk factors, including diabetes mellitus, hypertension, hypercholesterolemia, family history of CAD, and obesity, should be evaluated. Age greater than 70 years, MI within the preceding 12 months, and evidence of congestive heart failure significantly increase postoperative cardiac complications in patients undergoing gastrointestinal, urologic, and gynecologic surgery. Concurrently, the pelvic surgeon should attempt to determine the patient's “functional capacity.” The simple inability to walk three blocks or climb two flights of stairs portends a poor functional status. Physical examination evaluating the patient's overall status is mandatory. This evaluation, coupled with laboratory and radiologic information and an electrocardiogram (ECG), provides a baseline estimate as to actual perioperative cardiac risk. Numerous schema from Goldman and others intended to quantify cardiac risks have been designed, reported, and verified in an attempt to quantitate perioperative cardiac risk. The American College of Cardiology-American Heart Association (ACC-AHA) published revised practice guidelines based on qualitative analysis. Specific risk categories include ischemic heart disease, congestive heart failure, high-risk surgery, diabetes mellitus, renal insufficiency, and poor functional status ( Table 44.4). They define clinical risk stratification for noncardiac surgical procedures. Noncardiac procedures with high (>5%), intermediate (500 milliosmoles per kilogram), a fractional sodium excretion of less than 1%, and proteinuria. Elevated serum blood urea nitrogen out of proportion to creatinine (greater than 20:1 ratio) is strongly indicative of prerenal azotemia. Renal azotemia often is associated with minimal proteinuria; however, urine osmolality will be less than 350 milliosmoles per kilogram and fractional excretion of sodium will be greater than 1. The surgeon's degree of suspicion based on operative findings and events is an important aspect of assessment to exclude postrenal azotemia, particularly when the procedure required a difficult pelvic dissection, repair of a prolapse, or treatment of a malignancy. Renal ultrasonography can determine quickly the extent of unilateral or bilateral hydronephrosis and eliminate the possibility of clinically significant ureteral ligation or kinking. If an ultrasonograph is not diagnostic, an IVP may answer important questions, although the administration of contract may worsen the clinical condition of a patient with intrinsic renal insufficiency. Nonoliguric and oliguric renal insufficiency can occur in the postoperative patient, and physicians should not assume that “adequate” urine production indicates intact renal function. Importantly, oliguric renal failure portends a worse prognosis than nonoliguric renal failure. Using low-dose dopamine to increase renal perfusion (and convert an oliguric problem to a nonoliguric situation) has no proven efficacy and, if unsuccessful, should be discontinued, because it creates a propensity for cardiac

arrhythmias and alters blood supply to other vital organs. The morbidity and mortality (fivefold) of renal failure in the postoperative patient is higher than that of individuals treated medically. These adverse outcomes probably are related to many factors, from perioperative ischemic insults to the volume depletion and other changes associated with contrast used for preoperative testing (i.e., IVP, CT) or mechanical bowel preparation. Common predisposing comorbid illnesses, including diabetes mellitus, hypertension, and cardiopulmonary insufficiency, carry their own inherent risks for causing renal insufficiency. Medications such as ace inhibitors and nonsteroidal antiinflammatory drugs, used frequently in the perioperative patient, also increase risk. Methods to promote perioperative renal protection include practicing good surgical technique, limiting the use of nephrotoxic drugs, maximizing cardiopulmonary function by paying attention to intraoperative blood and volume losses, and replacing intravascular losses. Although invasive monitoring to guide fluid replacement may be helpful, its use and the administration of pharmacologic agents (dopamine, mannitol, furosemide) or other interventions has not been conclusively proven beneficial. Hopefully, continued research on adenosine and prostaglandin administration will prove fruitful. Women with chronic renal insufficiency are at risk for acute renal insufficiency during the perioperative period. Preoperative internal medicine or nephrology consultation should be considered in an attempt to optimize volume, use dialysis strategically, and manage coexisting morbidities. Management Dialysis is the cornerstone of perioperative management of renal insufficiency once postrenal obstructions are relieved. Specialists should be consulted to treat acidemia, recalcitrant hyperkalemia, symptomatic volume overload, or impending cardiovascular failure. All medication dosages and schedules should be reviewed regularly to maximize efficacy and minimize toxicity. In conclusion, although perioperative renal failure represents a significant danger to the operative patient, insufficient scientific data exist to identify patients at risk and to develop guidelines for postoperative renal surveillance and management. The surgical community awaits a more specific definition of renal failure, additional research regarding renal protection strategies, and better, less invasive treatments for patients with perioperative renal insufficiency.

WOUNDS AND INCISIONS Once surgery has been deemed appropriate therapy, selecting the operative approach is possibly the most important initial surgical decision. Choice of approach is related primarily to physician bias and comfort, related to previous teaching and experience. However, surgical indications, disease process, previous abdominal incision (and resultant adhesions), patient preference, and existing medical comorbidity should be considered carefully prior to solving this sometimes complex problem. Adequate intraoperative exposure must be a primary consideration. Although a vertical lower midline incision should be the first consideration, some, particularly those with a large body mass index, may be best managed with an upper abdominal incision, avoiding trauma to the infection-prone panniculus. A panniculectomy may be necessary or appropriate. It is evident that many, if not most, gynecologic (even oncologic) and obstetric procedures can be completed safely through a more cosmetic transverse lower abdominal approach. The transverse approach may offer the additional advantage of less pain and diminished postoperative pulmonary dysfunction. Although the Pfannenstiel incision, when combined with a table-stabilized self-retaining retractor, will often suffice, preoperative consideration of a Maylard (rectus muscle splitting) incision or intraoperative conversion to a Cherney (rectus splitting–incision at pubic insertion) may be necessary. Regardless of the initial approach, to avoid hernia risks the incision should not be placed perpendicular to a previous incision. Placing the long axis of the incision in the direction of maximal skin tension creates the most aesthetically pleasing scar. A more contemporary approach to numerous gynecologic procedures involves laparoscopy. Many, if not most, routine procedures and a significant proportion of complex procedures used for the management of benign or malignant disease can be safely completed laparoscopically, with less physiologic stress, less pain, and a more rapid return to normal activity. Although there are many correct answers to surgical incision placement, performing an incomplete procedure because of a compromised incision is an unacceptable alternative and should be avoided by immediate conversion to a more accommodating approach (e.g., hand-assisted laparoscopy or an open operation). Operative laparoscopy and abdominal approach are associated with additional inherent surgical risks, so they should not replace vaginal surgery. Although laparoscopy may be complementary to “convert” abdominal procedures and to complete (extensive) operations, stand-alone vaginal surgery should remain an important arm of the obstetrician-gynecologist's surgical repertoire, because it results in less morbidity. Vaginal surgery will suffice for many, if not most, gynecologic procedures. Although lesser invasive (laparoscopic or vaginal) surgical approaches are associated with diminished physiologic stress, their use does not negate the need for appropriate preoperative evaluation, preparation, and management. Wound Preparation Although many rituals of preoperative skin preparation exist, only a few add documented value to patient care. Preoperative hair removal is not deemed necessary; however; if performed, clipper removal at a time close to surgery avoids microinjury and is associated with a lower risk of wound infection than is razor preparation. Routine preoperative skin antisepsis is necessary to minimize the infectious potential of normal flora, as well as of pathogens. Bacterial concentrations in moist body areas (i.e., perineum) reach 10 6 /cm 2 of tissue and are greater than those in drier (abdomen) areas, reaching 10 3 /cm 2 of tissue. These areas also harbor different ratios of aerobic to anaerobic bacteria. Preparation solutions differ significantly in the immediate or late (>3 hours) mean bacterial reduction. Regardless of solution used, the preparation should be allowed to “set” for no less than 5 minutes to attain its maximal activity and bacterial reduction. Surgical Technique Creating an abdominal incision with a single bold knife stroke through the skin and subcutaneous tissue avoids a stairstep effect in the subcutaneous tissues. The use of multiple knife blades to create the incision offers little benefit and contributes to expense. Avoiding the creation of subcutaneous dead space is important because dead space increases the risk of wound infection and poor wound outcome, even when closed. The use of electrical or laser coagulation techniques to create an incision saves little time and adds little value. Their use creates “devitalized” tissue, which may contribute to an adverse wound environment and increase the risk of surgical site infection, resulting in acute or late poor wound outcome. Surgical technique may be of specific impact in those already at high risk because of thick (=3.0 cm) subcutaneous tissues. The rationale for proper suture selection rests on the need for tensile strength, the duration of retained tensile strength, and the biologic effects on the involved tissue ( Table 44.9). In general, the suture's chemical composition is more important than the physical configuration, although braided and multifilament sutures may potentiate infection. When compared with other absorbable materials, polyglycolic acid suture is associated with less inflammation and decreased pain, and in animal models it lessens the risk of infection. The incorporation of permanent suture material creates an increased risk of (˜10%) chronic wound problems (i.e., pain, wound sinus); however, their use is associated with a lessened risk of developing a fascial hernia. Vascular pedicle ligation can be accomplished safely using short-term absorbable sutures, because the risk of bleeding (1.5-mm vessels) is minimal 96 hours after ligation. There is little clinical benefit to be gained from peritoneal closure.

TABLE 44.9. Suture material

Suture size selection should be based on specific needs. Excluding fascial closure, there is little rationale for using suture larger than 2-0. Additional wound suture material increases the inflammatory response and may predispose to infection, and the tensile strength of a 2-0 absorbable suture is adequate for nearly all pedicle ligations. Although it seems basic, correct suture-tying techniques can contribute to surgical outcome. Complex knots (e.g., surgeon's) impart greater tensile strength than multiple simple square knots. Avoiding excess knots minimizes the amount of suture material, lessens inflammatory volume, and potentially decreases the risk of poor

wound outcome. Although all knots should be secure, tightly tied fascial sutures strangulate the incorporated tissues, increase ischemia, and lessen wound strength. Although good surgical technique always has been tied to attempts to prevent adhesion formation, the development of hyaluronic acid products for adhesion prevention has opened new avenues. The astute surgeon recognizes the different tissue-related healing curves and selects a suture designed to promote good outcome. Although short-term wound results are important, the risk of late problems (e.g., hernia) can be minimized with appropriate closure technique. A suture length–to–wound length ratio of greater than 4 decreases the risk of hernia formation and can be accomplished by placing sutures at least 1.5 cm from the fascial edge and 1.5 cm apart. Subcutaneous tissue closure is typically unnecessary. If required to lessen tension on wound edges, small minimally reactive sutures should be chosen. Closed suction and subcutaneous drain placement are of little apparent benefit, and drains increase the risk of surgical site infection. The best method or material for epithelial approximation should be related to the potential time, cost, need for return visits, and cosmetic results.

GASTROINTESTINAL CARE Perioperative management of the gastrointestinal tract during pelvic surgery focuses on the following three aspects: (a) avoidance of operative injury and reduction of subsequent complications, (b) the role of early feeding, and (c) the appropriate diagnosis and management of patients with liver disease. Excluding the need to address important aspects of coexisting intestinal diseases (e.g., inflammatory bowel disease) and issues related to cancer screening, gastrointestinal preparation values the decisions related to the incorporation of bowel preparation. Mechanical intestinal preparation offers a number of potential benefits ( Table 44.10). Although suboptimal preparation may increase the risk of contamination, most pelvic surgeons are inclined to use a mechanical bowel preparation when the risk of manipulation or injury is substantial. Several regimens are used. Fleets Phospho-Soda (90 mL) is cost effective, well tolerated, and gives excellent results.

TABLE 44.10. Mechanical bowel preparation potential benefits

Intestinal bacterial flora colony counts are high in the colon and distal ileum, decreasing dramatically in the more proximal small intestine unless obstructed. Classically the mechanical regimen is combined with oral antibiotics, typically including erythromycin or neomycin or both, to reduce luminal bacterial content. Although debate persists, the benefit of oral antibiotics over intravenous antibiotics is apparently small, and oral use is associated with significant gastrointestinal symptoms. Early Oral Feeding and Nasogastric Suction Nasogastric suction often has been incorporated into the postoperative care of women undergoing difficult pelvic operations. Reports not only question its use but deny its benefit as it relates to symptomatic relief, lessened risk of ileus, need for reinsertion, and protection of intestinal anastomosis. Regardless of procedure, fewer than 10% of patients require nasogastric suction, and it appears to be of little benefit to the other 90%. As importantly, early oral feeding is well tolerated, does not increase the risk of ileus or symptoms, and may shorten hospital stay. Liver Disease The preoperative detection of liver disease may not be easy, because serum enzymes (transaminase) may not be elevated in patients with late-stage disease. However, the historical and physical findings of significant alcohol intake or exposure to hepatotoxins should raise suspicion. Physical evidence of malnutrition, signs of hepatic encephalopathy, or jaundice usually signify advanced disease. Laboratory evidence of hepatic dysfunction includes an elevated (10-fold) increase in serum transaminase levels, abnormal coagulation study results, elevated total bilirubin values, and low serum albumin levels. Classic surgical risk indices have been based on the Child classification, initially detailed in the 1960s to determine risks associated with surgical management of those with esophageal varices. These indices have been applied to the risks in those with liver disease who are undergoing abdominal surgery. Perioperative management of women with hepatic disease requires particular attention to drugs (and to their dosing and scheduling) which are metabolized by the liver. For example, patients receiving benzodiazepines and narcotics are at risk for drug accumulation and overdosage due to poor hepatic clearance. Patients with acute liver disease (vital or other) should not undergo elective surgery until stabilization of liver dysfunction has been secured. Those with chronic liver disease and hepatic encephalopathy in an attempt to stabilize liver function may benefit from subspecialty consultation.

INFECTION Perioperative surgical site infection occurs in the practice of every gynecologic surgeon. Knowledge of prophylactic measures and the optimal treatment regimen for those with active or acquired infection are expected of every pelvic surgeon. Numerous factors including obesity, diabetes, use of steroids, and even length of preoperative hospitalization are all risk factors for infectious perioperative morbidity ( Table 44.11). Although addressing each of them is beyond the scope of this chapter, numerous technical methods can modulate these risks. Proper use of prophylactic antibiotics lessens infectious risks in many pelvic procedures. Numerous antibiotics and regimens are efficacious. In most situations a single dose, administered properly, is as effective as most multiple regimens. However, in longer procedures (i.e., =3 hours) when the duration of the procedure is longer than the half-life of the chosen drug or if blood loss exceeds 1,500 milliliters, repeat dosing improves its benefit. The use of antibiotics prophylactically can result in patient hypersensitivity and the development of resistant strains of bacteria. Although careful thought is never to be discouraged, the potential benefit of controlling operative site infection with antibiotic prophylaxis usually outweighs other patient risks. The authors use a second-generation cephalosporin, which has better anaerobic coverage than first-generation cephalosporins and minimizes the loss of Gram-positive coverage that typically occurs with third-generation cephalosporins, recognizing the lack of enterococcus coverage. Patients allergic to penicillin have an 8% risk of cross-reactivity with cephalosporins, but anaphylaxis risk is estimated at 0.0001% to 0.1% in non–penicillin-allergic patients to 0.02% in penicillin-allergic patients. Doxycycline or metronidazole represents an excellent alternative in patients undergoing abdominal or vaginal hysterectomy. Although the benefit of antibiotic prophylaxis has not been demonstrated in patients undergoing laparoscopy-assisted hysterectomy, it would seem prudent to use these drugs. The recommendation of prophylaxis for other procedures varies ( Table 44.12). Specific prophylaxis for those women at risk for endocarditis is imperative ( Table 44.13).

TABLE 44.11. Infection risk factors

TABLE 44.12. Antimicrobial prophylactic regimens by procedure

TABLE 44.13. Prophylactic regimens for prevention of endocarditis in susceptible patients undergoing genitourinary or gastrointestinal procedures.

Treatment for Existing Infection Prophylactic drugs are administered to patients in the absence of preoperative evidence of active infection. However, gynecologic patients with an obvious infection prior to surgery should receive an antibiotic treatment regimen appropriate for the type of infection. For example, patients with tuboovarian abscess should be treated with a broad-spectrum antibiotic regimen covering Gram-positive, Gram-negative, and anaerobic bacteria with a single or multiple drug regimen, according to patient acuity. Postoperatively, antibiotics should be given for documented infection. It is clear that most postoperative febrile illness is not related to a documented infection. When postoperative fever, leukocytosis, and clinical picture suggest postoperative infection, an examination to identify potential sources should be performed. Source-directed therapy is appropriate. Adjunctive procedures, including CT or ultrasonographically guided drainage, may hasten recovery in patients with fluid collections or abscesses. In general, patients should receive adequate, broad-spectrum antibiotic treatment prior to surgical intervention.

THROMBOEMBOLISM Thromboembolic phenomena, including deep venous thrombosis (DVT) and pulmonary embolism (PE), are inevitable following gynecologic surgery. A 1988 meta-analysis estimated a 6% to 7% proximal DVT risk and a 0.5% to 0.8% fatal PE risk in unprotected patients over 40 years of age undergoing abdominal surgery. This risk makes surgeons anxious because of the difficulty of diagnosing venous thromboemboli and the potential catastrophic outcome of fatal PE. The impact and importance of postoperative thromboembolic prophylaxis is no longer a question. In a survey by the American College of Surgeons, approximately 96% of surgeons claimed they regularly used antiembolic prophylaxis. However, patient poll and chart review suggest that only one third of patients received adequate prophylaxis. Despite the perception that venous emboli are a rare and unimportant event, all gynecologic patients should be considered at risk. The misconception regarding incidence is fostered by the high occurrence (˜50%) of DVT and PE postdischarge. Additionally, surgeons have been anxious about the potential side effects of prophylaxis, because low-dose unfractionated heparin or low-molecular-weight heparin administration has been associated with postoperative bleeding and wound hematomas. The infrequent occurrence of heparin-induced thrombocytopenia also has created some anxiety regarding prophylaxis. Finally, the initial or primary cost of thromboprophylaxis has deterred some from its use, although some have agreed that prophylaxis is cost effective. Thromboembolic phenomena can be clinically silent, whereas complications related to prophylaxis are not easily missed by the gynecologic surgeon or the patient. Furthermore, according to the Rochester Epidemiology Project and other studies, the incidence of PE and DVT is increasing as women age. Prophylaxis in the gynecologic practice will become more important during the next decade. Despite these common misperceptions, principles of good gynecologic perioperative care demand careful attention to the role and benefit of thromboprophylaxis. The American College of Chest Physicians published guidelines regarding perioperative thromboprophylaxis. Risk factor stratification ( Table 44.14) allows delineation of low risk, moderate risk, high risk, and highest risk categories, with coincident risk of developing DVT, clinical PE, and fatal PE. Proposed recommendations for thromboprophylaxis strategies ranged from aggressive mobilization in the lowest risk patients, to the use of pneumatic compression hose, elastic stockings, or pharmaceutical methods with heparin (unfractionated or low-molecular- weight). Patients at highest risk are deemed candidates for combined mechanical and heparin prophylaxis.

TABLE 44.14. Levels of thromboembolism risk in surgical patients without prophylaxis

A report suggested that the frequency of DVT following gynecologic procedures was approximately 16% (a range of 4%–38%). Fatal PE was reported in only 0.4% of the pooled sample. The cited risks for postoperative thromboembolic event included age greater than 40, history of venous embolism, surgery for cancer, and an abdominal surgical procedure. It was suggested that gynecologic oncology patients were at particularly high risk, fulfilling Virchow's triad of advanced age, cancer, and the hypocoagulable state; venostasis related to pelvic mass compression; vascular injury due to lymph node dissection; postoperative immobility; and the thrombogenic effect of chemotherapy. Importantly, the authors noted a 75% reduction in fatal PE (from 0.4% to 0.1%) with the use of appropriate thromboprophylaxis. The strongest evidence presented for thromboprophylaxis was a relative risk reduction of 64% with use of low-dose unfractionated heparin (reductive from 20% of patients to 7%). A dose response was suggested. Gynecologic cancer patients derived less protection than the noncardiac patients from the twice daily administration of low-dose unfractionated heparin. It would appear that dosing 3 times a day is more appropriate. Although of concern, the potential risk of bleeding complications was not reproducible. Importantly, aspirin use was deemed insufficient for thromboprophylaxis. Low-molecular-weight heparin has been associated with fewer bleeding complications and is being studied in comparison with low-dose unfractionated heparin. However, low-molecular-weight heparin, although extensively studied, did not meet the criteria for study inclusion by the American College of Chest Physicians, so it was not included in their recommendations. In general, patients undergoing brief procedures require no prophylaxis except early mobilization. Patients having major gynecologic procedures for benign disease without additional risk factors should receive low-dose unfractionated heparin twice daily, low-molecular-weight heparin, or perioperative external pneumatic compression hose. For the highest risk patients, dual prophylaxis with either low-dose unfractionated heparin plus elastic stockings or external compression hose, low-dose unfractionated heparin 3 times daily, or low-molecular-weight heparin given in daily doses of at least 3,400 anti-Xa units may be useful for DVT/PE prevention. The duration of antithrombotic therapy in the highest risk patients is being studied, because there is some speculation that anticoagulation continued beyond the intermediate postoperative period may reduce the risk of thromboembolic phenomena as well as the risk of death.

NEUROLOGIC COMPLICATIONS Avoiding serous perioperative neurologic sequelae, including stroke, seizure, altered mental status, and operation-associated nerve injury, is a vital aspect of perioperative management. Most general anesthetics adversely affect central nervous system function. Even short-term anesthetic use impairs psychomotor performance for 5 hours and sleep patterns for 24 hours or more. Women at risk for cerebrovascular disease (as evidenced by transient ischemic attacks, peripheral vascular disease, or other events) may require specific preoperative evaluation and neurologic consultation. Although an asymptomatic carotid bruit may signal the need for imaging studies or consultation, it does not, by itself, increase perioperative risk. Presurgical neurologic and radiologic evaluation, extracranial carotid endarterectomy, systemic anticoagulation, and platelet inhibition are some of the potential perioperative risk-sparing procedures to be considered. Central nervous system function is important for those women with arterial disease who are at risk for other thrombotic events and, regardless of risk, the adverse effect of perioperative hypotension (even of short duration) should be avoided. The overall risk of recurrent stroke approaches 3.0% but, fortunately, the risk of stroke is 0.2% to 0.7% in those with no such history. Cerebral blood flow is unstable and brain metabolism is depressed for 6 to 8 weeks following a completed stroke, making it prudent to avoid any significant elective

operations during this interval. The risk of a second infarct may approach 20%, and the mortality rate is high (˜25%). Preoperative central nervous system (CNS) scanning may assist in determining the time of resolution of the initial infarct, when operative procedures are less likely to result in repeated infarct. Emergency procedures in those at risk for CNS infarct should completed in a perioperative environment that maintains an elevated blood pressure and avoids events that may increase the risk of CNS hypoperfusion. Vertebral basilar ischemic episodes are associated with a lower perioperative stroke risk than is carotid ischemia. Although perioperative neurologic evaluation may be appropriate, these patients usually are managed medically, with minimal surgical risks. Operative Neurologic Deficits Nearly 2.0% of gynecologic procedures are associated with postoperative lower extremity neurologic deficits, regardless of attention paid to preoperative care or positioning. Multiple nerves are at risk, regardless of positioning, surgical duration, or approach. Mechanisms of nerve injury include prolonged compression (e.g., femoral nerve), excessive traction, stretch (e.g., sciatic nerve) or transection (e.g., lateral femoral cutaneous nerve). Regardless of the extent of injury, attention to positioning, or retractor placement, these problems occur and are managed by evaluation and rehabilitation. Fortunately, most resolve. Attention to minimizing hip flexion and external rotation potentially lessens the risk of traction injury. Ilioinguinal and iliohypogastric nerve injuries can be diagnosed with relief after local infiltration of nerve block or excision.

SUMMARY POINTS The primary goal of the pelvic surgeon involves prospective preoperative recognition, evaluation, and management of existing medical comorbidities. Every pelvic surgeon should commit to the development of operative techniques that form a surgical skill set geared to the safe completion of indicated surgical procedures. Appropriate consultation should be sought if the level of qualification does not fit the necessary procedure. A thorough, multisystem physical examination is an essential part of the preoperative evaluation to detect important coexisting disease. The suspicion or diagnosis of coexistent morbidities which have not been diagnosed previously requires preoperative investigation. Laboratory and radiologic investigation should be conducted patiently and procedures directed in a safe, cost-effective manner. Coexisting or occult cardiac disease represents a significant contributor to perioperative morbidity. Particular attention to the detection, evaluation, and management of cardiac risk factors including CAD, hypertension, congestive heart failure, valvular disease, arrhythmias, and hypercholesterolemia is imperative. Particular attention to ACLS guidelines is important. Perioperative assistance by personnel experienced in cardiac care should be invoked when necessary. Perioperative pulmonary care is an important concern. The procedure should be designed to minimize pulmonary risks when possible. Postoperative attention to pulmonary toilet may render the best possible outcome. Perioperative renal failure is associated with high morbidity and mortality. No intervention, including prophylactic low-dose dopamine, mannitol, or furosemide therapy, has been shown to be of any benefit in reducing the incidence of perioperative renal insufficiency. The optimal surgical incision is chosen according to the procedure done, existing comorbidity, disease process, and patient preference. Appropriate preoperative skin antisepsis is necessary and, in combination with intravenous antimicrobial agents, will decrease the incidence of wound infection. Good surgical technique is mandatory for wound health and maximizing patient outcome. Classically, bowel preparation (mechanical, with or without antibiotic) is undertaken prior to many gynecologic procedures. Nasogastric suction has not been shown to lessen the risk of ileus, and only 10% of patients require insertion of a nasogastric tube postoperatively. Most patients can be managed without nasogastric suction. Proper use of prophylactic antibiotics lessens infectious risks in many pelvic procedures. Single-agent broad-spectrum antibiotic use, such as cephalosporin, doxycycline, or metronidazole, appears to be safe and effective for women undergoing abdominal or vaginal hysterectomy. The American College of Chest Physicians published guidelines regarding perioperative thromboprophylaxis. Patients at low risk should be mobilized aggressively postoperatively. Those at moderate and high risk should use pneumatic compression hose, elastic stockings, or pharmaceutical methods. Patients at highest risk may also benefit from combined mechanical and pharmaceutical prophylaxis. Approximately 2% of gynecologic procedures are associated with postoperative lower extremity neurologic deficits. Special attention should be paid to patient positioning, surgical duration, proper use of surgical retractors to prevent excessive traction and stretch, and careful operative techniques. SUGGESTED READINGS General Considerations Rutkow IM. Surgical operation in the United States. Then (1983) and now (1994). Arch Surg 1997;132:983–990.

Surgical Indications and Consent Guidelines for referral to a gynecologic oncologist: rationale and benefits. Gynecol Oncol 2000;78:S1–S13. Birkmeyer JD, Siewers AE, Finlayson E, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346:1128–1135. Montz HG. “Curbside” consultations in gynecologic oncology: a close look at a common practice. Gynecol Oncol 1999;74:456–459.

Preoperative Evaluation and Testing Orr JW Jr. Preoperative care. In: Gershenson D, ed. Operative gynecology. Philadelphia: WB Saunders, 2001. Hodges PJ, Kam PC. The perioperative implications of herbal medicines. Anaesthesia 2000;57:889–899. Fischer SP. Cost-effective preoperative evaluation and testing. Chest 1999;115:96S–100S. Roizen MF. More preoperative assessment by physicians and less by laboratory tests. N Engl J Med 2002:342:204–206. Brooks-Brunn JA. Predicators of postoperative pulmonary complications following abdominal surgery. Chest 1997;111:564–571. Eagle KA, Brundage BH, Chaitman BR, et al. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery: report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Circulation 2002;105:1257. Fleisher LA, Eagle KA. Lowering cardiac risk in noncardiac surgery. N Engl J Med 2001;345:1677–1682. Raymer K, Yang H. Patients with aortic stenosis: cardiac complications in non-cardiac surgery. Can J Anaesth 1998;45:855–859. Proceedings of the American College of Chest Physicians Fifth Consensus on Antithrombotic Therapy 1998. Chest 1998;114:439S–769S. Eagle KA, Rihal CS, Mickel MC, et al. Cardiac risk of noncardiac surgery: influence of coronary disease and type of surgery in 3368 operations. CASS Investigators and University of Michigan Heart Care Program. Coronary Artery Surgery Study. Circulation 1997;96:1882–1887. American Heart Association. ACLS Provider Manual. Dallas: American Heart Association, 2001.

Perioperative Pulmonary Evaluation Smetana GW. Preoperative pulmonary evaluation. N Engl J Med 1999;340:937–944.

Renal Disease Novis BK, Roizen MF, Aronson S, et al. Association of preoperative risk factors with postoperative acute renal failure. Anesth Analg 1994;78:143–149. Gilmour DT, Dwyer PL, Cary MP. Lower urinary tract injury during gynecologic surgery and its detection by intraoperative cystoscopy. Obstet Gynecol 1999;94:883–889.

Wounds and Incisions Orr JW Jr, Orr PF, Bolen D, et al. Radical hysterectomy: does the type of incision matter? Am J Obstet Gynecol 1995;173:399–406. Labib M, Palfrey S, Paniagua E, et al. The postoperative inflammatory response to injury following laparoscopic assisted vaginal hysterectomy versus abdominal hysterectomy. Ann Clin Biochem 1997;34:543–545. Orr JW Jr, Orr PF. Perioperative care. In: Greer BE, Montz FJ. Atlas of clinical gynecology, vol 4. 1998. Soper DE, Bump RC, Hurt WG. Wound infection after abdominal hysterectomy: effect of the depth of subcutaneous tissue. Am J Obstet Gynecol 1995;173:465–469. Vrijland WW, Tseng LN, Eijkman JH, et al. Fewer intraperitoneal adhesions with use of hyaluronic acid-carboxymethylcellulose membrane: a randomized clinical trial. Ann Surg 2002;235:193–199.

Gastrointestinal Care Zamora O, Pikarsky AJ, Wexner SD. Bowel preparation for colorectal surgery. Dis Colon Rectum 2001;44:1537–1549. Cheatham ML, Chapman WC, Key SP, et al. A meta-analysis of selective versus routine nasogastric decompression after elective laparotomy. Ann Surg 1995;221:469–478. Andrews FJ. Early postoperative feeding after major gynecologic surgery: evidence-based scientific medicine. Am J Obstet Gynecol 2001;185:1–4. Pearl ML, Frandina M, Mahler L, et al. A randomized controlled trial of a regular diet as the first meal in gynecologic oncology patients undergoing intraabdominal surgery. Obstet Gynecol 2002;100:230–234.

Infection ACOG Practice Bulletin. Clinical management guidelines for obstetrician-gynecologists. Washington, DC: ACOG, 2001. Scher KS. Studies on the duration of antibiotic administration for surgical prophylaxis. Am Surgeon 1997;63:59–62. Cinat ME, Wilson SE, Din AM. Determinants for successful percutaneous image-guided drainage of intra-abdominal abscess. Arch Surg 2002;137:845–849.

Thromboembolism Gertz W, Heit J, Clagett G, et al. Prevention of venous thromboembolism. Chest 2001;119:132S–175S. Silverstein MD, Heigh JA, Mohr DV, et al. Trends in the incidence of deep vein thrombosis and pulmonary embolism: 1. 25-year population-based study. Arch Intern Med 1998;158:585–593. Maxwell GL, Synan I, Dodge R, et al. Pneumatic compression versus low molecular weight heparin in gynecologic oncology surgery: a randomized trial. Am Coll Obstet Gynecol 2001;98:989–995.

Neurologic Warner MA, Warner DO, Harper CM, et al. Lower extremity neuropathies associated with lithotomy positions. Am Soc Anesth 2000;93:938–942. Wong GY, Warner DO, Schroeder DR, et al. Risk of surgery and anesthesia for ischemic stroke. Am Soc Anesth 2000;92:425–434.

Chapter 45 Pelvic Organ Prolapse Danforth’s Obstetrics and Gynecology

Chapter 45 John O.L. DeLancey and Kris Strohbehn

Pelvic Organ Prolapse

INTRODUCTION PELVIC FLOOR AND THE NATURE OF GENITAL PROLAPSE Epidemiology of Surgically Managed Pelvic Organ Prolapse and Urinary Incontinence Prevalence Vaginal Delivery and Pelvic Organ Prolapse BASIC ANATOMY AND PATHOPHYSIOLOGY OF THE PELVIC FLOOR Viscerofascial Layer Levator Ani Muscles Interaction Between the Muscles and Fasciae ISSUES RELATING TO INCONTINENCE CAUSES OF PELVIC FLOOR DAMAGE DIAGNOSIS AND CLASSIFICATION EVALUATING INDIVIDUAL ELEMENTS OF SUPPORT Anterior Vaginal Wall Uterus and Vaginal Apex Posterior Vaginal Wall Enterocele Rectocele Prolapse Subsequent to Hysterectomy PELVIC ORGAN PROLAPSE CLASSIFICATION Measurements Describing Prolapse Type and Size Pelvic Floor Measurements SYMPTOMS Symptoms of Anterior Wall Prolapse Symptoms Associated With Prolapse of the Uterus, Prolapse of the Vaginal Apex, or Enterocele Symptoms of Rectocele WHO SHOULD BE TREATED SURGICALLY SURGICAL PROCEDURES Vaginal Hysterectomy and Anteroposterior Colporrhaphy Basic Principles Underlying the Surgical Repair of Prolapse Preoperative Considerations Technique of Vaginal Hysterectomy Management of the Cul-de-sac and Suspension of the Vaginal Apex Anterior Colporrhaphy Posterior Colpoperineorrhaphy Le Fort Partial Colpocleisis Vaginal Repair of the Posthysterectomy Enterocele Posthysterectomy Prolapse Paravaginal Repair Conservative Measures SUMMARY SUMMARY POINTS SUGGESTED READINGS Epidemiology of Surgically Managed Pelvic Organ Prolapse and Urinary Incontinence Pelvic Floor and the Nature of the Genital Prolapse Basic Anatomy and Pathophysiology of the Pelvic Floor Issues Relating to Incontinence Causes of Pelvic Floor Damage Diagnosis and Classification, Evaluating Individual Elements of Support and Pelvic Organ Prolapse Classification Symptoms Surgical Procedures

INTRODUCTION Pelvic organ prolapse is a condition unique to the field of obstetrics and gynecology. It has helped define our specialty and continues to lie in the exclusive province of the obstetrician-gynecologist. Although often discussed as a purely mechanical phenomenon, prolapse is associated with significant functional problems. Stress urinary incontinence, micturition difficulties, and problems with defecation are all associated with prolapse. These functional derangements are not simply results of altered support of the bladder and rectum but have to do with the innervation and musculature of the urinary and intestinal tracts, as well. This chapter reviews the structural and functional aspects of prolapse necessary to understand and manage these conditions.

PELVIC FLOOR AND THE NATURE OF GENITAL PROLAPSE The pelvis lies at the bottom of the abdominopelvic cavity, and the pelvic floor closes the canal within the bony pelvis ( Fig. 45.1). The pelvic floor forms a supportive layer that prevents the abdominal and pelvic organs from falling through the opening within the pelvic bones. Its structural role can best be appreciated by considering a surgeon's hand placed through a transabdominal incision that pushes caudally on the pelvic organs. All of the structures that prevent this hand from passing through the pelvic canal constitute the pelvic floor. In addition to this supportive role, the pelvic floor must accommodate conception and parturition, while also controlling storage and evacuation of urine and feces. To understand the pelvic floor and genital prolapse, it is necessary to understand the mechanical strategies that evolution has put in place to prevent downward descent of the pelvic organs, as well as the process by which genital prolapse occurs. As Victory Bonney pointed out, the phenomenon of prolapse is similar to the maneuver that a scrub nurse uses to evert the in-turned finger of a surgical glove ( Fig. 45.2). Compressing the air within the glove drives the invaginated finger outward in much the same way that increases in intraabdominal pressure force the vagina and the uterus to prolapse. It is not the weight of the uterus that is important in the development of prolapse, but rather the forces placed on the pelvic floor by increases in intraabdominal pressure.

FIG. 45.1. Sagittal section of the abdomen and pelvis shows the relation of the pelvic floor to the abdominal cavity. (From Kelly HA. Gynecology. Baltimore: Appleton

and Co, 1928:64, with permission.)

FIG. 45.2. A: Diagrammatic representation of the vagina within the abdomen shows how increases in abdominal pressure (arrow) force the vagina to prolapse. B: This prolapse may be prevented by (C) constricting the lower portion of the vagina, (D) suspending the vagina from the pelvic walls, and (E) forming a flap-valve closure, wherein the vagina is pinned against surrounding structures.

Two mechanical principles explain how the pelvic floor prevents prolapse (see Fig. 45.2). First, the uterus and vagina are attached to the walls of the pelvis by a series of ligaments and fascial structures that suspend the organs from the pelvic sidewalls. Second, the levator ani muscles constrict the lumina of these organs, forming an occlusive layer on which the pelvic organs may rest. It is a combination of these two factors—suspension of the genital tract by the ligaments and fasciae and closure of the pelvic floor by the levator ani—that holds the vagina over the levator ani muscles and forms a flap-valve closure. This flap-valve mechanism is instrumental in keeping the posterior cul-de-sac closed and preventing the development of an enterocele.

Epidemiology of Surgically Managed Pelvic Organ Prolapse and Urinary Incontinence Prevalence Olsen and colleagues reported the prevalence of surgery for pelvic organ prolapse and urinary incontinence within a defined population. They studied 149,554 women over 25 years of age who were members of the Kaiser Permanente Northwest Health Maintenance Organization. During 1995, they identified 384 women who had surgical treatment for either pelvic organ prolapse or urinary incontinence, or both of these problems. A woman's lifetime risk for needing a single operation by age 80 years was 11.1%. Among this group of women, there was great variety in types and sizes of prolapse ( Table 45.1), and the overall incidence of surgery increased with age ( Fig. 45.3). The higher incidence was primarily due to the increase in surgery among older women for prolapse rather than for incontinence. Repeated operation was remarkably common, with 29.2% of patients requiring a second surgery. There is increasing risk of pelvic organ prolapse with subsequent parity ( Fig. 45.4), and the study of the obstetric factors that cause pelvic floor injury has been neglected.

TABLE 45.1. Preoperative prolapse severity according to operative site

FIG. 45.3. Age-specific incidence of surgery for pelvic organ prolapse or urinary incontinence. (From Olsen AL, Smith VJ, Bergstrom JO, et al. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 1997;89:501, with permission.) UI, urinary incontinence; POP, pelvic organ prolapse.

FIG. 45.4. Increase in risk of needing surgery for prolapse with increasing parity. (From Mant J, Painter R, Vessey M. Epidemiology of genital prolapse: observations from the Oxford Family Planning Association study. Br J Obstet Gynaecol 1997;104:579–585, with permission.)

These data show the remarkable frequency with which women require surgery for incontinence and prolapse but certainly underestimate the prevalence of these problems within the population. One would expect that many women who have prolapse and incontinence choose not to have surgery, so these data represent only those women operated on. Therefore, serious consideration needs to be given these conditions. Vaginal Delivery and Pelvic Organ Prolapse It is commonly accepted that vaginal delivery increases the likelihood that pelvic organ prolapse will develop. Mant, using the Oxford family planning cohort, found that the likelihood of a woman developing prolapse increased approximately eight-fold after two vaginal births and about 12-fold with four or more vaginal deliveries. Timonen and co-workers compared information concerning parity in 1,422 patients operated on for prolapse in their hospital between 1955 and 1965 with national parity data in Finland; these data are displayed in Figure 45.5. Only 4% of women with prolapse are nulliparous, while among the Finnish population 13.3% of women are nulliparous. In addition, they evaluated the size of the largest infant delivered by women in whom prolapse subsequently developed ( Fig. 45.6) and compared that with birthweight information for Finland (N = 52,955) during the same period (1957–1958). These data lend some sense of the degree to which vaginal birth influences pelvic organ prolapse.

FIG. 45.5. Parity information for women operated on for pelvic organ prolapse compared with national average. (From Timonen S, Nuoranne E, Meyer B. Genital prolapse: etiological factors. Ann Chir Gynaecol Fenn 1968;57:363, with permission.)

FIG. 45.6. Birthweight of largest infant delivered vaginally to women with pelvic organ prolapse and normative birthweight data for Finland. (From Timonen S, Nuoranne E, Meyer B. Genital prolapse: etiological factors. Ann Chir Gynaecol Fenn 1968;57:363, with permission.)

BASIC ANATOMY AND PATHOPHYSIOLOGY OF THE PELVIC FLOOR Viscerofascial Layer The topmost layer of the pelvic floor is a combination of the pelvic viscera and their connections to the pelvic walls and will be referred to as the viscerofascial layer. Although it is common to speak of the fasciae and ligaments as separate from the pelvic organs, unless these fibrous structures have something to attach to (e.g., the pelvic organs) they have no structural integrity. The uterus and vagina are attached to the pelvic walls by the fibrous tissue referred to as the endopelvic fascia. It forms a sheetlike mesentery that is continuous from the uterine artery to the point at which the vagina fuses with the levator ani muscles as it passes through the urogenital hiatus. The tissues that connect the uterus are called the parametria, and those that attach to the vagina are the paracolpium. Although they are given regional names, they are actually one continuous entity. The parametria comprise the cardinal and uterosacral ligaments. These are two different elements of the same tissue ( Fig. 45.7). The uterosacral ligaments are the visible and palpable medial margin of the cardinal–uterosacral ligament complex. As is true of the remainder of the parametria, they contain smooth muscle, nerves, and blood vessels and are not the same type of tissue seen in the fascia of the rectus abdominus muscle, which is dense regular connective tissue.

FIG. 45.7. Sagittal section of the pelvis shows the support structures of the genital tract. A: The bladder, urethra, and uterine corpus (above the cervix) have been removed. B: All of the pelvic organs have been removed to show the levator ani muscles.

Opposite the external cervical os, the sheet of tissue that attaches the genital tract to the pelvic wall arbitrarily changes name from the parametrium to the paracolpium. The paracolpium has two portions ( Fig. 45.8). The upper portion (i.e., level I) consists of a relatively long sheet of tissue that suspends the vagina by attaching it to the pelvic wall in an area similar to that of the cardinal–uterosacral ligament complex. It is this portion that prevents the upper vagina from prolapsing after the uterus has been removed.

FIG. 45.8. Support structures of the vagina after hysterectomy. The bladder has been removed to expose the vagina. A: The paracolpium. B: The different levels of support structures. C: The details of the pubocervical and rectovaginal fasciae after a wedge of vagina and urethra has been removed (inset). (From DeLancey JOL. Anatomic aspects of vaginal eversion after hysterectomy. Am J Obstet Gynecol 1992;166:1717, with permission.)

In the midportion of the vagina, the paracolpium attaches the vagina laterally and more directly to the pelvic walls (i.e., level II). This stretches the vagina transversely between these two lateral attachments (see Fig. 45.8B). This arrangement has functional significance. The structural layer that supports the bladder (i.e., pubocervical fascia) is composed of the anterior vaginal wall and its attachment through the endopelvic fascia to the pelvic wall. The term fascia is used commonly, but this is not a layer separate from the vagina. A term has been proposed for this layer, the fibromuscular layer of the vagina, that contains both smooth muscle and connective tissue. We use these terms interchangeably in the remainder of the text, because the surgeon generally refers to this layer as fascia. Similarly, the posterior vaginal fibromuscular layer and its connection to the pelvic walls forms the restraining layer that prevents the rectum from protruding forward. In the distal vagina (i.e., level III), the vaginal wall is attached directly to surrounding structures without any intervening paracolpium. The support that lies under the urethra has special importance for urinary incontinence. The endopelvic fascia in this region is better developed and is tougher than the tissues of the upper vagina, in the area under the bladder. This provides better support for the vesical neck than for the bladder. This layer of suburethral endopelvic fascia attaches laterally to the arcus tendineus fasciae pelvis and also to the medial border of the levator ani muscles. Loss of this normal support of the urethra at the vesical neck is responsible for stress incontinence of urine. Levator Ani Muscles Below the uterus, vagina, bladder, and rectum lie the levator ani muscles (see Fig. 45.7B). The medial portion of the levator ani muscles frequently has been called the pubococcygeus, but the pubovisceral muscle is a better term because it extends from the pubic bone and attaches to the vagina and rectum, with only a few

insignificant fibers ending in the coccyx (Lawson, 1971). This strong, robust, fatigue-resistant striated muscle starts on the inner surface of the pubic bone near the midline and passes behind the rectum, to return to the pubic bone on the other side. The lateral walls of the vagina are attached to this muscle, which then inserts into the anus in the intersphincteric groove between the internal and external anal sphincter, as well as passing around its dorsal surface in a slinglike loop. The normal resting tone of this muscle squeezes the rectum, vagina, and urethra closed by compressing them against the pubic bone. The pubo-rectalis muscle originates lateral to the pubovisceral muscle and forms a sling behind the rectum at the anorectal angle. Arising from the lateral pelvic walls (at the tendineus arch of the levator ani muscles) is the iliococcygeal muscle that forms a horizontal shelf on which the upper pelvic organs rest. Interaction Between the Muscles and Fasciae The interaction between the pelvic floor muscles and ligaments is critical to proper function. As long as the pelvic floor musculature functions normally, the pelvic floor is closed, and the ligaments and fasciae are under no tension. They simply act to stabilize the organs in their position above the levator ani muscles. When the pelvic floor muscles relax or are damaged, the pelvic floor opens. The vagina lies between the high intraabdominal pressure and low atmospheric pressure, where it must be held in place by the ligaments. Although the ligaments can sustain these loads for short periods, if the pelvic floor muscles do not close the pelvic floor, then it is more likely that the connective tissue will become damaged and eventually fail to hold the vagina in place.

ISSUES RELATING TO INCONTINENCE The mechanisms of urinary and fecal incontinence are more difficult to explain than the mechanical basis of genital tract support, and they are not understood entirely. As is true for genital prolapse, urinary continence depends on both support of the urethra and its ability to remain closed as a result of its innate constriction. Normal support of the urethra allows it to be compressed closed by increases in abdominal pressure (i.e., pressure transmission). Although both support of the urethra and pressure transmission are important for stress urinary continence, they are not the only factors involved. Stress incontinence may occur in women with normal urethral support of the vesical neck and urethra but with inadequate constriction by the muscles within the urethral walls. If the a-adrenergically innervated smooth muscle of the vesical neck does not function normally and the vesical neck is open, this is referred to as type III incontinence. This type of incontinence is differentiated from types I and II incontinence, which involve different degrees of support loss. When the proximal urethra is patent at rest, urine already lies below the area where pressure transmission has its effect and, therefore, bypasses this mechanism of continence. The support of the urethra is not due exclusively to fascial structures; active muscle contraction plays an important role in urethral support. In addition, the musculature of the urethra contracts during a cough and increases urethral closure. Therefore, neuromuscular mechanisms play an important role in the pathophysiology of stress incontinence and must be considered in understanding this group of diseases. By the same token, anal continence can be explained only partially by the mechanical structures that support the anorectum and by preservation of the anorectal angle through contraction of the puborectalis. Anal incontinence can occur with normal support of the anorectum and preserved anorectal angle. There are other important factors in maintaining anal continence, including normal integrity of the muscles of the anal sphincter complex and their innervation.

CAUSES OF PELVIC FLOOR DAMAGE Several factors influence the development of genital prolapse and urinary incontinence ( Table 45.2). The inherent strength of the muscles and connective tissue of an individual, damage to these structures that occurs with birth, the rate at which they deteriorate with age, and the loads that they are subjected to during life all play a role in the development of genital prolapse and incontinence. Understanding the interplay among these various factors is critical to appreciating the pathophysiology of genital prolapse.

TABLE 45.2. Factors involved in pelvic organ prolapse

Although it is obvious that the native strength of the levator ani muscles and endopelvic fascia plays an important role in prolapse, the changes that vaginal birth causes in the pelvic floor deserve specific examination. Distension of the vagina during vaginal birth usually is blamed for prolapse, yet the cervix undergoes a much greater degree of dilation during parturition but recovers sufficiently so that forceful dilation is needed if dilation and curettage is done later in life. Damage to the levator ani muscles and their innervation is common during vaginal birth, and loss of the muscle's ability to support the pelvic organs and unload the ligaments may be partially responsible for the breakage and elongation of the ligaments later in life. Certainly, some damage to the connective tissue of the parametria and paracolpium occurs during childbirth and can contribute to subsequent prolapse. Although usually thought of as stretching, prolapse also includes instances in which the connective tissue supports rupture. This acute injury mechanism is much more likely than stretching, because usually it is rupture that occurs with overload of other ligaments in the body (e.g., knee). Chronic injury through attenuation of the endopelvic fascia with age contributes and explains why prolapse occurs many years after childbirth. Failure of connective tissue to heal the many minor injuries that occur from day to day may also play a role. Age and chronic disease may cause weakening of the tissues, and heavy lifting or chronic coughing may cause progressive damage. This wear and tear on the connective tissues depends on the magnitude of the stresses placed on them, as well as how well damage is repaired by the body.

DIAGNOSIS AND CLASSIFICATION Determining the type and severity of prolapse in any given patient is a skill that should be acquired through practice and careful observation. Characterizing the degree of support loss as normal or abnormal depends on comparisons with the findings in normal multiparous women in the examiner's experience. It is, therefore, helpful to perform the same examination on a sufficient number of asymptomatic patients without prolapse to become familiar with the range of normal support. In performing an examination to determine the type and severity of prolapse, the practitioner has two important points to consider: Examination must be made with the patient straining forcefully enough that the prolapse is at its greatest. The examiner must examine each element of support independently. If a patient is not able to strain sufficiently in the lithotomy position so that the prolapse is at its largest, examination in the standing position may be necessary. This is a critical point, because it is only when the prolapse can be seen in its fullest extent that all of its various elements can be assessed. If the entire extent of the prolapse is not observed, some element may be overlooked. For example, a large cystocele may be seen initially when the patient strains. It may be only with continued effort by the patient that an enterocele and prolapse of the vaginal apex can be demonstrated. To make sure all aspects of the prolapse can be evaluated, the patient should be asked how large her prolapse is at its largest, and the physician should persist in the examination until that size is achieved. Once the prolapse is visible, the elements of the vagina and pelvic organs that have prolapsed can be evaluated. The examination should then focus on what specific defects in support are present and how severe the prolapse is, and there should be some evaluation of the cause of the prolapse. Once the prolapse is maximally developed, the physician should begin by identifying how much the anterior wall, cervix, and posterior wall have prolapsed. The anterior and posterior walls should be examined separately by retracting the opposite wall with the posterior half of a vaginal speculum. A stepwise, site-specific examination is important because a large cystocele, for example, may hold a potential rectocele in place and, therefore, hide it. If a rectocele is not recognized preoperatively, its repair may be overlooked and the defect can become symptomatic postoperatively. These observations have been confirmed on dynamic colpoproctography imaging studies of the pelvic floor, with contrast placed in the bladder and rectum while the patient strains in the standing position. If the patient has a full bladder when imaging is performed, a rectocele can be obstructed and will not be evident until the patient empties her bladder. Care must also be taken to assess how much of the loss of support is from a defect of the apical (level 1) support. It is not uncommon to correct the apical defects and find that much of what was considered a cystocele and rectocele have been corrected. Examination while under anesthesia is used to evaluate pelvic masses, but it is not the optimal time to identify defects caused by prolapse because of the patient's inability to perform the Valsalva maneuver and because of loss of normal levator tone.

EVALUATING INDIVIDUAL ELEMENTS OF SUPPORT Anterior Vaginal Wall Examination of the anterior vaginal wall should establish the status of urethral support, as well as bladder support. The urethra is fused with the lower 3 to 4 cm of the vaginal wall, and abnormal support in this region is properly referred to as a urethrocele ( Fig. 45.9). Defective support of the upper portion of the vagina is called a cystocele, because the bladder lies adjacent to this portion of the vaginal wall ( Fig. 45.10). The urethrovesical crease, normally visible on examination, forms the line of demarcation between these two areas of support ( Fig. 45.11). When support of the entire anterior wall is defective, the term cystourethrocele is used.

FIG. 45.9. Displacement cystourethrocele with intact rugal folds caused by lateral detachment of the pubocervical fascia. (Copyright © DeLancey, 1993.)

FIG. 45.10. Distension cystourethrocele caused by midline failure of the pubocervical fascia. (Copyright © DeLancey, 1993.)

FIG. 45.11. Angulation in the anterior vaginal wall, called the urethrovesical crease (arrow), indicates the location of the urethrovesical junction. (Copyright © DeLancey, 1993.)

The anterior vaginal wall should be above the hymenal ring during straining. Descent of the lower anterior vaginal wall to the level of the hymenal ring during straining is characteristic of a urethrocele and is seen often in patients with stress urinary incontinence. This is due to loss of urethral support and corresponds to the loss of the posterior urethrovesical angle on radiographic studies of patients with stress incontinence. The lower anterior vaginal wall is mobile in all women and may move significantly in continent multiparas. Therefore, motion of this region does not establish stress incontinence but rather indicates the degree to which the support of the urethra has failed. Descent below the hymenal ring is definitely abnormal and indicates a cystourethrocele whether or not stress incontinence is present. The anterior vaginal wall above the urethrovesical crease usually lies in a flat plane at about a 45-degree angle from the horizontal (see Fig. 45.11). Descent below the level of the hymenal ring is significant. This descent can be caused by one of three entities: Separation of the paravaginal attachment of the pubocervical fascia from the white line due to detachment from the ischial spine Loss of the vagina's attachment to the cervix Tearing in the pubocervical fascia that results in herniation of the bladder through this layer Uterus and Vaginal Apex The vagina and cervix are fused with one another, and prolapse of the uterine cervix is associated invariably with prolapse of the upper vagina, as well. When the uterus descends below its normal level, the term uterovaginal prolapse is appropriate, although uterine prolapse commonly is used. In patients in whom the uterus has been removed, descent of the vaginal apex below its normal position in the pelvis is referred to as prolapse of the vaginal apex, and when the vagina turns entirely inside out, the term vaginal eversion is used. The location of the cervix customarily is used to gauge the severity of uterine prolapse ( Fig. 45.12). Its position relative to the hymenal ring should be noted while the prolapse is at its greatest. If the cervix is not visible because of a cystocele or rectocele, then its location may be palpated while having the patient strain. When the cervix descends to within 1 cm of the hymenal ring, there is a significant loss of support. In instances in which the uterus is not necessarily going to be removed, uterine support should be tested before it is assumed that the uterus is well supported. This can be done by grasping the cervix with a tenaculum or ring forceps and applying

traction until it stops descending. Occult prolapse, in which the cervix comes below the hymenal ring, can be detected in this way.

FIG. 45.12. Uterine prolapse with the cervix extending 3 cm below the hymen. (Copyright © DeLancey, 1993.)

In addition to determining how far the cervix descends, its length should be measured. Cervical elongation is frequent in individuals with prolapse, and the uterine corpus often may lie in its normal location. Awareness of cervical elongation preoperatively will allow the surgeon to proceed expeditiously with the hysterectomy, rather than hoping with every pedicle that the uterine arteries will soon appear. Posterior Vaginal Wall The posterior vaginal wall is the site of both rectoceles and enteroceles. Evaluation and correction of these two problems challenge even the most experienced gynecologic surgeon, and they are probably the most difficult to understand of all pelvic support defects. Because dyspareunia can follow repair, correction of asymptomatic posterior wall defects is not without risk. On the other hand, having a rectocele or enterocele develop after vaginal hysterectomy and anterior colporrhaphy is an undesirable outcome, and careful consideration of the support of the posterior vaginal wall is important. Three questions should be asked by the physician when examining the posterior wall. Is it supported normally? If not, is it a true rectocele or a pseudorectocele? Is an enterocele present? A rectocele is present when the anterior rectal wall and overlying vagina protrude below the hymenal ring. An enterocele exists when the cul-de-sac becomes distended with the intestine and bulges the posterior vaginal wall outward. There are also occasions in which the posterior wall appears to bulge into the vagina, not because of poor support of the rectal wall but because of a deficiency in the perineal body. This has been referred to by Nichols and Randall as a pseudorectocele and can be differentiated easily from a true rectocele because the anterior rectal wall contour is normal on rectal examination. Another type of pseudorectocele may be suspected when there is apical descent of the upper vagina or cervix and apparent loss of posterior support. However, often when the normal apical support is restored (by temporarily supporting it with ring forceps in the office or after surgical repair of the apical descent), a suspected rectocele is not evident. This is important to determine preoperatively, because loss of tone of the levator ani muscle and anal sphincter muscle with muscle paralyzing agents during anesthesia make it harder to establish the existence of a true rectocele. Enterocele There is always a cul-de-sac between the upper vagina and the rectum. This allows a culdocentesis to be performed and a colpotomy to be made through the posterior vaginal wall at the beginning of a vaginal hysterectomy. The peritoneal pouch normally extends 3 to 4 cm beyond the junction of the vagina and cervix. Therefore, the absence of an enterocele in normal women must be explained by factors that keep the cul-de-sac closed rather than by the absence of a peritoneal space between the upper vagina and rectum. It is the suspension of the upper vagina near the sacrum in a position where it may rest over the rectum and intact levator plate that keeps this space closed. There are two types of enteroceles: pulsion enterocele and traction enterocele. A pulsion enterocele exists when the cul-de-sac is distended and appears as a bulging mass that is inflated by increases in abdominal pressure. This may occur with either the vaginal apex or uterus well suspended, in which case the cervix or vaginal apex is at a normal level and the enterocele dissects between the vagina and the rectum. When an enterocele is associated with prolapse of the uterus or vaginal apex, then the prolapse and enterocele occur together. A traction enterocele represents a situation in which prolapse of the uterus pulls the cul-de-sac peritoneum down with it, but there is no bulging or distension of the cul-de-sac when abdominal pressure rises. This condition usually is found at the time of vaginal hysterectomy when the cervix has prolapsed. It represents a potential enterocele rather than an actual enterocele, because there is no bulging mass separate from the uterus. Unlike uterine prolapse, which is obvious because of the protrusion of the easily recognized uterine cervix, enteroceles and rectoceles rarely are evident on examination. Therefore, the key to detecting an enterocele lies in actively looking for it whenever a patient who has prolapse is examined. Detection of an enterocele is performed best in the awake, straining patient by noting a mass of small intestine between the rectum and vagina; it may not be suspected in a supine individual at rest. Anatomically, an enterocele extends from the apex of the vagina downward, whereas a rectocele typically begins in the lower portion of the vagina. An enterocele sometimes is evident as a bulge that overrides the more caudal rectocele ( Fig. 45.13). Careful inspection of the posterior vaginal wall with a speculum retracting the anterior wall sometimes can suggest that an enterocele is present. The key to detecting a pulsion enterocele lies in palpating the small bowel between the vagina and rectum during rectovaginal examination, with the patient straining so that the prolapse is protruding. To do this, an index finger is placed in the rectum and a thumb is placed in the vagina. Then, with the patient straining, the rectovaginal space may be palpated to detect the bulge of the enterocele and the presence of small bowel, omentum, or large bowel.

FIG. 45.13. “Double hump” sign of an enterocele overriding a rectocele. (Copyright © DeLancey, 1993.)

Rectocele The hallmark of a typical rectocele is the formation of a pocket that allows the anterior rectal wall to balloon downward through the introitus. When a rectal examination is performed with the prolapse fully developed, a rectocele exists if there is an extension of the rectal lumen below the axis of the anus ( Fig. 45.14). This not only provides the diagnosis but also illustrates the mechanism by which rectoceles create their symptoms. As long as the anterior rectal wall has a smooth contour and no sacculation, even though it may be more mobile than normal, stool will pass through the anus. However, when a pocket develops as the patient strains, stool becomes trapped in it, and difficulty with evacuation can occur.

FIG. 45.14. A: Pelvic examination shows a rectocele. B: Lateral bead chain cystourethrogram with the patient supine and with contrast in the vagina and rectum shows a protruding rectocele. (Copyright © DeLancey, 1993.)

Prolapse Subsequent to Hysterectomy Special consideration should be given to patients who have prolapse after hysterectomy to assess whether or not prolapse of the vaginal apex is present. When the uterus is in situ, the cervix calls attention to the poor support of the cervix and upper vagina. In instances of posthysterectomy vaginal prolapse, descent of the vaginal apex is more easily missed. If it is overlooked and an anteroposterior colporrhaphy is not accompanied by suspension of the vaginal apex, the colporrhaphy will fail to cure the apical prolapse, and the problem is not corrected. Overlooking apical support loss also can lead to overly aggressive excision of vaginal tissues during anteroposterior colporrhaphy and a shortened vagina. Examination of patients who have previously had a hysterectomy should include a specific effort to determine the location of the vaginal apex when the prolapse is at its largest. The apex is identified by the vaginal scar at the hysterectomy site ( Fig. 45.15). Vaginal prolapse is present when the hysterectomy scar lies below the level of the hymenal ring. If the apex descends to within the lower one third of the vagina with straining, a significant deficit in support of the apex is present, and the vagina should be resuspended during repair.

FIG. 45.15. Eversion of the vagina after hysterectomy. Note that the vaginal apex, indicated by the puckered scar where the cervix had been removed, lies below the hymenal ring. (Copyright © DeLancey, 1993.)

PELVIC ORGAN PROLAPSE CLASSIFICATION Several classification systems have been used to describe the sizes and types of pelvic organ prolapse in an individual woman. Because many of these systems use similar words to indicate different degrees of prolapse, confusion has arisen concerning the size of prolapse. For example, grade 2 uterine prolapse in some systems indicates that the cervix descends halfway between its normal position and the introitus. In other classifications, grade 2 can mean that one half of the uterus is outside the introitus. A system that standardizes terminology has been adopted by several groups, including the International Continence Society, the American Urogynecologic Society, and the Society of Gynecologic Surgeons. This standardized terminology (Pelvic Organ Prolapse Quantification, or POP-Q) provides a system that can describe the type of prolapse, as well as quantify the degree of prolapse in each area. Although this standardized system seems somewhat cumbersome when described in writing, in actual practice it is quite simple. The following section first considers the measurements that describe the type and size of prolapse and then discusses the measurements concerned with the changes in the urogenital hiatus in the levator ani muscles through which the prolapse descends. Measurements Describing Prolapse Type and Size To describe the nature of a woman's prolapse, it is necessary to do the following: (a) document what part or parts of the genital tract have prolapsed and (b) indicate how far down each part of the vaginal wall or cervix has descended. Prolapse description must include a consideration of anterior vaginal wall descent, posterior wall descent, and uterine descent (or prolapse of the vaginal apex after hysterectomy). Furthermore, because different parts of the anterior wall might suffer support damage, the system provides for determining the status of each level of vaginal support. For example, the distal anterior vaginal wall adjacent to the urethra may be well supported, while the portion of the vagina under the bladder may prolapse. This system addresses the need to make individual assessments of different parts of the vaginal wall. The three levels of vaginal support (see Fig. 45.8) must be assessed, corresponding to the different anatomic regions of vaginal support. Level I: support of the vaginal apex and uterus Level II: support of the bladder and rectum Level III: support of the urethra and perineal body In levels II and III, the anterior and posterior vaginal wall are considered separately, while in level I the cervix (or vaginal apex) and posterior fornix must be assessed. To understand the POP-Q classification system, refer to Figure 45.16. The size and type of prolapse are measured by determining the location of a series of points on the anterior and posterior vaginal walls relative to the hymenal ring. Points at each of the three levels are measured. Positive numbers reflect measurements of the vaginal points that have prolapsed below the level of the hymen and negative numbers reflect measurements above the hymen. It should be noted that this descriptive scheme does not distinguish between rectocele and enterocele but simply provides a way to quantify the amount of vaginal wall descent in each specific area. Additional examination and written comments concerning these important differences should be made.

FIG. 45.16. Six sites (points A ant, B ant, C, D, B post , and A post ), genital hiatus (gh), perineal body (pb), and total vaginal length (tvl) used for pelvic organ support quantitation.

A summary of the measurements obtained during the POP-Q examination is noted in Table 45.3.

TABLE 45.3. Stages of pelvic organ prolapse based on measurement of specific sites

I. Vaginal points (vaginal profile): patient straining maximally A. Level I: apex and cervix: points C and D (D is omitted if patient has had a hysterectomy) B. Level II: midvaginal points 1. Anterior wall: B a or B ant 2. Posterior wall: B p or B post C. Level III: distal vagina (perineum and urethrovesical neck) 1. Anterior wall: A a or A ant 2. Posterior wall: A p or A post 3. External measurements: obtained with patient at rest and again straining A. Genital hiatus: length gh B. Perineal body: length: pb II. Internal digital measurements: obtained with patient at rest with vaginal apex restored to normal position A. Total vaginal length: tvl To measure the lower third of vaginal support (level III), the location of a pair of points that normally lies 3 cm above the hymenal ring is assessed. Points measured at level III are called points A. One can imagine marking these vaginal points 3 cm above the hymen with a marker and then recording the position of these points in relation to the hymen with the subject straining maximally. Anteriorly, this corresponds to the approximate location of the urethrovesical junction, and this measurement assesses urethral descent (A a or A ant). Posteriorly, this region is normally occupied by the tissues of the perineal body (A p or A post). By definition, the highest possible position of either point A ant or A post is 3 cm above the hymen (-3), and the lowest position is 3 cm below the hymen (+3). To assess midvaginal support (level II), the most dependent part of the vaginal wall above point A ant is used. This point is called B a , or B ant, on the anterior wall and B p , or B post, on the posterior wall. This is, therefore, not a fixed point along the surface of the vagina, but rather it is marked at whatever location is the most caudal (distal) portion of that vaginal segment at maximal prolapse protrusion. In a normally supported vagina, this will be the same as point A ant, whereas in a woman with procidentia it will be the same as point C. The same is true for the posterior vaginal wall at point B post. Point C corresponds to the most distal portion of the uterine cervix or of the hysterectomy scar in the vagina in those patients who have had the uterus removed. Point D denotes the posterior fornix (that point at which the posterior vaginal wall changes direction). In addition to the positions of these points, the total length of the vagina is noted. Once these data have been gathered, a simple line diagram can be constructed by plotting these points relative to the hymen to provide a graphic representation of the prolapse ( Fig. 45.17).

FIG. 45.17. Diagram of prolapse sites. (From Viereck V, Peschers U, Singer M, et al. Metrische Quantifizierung des weiblichen Genitalprolapses: eine sinnvolle Neuerung in der Prolapsdiagnostik? Geburtshilfe Frauenheilk 1997;57:177, with permission.)

So far, each element of the prolapse has been considered separately. It is also possible to give an overall description to the size of the prolapse by looking at the most dependent part of the protruding vagina or uterus. In this way, different stages—stage 0 through stage IV-can be defined; a description of these is shown in Table 45.3. A briefer version of the original description of the POP-Q staging system has been summarized, also. The choice of the term stage here is somewhat unfortunate, because most women with stage I and II support are anatomically normal. The implication that they have a stage of prolapse is incorrect. Hopefully this problem with terminology will be corrected in the future. Pelvic Floor Measurements In normal women, the levator ani muscles close the pelvic floor. In women with pelvic organ prolapse, the urogenital hiatus within the levator ani muscles is the opening through which the vagina prolapses. This hiatus is enlarged in women with pelvic organ prolapse. The size of the urogenital hiatus and thickness of the perineal body can easily be measured to describe the changes that have occurred in the pelvic floor. The anteroposterior diameter of the genital hiatus extends from the arch of the pubic bone to the front of the perineal body, while the thickness of the perineal body is measured from the anterior margin of the perineal body to the center of the anal verge. The urogenital hiatus is held closed by the constant activity of the levator ani muscles, and the diameter of this opening enlarges in many women with prolapse. This classification system is detailed and specific. It requires careful examination and assessment. Although at first it seems quite detailed, it is simply the quantitative documentation of the individual defects that experienced surgeons always have found necessary to assess. Some clinicians will not find it expedient to measure each of these sites, but intelligent, detailed analysis of each site of support is important to plan properly any repair for pelvic organ prolapse.

SYMPTOMS All types of prolapse have several symptoms in common. Once the vagina prolapses below the introitus, it becomes the structural layer between the high pressures in the abdominal space and the relatively low atmospheric pressure. The downward force that this pressure differential creates puts tension on the fasciae and ligaments that support the vagina and uterus. This results in a dragging sensation where the tissues connect to the pelvic wall, usually identified by patients as occurring in the groin, and in sacral backache caused by traction on the uterosacral ligaments. This type of discomfort resolves when the patient lies down and the downward pressure is reduced. In addition, exposure of the moist vaginal walls leads to a sensation of perineal wetness that may be confused with urinary incontinence, and it also can give rise to ulceration of the vaginal wall. Most patients have an underlying sense of insecurity that is difficult for them to describe and is often expressed as a feeling that “something is just not right.” Sometimes patients who feel the cervix or vagina protruding have fears that they have a cancer and may be relieved to find that the condition is related to prolapse. Although patients may find it difficult to put their symptoms into words, the symptoms can cause significant distress and should not be ignored. Symptoms of Anterior Wall Prolapse The symptoms of cystourethrocele are varied, and the two primary ones are paradoxical. On the one hand, loss of support of the urethra and the lower vaginal wall is associated with stress urinary incontinence, whereas loss of support of the upper anterior vaginal wall and bladder base can cause difficulty in emptying the bladder. This inability to empty the bladder completely is probably related to voiding by the Valsalva maneuver. If there is a detrusor contraction, there should be no reason for a woman with a cystocele not to empty her bladder, and many women with a significant cystocele have normal postvoid residual urine volumes. When a woman strains to void, however, the cystocele simply gets bigger, and no impulse is provided for urine to flow through the urethra. In addition to these functional symptoms, many patients with a cystourethrocele complain of urinary urgency and frequency. This probably arises from stretching of the bladder base that accompanies its prolapse through the vaginal introitus; it is often less pronounced at night when patients are supine. Patients have a varying amount of support loss under the urethra or bladder, and symptoms vary along the spectrum from incontinence to urinary retention. As is true for other forms of prolapse, it is important to correlate a patient's symptoms with the physical findings so these problems can be addressed. Symptoms Associated With Prolapse of the Uterus, Prolapse of the Vaginal Apex, or Enterocele Few specific symptoms are related to prolapse of the uterus, prolapse of the vaginal apex, or enterocele. Patients with these conditions usually complain of the generalized symptoms of prolapse mentioned above. Some have urgency and frequency, probably related to pressure of the prolapse on the bladder base, but this is variable. In addition, patients with large, thin enteroceles occasionally have a sense of impending rupture. Although this is an uncommon problem, it should not be overlooked. Symptoms of Rectocele The cardinal symptom of a rectocele is difficulty in emptying the rectum. As a woman bears down to evacuate the rectum, stool is pushed into the rectocele, and the harder she strains, the bigger the rectocele becomes. Because constipation is common in older women, it is important to differentiate between infrequent bowel movements due to poor colonic motility or inadequate dietary fiber and difficulty due to a rectocele. Many women have found that if they press between the vagina and rectum to elevate the rectocele, this maneuver helps with defecation. This finding supports the fact that the rectocele is the source of the problem.

WHO SHOULD BE TREATED SURGICALLY A decision about when an operation for prolapse should be performed is based on the individual woman's situation. It depends on the size of the prolapse, the presence or absence of symptoms, and whether or not physiologic complications have arisen because a prolapse is present. Conservative treatment with a pessary may be considered. There is little data on continuance rates with long-term pessary use. Contrary to prior recommendations that pessaries should be offered only to those women who are not operative candidates or to those awaiting surgery, a survey of urogynecologists found that 77% would offer a pessary for symptomatic patients as first-line treatment. When the prolapse lies at or above the level of the hymenal ring, surgery should be performed only if definite symptoms are present and can be attributed reliably to the prolapse. Examples of this include: A patient with a cystourethrocele at the level of the hymen who has significant stress incontinence A woman with the cervix in the lower one third of the vagina, who has the characteristic dragging discomfort of a prolapse that resolves when she lies down or puts a supportive pessary in place A patient with a small rectocele in whom a definite pocket can be detected on rectal examination and elevation of the perineum relieves problems with defecation It is unusual for women in whom the uterus has not yet descended to the level of the hymenal ring to have symptoms caused by descent of the uterus. Symptoms of pressure, back pain, or feelings that something is coming out should be studied to confirm that they are related to the prolapse, and not to other factors, before a decision to operate is made. If the symptoms go away when a pessary is placed or a patient notices prompt relief when she lies down, this helps to confirm that the prolapse is the source of the symptoms. A woman with low back pain whose discomfort persists after she is supine is more likely to have arthritic or musculoskeletal pain than pain caused by uterine prolapse, and the results of surgery are likely to disappoint her. Among patients whose prolapse descends several centimeters below the hymenal ring, symptoms are usually present, and surgery will relieve this distress. Some women with a large prolapse will deny any symptoms, because they do not wish to undergo surgery or because they simply are not troubled by the prolapse. Their wishes should be respected. On the other hand, potential complications from the prolapse should be considered and discussed frankly. A prolapse should be repaired, despite a lack of symptoms, if a patient has recurrent urinary tract infections associated with an increased postvoid residual urine volume. Ureteral dilation caused by the prolapse, which leads to an impairment of renal function, occurs in some patients with large prolapses and should be considered an indication for repair. An intravenous pyelogram and renal function testing can detect these abnormalities and indicate the need for treatment. The intravenous pyelogram should include films made with the patient standing and the prolapse present, because the supine position may mask significant dilation.

SURGICAL PROCEDURES Vaginal Hysterectomy and Anteroposterior Colporrhaphy Vaginal hysterectomy with appropriate anteroposterior colporrhaphy is the operation most often performed for the treatment of uterovaginal prolapse. Because simple removal of the uterus does nothing to correct the prolapse of the vaginal walls that accompanies descent of the cervix, it is necessary to consider how an operation works to cure these conditions properly. Basic Principles Underlying the Surgical Repair of Prolapse Removal of the uterus does not improve support of the genital tract. Successful repair depends on the way in which a surgeon resuspends the vagina and repairs other defects in pelvic organ support when performing a vaginal hysterectomy and anteroposterior repair. Once the uterus has been removed, the vaginal apex should be attached to a point that is higher in the pelvis than the vagina to elevate it to a normal position. During the vaginal hysterectomy, the cardinal and uterosacral ligaments can be shortened and used to resuspend the vagina. In addition, techniques such as McCall culdoplasty anchor a suture to a point on the uterosacral ligaments higher than the preoperative position of the vagina and use this suture to pull the vaginal apex to a point higher in the pelvis. It is the elevation of the vagina rather than removal of the uterus that is critical. Preoperative Considerations In addition to the usual thorough history, physical examination, and other medical measures necessary prior to any major operation, preoperative preparation for vaginal hysterectomy should include a pelvic examination, Pap smear, and urinalysis. Special care should be taken, as previously described, to determine the exact type and degree of prolapse and to detect any potential ovarian malignancy. A saline enema (e.g., Fleet enema) the night before surgery empties the rectum in the

event that a rectal examination is required during surgery and provides more room posteriorly, improving exposure. Patients should receive a prophylactic antibiotic to minimize the risk of postoperative infections (cefazolin sodium 1 g intramuscularly or intravenously, just before going to the operating room, is appropriate). Cervical cytologic examination results should be documented prior to hysterectomy, and if abnormal uterine bleeding is present, the endometrium should be sampled and tested preoperatively. Finally, special consideration should be given to assessing whether or not concomitant cystocele, stress incontinence, enterocele, or rectocele is present. Before the hysterectomy is begun, a sterile preparation of the vulva and vagina is performed. An indwelling urinary catheter usually is not placed at this point, to allow the bladder to accumulate enough urine so that an inadvertent cystotomy may be recognized easily. Care should be taken in positioning and draping the patient to avoid compression of the femoral or peroneal nerve, but there should be adequate access to the operative site. An examination should be performed under anesthesia to, once again, assess whether or not any abnormalities exist that would make vaginal hysterectomy impossible and would indicate a switch to an abdominal approach. Technique of Vaginal Hysterectomy The elements of vaginal hysterectomy in the United States are derived from the operation described by Heaney ( Fig. 45.18). A circumscribing incision is made through the vaginal wall at the cervicovaginal junction (see Fig. 45.18A). Anteriorly, the incision is placed at the lower edge of the bladder to allow entry into the vesicocervical space. The lower edge of the bladder can be palpated against the cervix to determine where this incision should be made. In addition, the junction of the vaginal rugae and the smooth surface of the cervix can be used as landmarks to identify the proper site of incision. Some operators inject saline into the tissues around the cervix to facilitate dissection. This injection should not contain epinephrine, because its use increases the incidence of postoperative pelvic infection. Several studies have documented decreased blood loss after infiltration of a dilute vasopressin solution at the cervical portio, without concomitant increased infection rates. The incision should be made through the full thickness of the vaginal wall. This can be determined by placing a retractor anteriorly while pulling forcefully on the cervix so that the vaginal wall is placed under tension. Once the vagina has been transected it separates, exposing the underlying cleavage plane. On the posterior aspect of the cervix, this exposes the area of the cul-de-sac. Blunt dissection here frees the peritoneum, which can then be grasped and entered (see Fig. 45.18B). Care should be taken not to dissect the peritoneum off the uterus, because that makes entry more difficult. Once the cul-de-sac incision has been extended laterally as far as the medial margins of the uterosacral ligaments, it can be sutured to the posterior vaginal cuff to minimize bleeding. At this point, it is appropriate to palpate the adnexal structures through the colpotomy to detect unsuspected tumors and to make a final decision concerning the feasibility of the vaginal approach, if any doubt existed. This is an easy time to acquire this valuable information, and if unsuspected disease is found, an abdominal approach can be taken.

FIG. 45.18. A: Vaginal hysterectomy is initiated by circumscribing the cervix at the cervicovaginal junction. B: The posterior cul-de-sac is opened. C: The peritoneum of the anterior cul-de-sac is incised. D: The base of the uterosacral and cardinal ligament usually is clamped in two bites. E: The upper cardinal ligament is clamped prior to its transection. F: The uterine fundus is delivered, and the connections between the adnexal structures and uterine corpus are clamped. G: Ligaments and vaginal cuff as they appear after hysterectomy. The posterior cuff is whip-stitched (inset). H: Technique for resuspension of the vaginal cuff and obliteration of the cul-de-sac: 1, placement of suture through exteriorized ligaments and vaginal wall; 2, reefing sutures placed in peritoneum; 3, modified internal McCall suture; 4, high purse-string suture to close the cul-de-sac. (From Mattingly RF, Thompson JD, eds. Te Linde's operative gynecology, sixth ed. Philadelphia: JB Lippincott Co, 1985:554, with permission.)

Next, the long blade of an Auvrard retractor is placed in the cul-de-sac, and attention is turned to entering the anterior cul-de-sac. Dissection along the vesicocervical plane is best performed by lifting the fascia anteriorly while cutting with Mayo scissors placed tip-down on the cervix. This avoids the problems that sometimes occur when a knife is used, because the scissors held in this orientation will not cut into the dense fibrous tissue of the cervix. Once the cleavage plane of the vesicocervical space is reached, gentle dissection with a gloved finger is all that is needed to elevate the bladder off of the cervix. If the bladder pillars are prominent, they can be clamped next to the cervix, transected, and ligated to provide better lateral exposure. The location of the peritoneum may be detected by moving a finger from side to side and feeling the way in which the two slippery mesothelial surfaces slide on one anther. The peritoneum can then be lifted, transilluminated, and incised (see Fig. 45.18C). After the incision is extended laterally, a suture is placed in its vesical edge and held with a hemostat to facilitate its identification at the time of reperitonealization. Once the cul-de-sacs have been entered, the ureter should be palpated to determine its position prior to shortening and transecting the cardinal ligaments. This is a critical step, because only if the ligaments are shortened will the vagina be held at a level higher than that of the prolapsed cervix. A retractor is placed in the lateral fornix of the vagina and a finger in the anterior cul-de-sac. Because the ureter must pass under the uterine artery and end in the bladder anterior to the examining finger, it can be palpated against the retractor blade. The characteristic snap of the ureter permits its identification, and it can be followed for some distance by moving the retractor and examining finger anteriorly and posteriorly. After the ureter is identified the uterosacral and cardinal ligaments are shortened by clamping them at a safe distance from the ureter, somewhat lateral to the cervix (see Fig. 45.18D, Fig. 45.18E). After the ligaments are cut and suture-ligated, the sutures are left long and marked for use in resuspending the vagina at the end of the procedure. The uterine arteries are similarly clamped, transected, and suture-ligated, but they are not tagged, to avoid dislodging the suture from this vascular pedicle. At this point, the corpus of the uterus usually is brought through the posterior colpotomy by pulling on it with a tenaculum, to expose the connections between the adnexal structures and the uterus (see Fig. 45.18F). When this is not possible because of a small vagina or large uterus, the uterus can be divided along its sagittal plane, and one half can be pushed up into the peritoneal cavity. A long tenaculum is left attached to the elevated half of the uterus to allow its later retrieval, while the contralateral adnexal pedicles are dealt with. Clamps are placed across the adnexal structures, with care taken not to include bowel or adjacent structures, and the pedicles are transected. Next, consideration should be given to removal of the ovaries. If the ovaries are to be removed, the mesovarium or infundibulopelvic ligament is clamped, the ovary is removed with or without the fallopian tube, and the remaining pedicle is suture-ligated. Management of the Cul-de-sac and Suspension of the Vaginal Apex Once the uterus has been removed and the ligaments have been prepared by shortening them, the reconstructive phase of the operation begins. These steps use the ligaments that were tagged during the hysterectomy (see Fig. 45.18G). The goals are to close the peritoneum and suspend the vaginal apex, while also correcting any existing enlargement of the cul-de-sac. The cul-de-sac peritoneum is attached to the uterus. Whenever the uterus prolapses, it pulls the cul-de-sac peritoneum with it. If this extension is not distended with bowel, it is referred to as a traction enterocele, and when the cul-de-sac is not only elongated but also distended with bowel, it is referred to as a pulsion enterocele, a true enterocele. When little prolapse has been present (i.e., the cervix does not extend below the hymenal ring), simple closure of the cul-de-sac and attachment of the ligaments to the vaginal cuff suffice. An indwelling catheter usually is placed at this time to drain the bladder. This brings the anterior peritoneum closer to the surgeon and, thereby, facilitates its closure. Closure of the vaginal cuff should accomplish two things: resuspension of the vaginal apex and closure of the cul-de-sac (see Fig. 45.18H). A suture is placed through the vaginal cuff and through the ends of the suspensory ligaments to reattach them to the vagina. Next, the cul-de-sac may be reefed and a suture placed through the uterosacral ligaments to bring them to the midline. Finally, the peritoneum is closed with a purse-string suture of 0 silk placed at the highest level possible, beginning on

the vesical peritoneum. A finger is placed in the peritoneum while this suture is tied, to avoid trapping intraabdominal contents in the suture, and is removed as the suture is snugged down. Anterior Colporrhaphy After the uterus has been removed and before the ligaments are reattached to the vaginal cuff, an anterior colporrhaphy is performed for patients with cystoceles ( Fig. 45.19). Scissors are used to undermine and incise the mucosa of the anterior vaginal wall as far as the midurethra (see Fig. 45.19A). The vaginal fascia is then mobilized widely from the vaginal epithelium, with use of a combination of sharp and blunt dissection, until the superior lateral sulcus of the vagina is reached (see Fig. 45.19B). Care should be taken to remain superficial in this dissection to avoid transecting the fascia that is to be plicated in the midline. Special attention should be given to adequately developing the fascia that lies under the urethrovesical junction, because this is critical to improving urethral support. Once mobilization has been accomplished, a series of plication sutures are taken in the fascia, just medial to the junction of the anterior and lateral vaginal walls (see Fig. 45.19C).

FIG. 45.19. A: The vaginal epithelium is undermined when commencing anterior colporrhaphy. B: The fascia is separated from the mucosa. C: The ascia is sutured together in the midline. Special care should be taken to ensure adequate support at the urethrovesical junction. D: Excess epithelium is trimmed. E: The epithelium is closed. A bite of the underlying fascia is included to minimize dead space. (From Mattingly RF, Thompson JD, eds. Te Linde's operative gynecology, sixth ed. Philadelphia: JB Lippincott Co, 1985:612, with permission.)

Plication of the suburethral fascia deserves special attention. Once the first suburethral suture has been placed and tied, it pulls some of the more lateral fascia into view, and this offers an opportunity to place a second suture in this same region and improve the elevation that was possible with the first suture. Once the remainder of the fascia has been closed, the cranial edge of the fascia at the vaginal apex should be tagged so that it may be included in the sutures that reattach it to the cardinal and uterosacral ligaments. This step will help elevate the upper margin of the fascia and pull it higher in the pelvis. An appropriate amount of vaginal epithelium is then trimmed (see Fig. 45.19D), and the epithelium is closed with running or interrupted sutures (see Fig. 45.19E). Posterior Colpoperineorrhaphy Next, a rectocele, if present, must be corrected. Although often undertaken at the end of a tiring operation, this repair requires the most skill and judgment of all phases of surgery for prolapse. Overcorrection of the posterior wall may leave the patient unable to have satisfying intercourse, yet a failure to reinforce the supportive position of the posterior vaginal wall may increase the likelihood that prolapse will recur. The goal of this portion of the operation is to reinforce the rectovaginal septum and to return the size of the introitus to normal. The posterior colporrhaphy is begun by establishing the size that the introitus should be at the end of the procedure. This is accomplished by estimating the amount of vaginal and perineal skin that must be removed to return the introitus to a normal diameter. Allis forceps are placed to grasp a solid bite of tissue at the hymenal ring bilaterally at the margins of the mucosa that is to be excised. These forceps are then approximated in the midline to bring the soon-to-be-united parts of the introitus together. This allows a preview of the degree to which the introitus will be narrowed. Adjustments may then be made in the placement of the forceps to obtain an optimal result before an irrevocable incision is made into the perineum. A patient with a normal introitus who has a rectocele above this level need not have any tissue excised, and a vertical incision can be made in the introitus. After the skin of the introitus is incised, the mucosa is undermined ( Fig. 45.20A). The vagina and perineal body are fused, so dissection in the 3 cm inside the hymenal ring must be done sharply. If the rectovaginal septum is thin, placing a finger in the rectum to guide dissection will minimize the likelihood of proctotomy.

FIG. 45.20. A: To begin posterior colporrhaphy, a transverse incision is made in the perineal body, and a vertical incision is extended toward the vaginal apex. B: The perirectal fascia is mobilized from the mucosa. C: The perirectal fascia is sutured, beginning above the site of the rectocele. D: The inner surface of the levator ani muscles are approximated in the perineal body, and redundant mucosa is excised. E: The mucosa and perineal skin are closed. (From Mattingly RF, Thompson JD, eds. Te Linde's operative gynecology, sixth ed. Philadelphia: JB Lippincott Co, 1985:578, with permission.)

The rectovaginal space begins at the top of the perineal body and lies between the rectovaginal fascia and the rectum. It is a loose areolar plane and dissects easily. Once it is entered, it should be developed laterally (see Fig. 45.20B) until the inferior lateral sulcus of the vagina has been reached and the inner surfaces of the levator ani muscles can be felt. The upper limit of the dissection depends on the location of the rectocele and should extend, at least, to a point several centimeters above the upper limit of the bulge. When a posterior repair is used in a patient who previously had a hysterectomy, the upper vaginal dissection should include a search for an enterocele. Once the rectovaginal space has been developed adequately, the repair may begin. With a finger depressing the rectum, the connective tissue lateral to the rectum is grasped with a needle and elevated, and the needle is retrieved (see Fig. 45.20C). A similar bite of the endopelvic fascia on the contralateral side is then taken. Successive sutures are then placed from that point to the introitus. After this is complete, it is prudent to make sure that no ridge has formed. If it has, the offending suture should be removed and replaced to avoid distortion. After this first layer has been placed, the space between the levator ani muscles may be narrowed by suturing their fascial covering, and the perineal body can be reconstructed (see Fig. 45.20D). This requires a second layer of sutures in the distal 3 cm of the vagina. After these sutures are placed, one final stitch usually is placed at the level of the hymenal ring to establish the triangular nature of the perineal body when seen in sagittal section. Finally, an appropriate amount of vaginal mucosa is trimmed, and the vaginal edges are reapproximated with interrupted or running absorbable sutures

(see Fig. 45.20E). When the operation is finished, the adequacy of the vagina for intercourse should be assessed. Despite precautions to avoid vaginal stenosis, it sometimes is noted after the repair is complete. If so, relaxing incisions may be made in the vaginal wall to relieve the stricture, or part of the repair may be taken down and resutured. Time rarely improves a bad initial repair, and correction in the operating room is far superior to taking the patient back to the operating room later for revision. Le Fort Partial Colpocleisis In elderly patients who will no longer be engaging in intercourse, a colpocleisis may be used to cure the prolapse. This operation carries less operative and postoperative morbidity and may be performed under local anesthesia in the rare patient whose medical condition puts her at too great a risk to have a regional anesthetic. It consists of denuding a rectangle of anterior and posterior vagina and sewing them together ( Fig. 45.21). This prevents the uterus and vagina from prolapsing. An adjunctive high perineorrhaphy at the time of the colpocleisis forms a shelf for the repair to rest on, so as to lessen the likelihood of prolapse recurring. Channels are left below the cervix and lateral to the closure to provide egress of cervical secretions and blood, should postmenopausal bleeding occur. Because the uterus is not removed and access to the cervix is blocked, care should be exercised in the preoperative evaluation to make sure that endometrial carcinoma is not present or likely to develop. Intraoperative dilation and curettage should be considered, as well, to exclude carcinoma.

FIG. 45.21. A–C: In Le Fort colpocleisis, rectangles of vaginal mucosa are removed from the anterior and posterior vaginal walls. D,E: The denuded areas are then sutured together, leaving (F) channels on each side open. (From Mattingly RF, Thompson JD, eds. Te Linde's operative gynecology, sixth ed. Philadelphia: JB Lippincott Co, 1985:562, with permission.)

Stress incontinence may follow colpocleisis if there is poor urethral support associated with the prolapse. In this instance, steps should be taken during the operation to improve the situation. Preoperatively, occult stress incontinence may be detected by examining the patient with a full bladder and having her cough with the prolapse reduced. If stress incontinence is elicited, steps must be taken to improve urethral support by plicating the endopelvic fascia under the urethra or considering an incontinence procedure such as a postoperative collagen instillation, a pubovaginal sling, or a tension-free vaginal tape procedure. Vaginal Repair of the Posthysterectomy Enterocele The description earlier in this chapter concerning care of an enterocele and the cul-de-sac involves management of an enterocele encountered with the uterus in situ during vaginal hysterectomy. Because enterocele after hysterectomy is a significant problem in its own right, it will be considered separately here. Certain features of enterocele, such as its occurrence after high retropubic urethral suspension, warrant special consideration. Repair of an enterocele should not only obliterate the cul-de-sac but also should pull the vagina over the rectum by shortening the ligamentous suspension of the vagina, so that the normal closure of this space may be maintained. In addition, if the vaginal wall has been stretched because of its protrusion below the pelvic floor, the excess vaginal wall must be excised. Although transabdominal enterocele repairs (e.g., Moschcowitz, Halban) are useful as prophylaxis at the time of abdominal repair, an enterocele that appears after hysterectomy can almost always be repaired vaginally. This is the preferred method, because it avoids the morbidity of an abdominal incision and is much easier to perform. The vaginal wall is opened over the enterocele sac. If a concomitant rectocele is present, this may be done by extending the incision made for the posterior repair up to the apex of the vagina. If a pure enterocele is present, the incision is made directly over the enterocele. In instances where the prolapsed vaginal wall has become stretched, a diamond-shaped incision may be made to excise the excess vaginal skin ( Fig. 45.22A). Once identified, the peritoneum is separated from the vaginal wall (see Fig. 45.22B), and the excess peritoneum is excised. Next, the remnants of the uterosacral ligaments are brought together under the peritoneum and attached to the vaginal apex to improve its support (see Fig. 45.22C). At that point, any excess vaginal wall is trimmed, a posterior repair performed if needed, and the vaginal incision closed. Most enterocele repairs should be accompanied by a posterior colpoperineorrhaphy.

FIG. 45.22. A: The enterocele is exposed, and a diamond-shaped piece of vaginal wall is excised to reduce the size of the stretched upper vagina. B: The peritoneum is dissected from the surrounding structures, and it is closed with a purse-string suture. C: The uterosacral ligaments are defined by creating opposing traction with a retractor and suture to complete repair, after which the vaginal wall is closed. (From Segala CJ. New technique for the repair of vaginal vault prolapse following hysterectomy. Int Surg 1969;51:36, with permission.)

Posthysterectomy Prolapse Prolapse subsequent to hysterectomy deserves special consideration. One third to one half of women will have had a hysterectomy by the age of 65, so prolapse in older women often will occur in the absence of a uterus. Although it might appear that these individuals would have only a cystocele or rectocele, because there could be no uterine prolapse, prolapse of the vaginal apex is often present. Apical prolapse must be recognized; otherwise, repair of a cystocele or a rectocele associated with prolapse of the vaginal apex is doomed to failure. Additionally, without recognition of the apical prolapse, the surgeon is likely to excise too much vaginal tissue with anterior and posterior colporrhaphy, leaving the patient with a very shortened, narrow vagina. When the apex of the vagina, marked by the scar at the former location of the cervix, prolapses below the hymenal ring, vaginal prolapse is present. This usually is accompanied by an enterocele and often is seen in combination with a cystocele or a rectocele. To cure these patients, the apex of the vagina must be returned to and held in its normal position in the pelvis, and the cystocele or rectocele must be repaired. This may be accomplished by suturing the vaginal apex to the sacrospinous

ligament ( Fig. 45.23A) or to the sacrum, using an interposing graft in an operation known as abdominal sacral colpopexy ( Fig. 45.23B), or by performing a high uterosacral suspension. Each of these operations is accompanied by repair of the enterocele and whatever other prolapse must be addressed. In women who are no longer interested in preserving coital function, a complete colpocleisis obliterates the vagina and prevents prolapse from occurring, and this provides an easy, safe, and effective technique to relieve the symptoms.

FIG. 45.23. A: The vagina is fixed to the sacrospinous ligament. (From Morely GW, DeLancey JOL. Sacrospinous ligament fixation for eversion of the vagina. Am J Obstet Gynecol 1988;158:872, with permission.) B: The vaginal apex is attached to the sacrum with an intervening graft. (From Addison WA, Timmons MC, Wall LL, et al. Failed abdominal sacral colpopexy: observations and recommendations. Obstet Gynecol 1989;74:480, with permission.)

Paravaginal Repair When a cystourethrocele persists or returns soon after anterior colporrhaphy, this often means that the loss of anterior vaginal wall support is due to a paravaginal separation of the pubocervical fascia from its normal lateral attachment to the arcus tendineus rather than to a midline defect in the fascia. In this instance, repeating the anterior colporrhaphy will not be effective, and repair of the separation will be required. Although the paravaginal repair usually is described as an operation to treat stress incontinence associated with cystourethrocele, it also is helpful in treating cystourethrocele in the absence of stress urinary incontinence, especially when an anterior colporrhaphy has failed. This retropubic operation reattaches the lateral margin of the pubocervical fascia to the pelvic sidewall along a line from the undersurface of the pubic symphysis to the ischial spine, along the line of the arcus tendineus. This reestablishes the normal lateral attachments of the pubocervical fascia and, therefore, prevents downward displacement of the bladder. It has not yet been established as a necessary procedure in the primary treatment of cystourethrocele without stress incontinence but, in instances of recurrent cystocele after anterior repair, simply repeating an operation that has not helped the patient is not desirable. Conservative Measures Vaginal pessaries are perhaps the oldest effective treatment for prolapse. Named after an oval stone (i.e., a pessos) used in certain Greek games, innumerable natural and man-made materials have been used to make pessaries. These objects are placed in the vagina. Because they are larger than the introitus, they are retained in the vagina above the pelvic floor musculature and thereby prevent the somewhat smaller uterine cervix from passing through the opening. When the pelvic muscles are relatively normal, these devices work well. When the pelvic floor opening is enlarged because of damage to the levator ani, a pessary may not be retained. There are countless varieties of pessaries. The ones most commonly used for prolapse are the doughnut-shaped pessaries, either permanently inflated or inflatable. The largest pessary that the vagina will accommodate without undue pressure on the vaginal walls is inserted. Often, patients can be taught how to place and remove them. Having the patient remove the pessary at night minimizes the discharge inevitably associated with its use. Adjunctive treatment with topical or systemic estrogen helps the vaginal mucosa tolerate the foreign body. Because pessaries may cause erosion and ulceration, they should be checked periodically. If an ulcer is developing, pessary use should be discontinued or a smaller one fitted, and the ulcer should be monitored until it heals.

SUMMARY Pelvic organ prolapse affects between 5% and 10% of women. It arises due to damage to the levator ani muscles and endopelvic fascia that occurs at the time of vaginal birth and the deterioration that occurs in these tissues with advancing age. Women who have these conditions express distress from the constant, unrelenting feelings of pressure and incontinence and from fear that this protrusion may lead to other problems. The term pelvic organ prolapse encompasses a variety of related conditions that arise when the muscles and connective tissues of the pelvic floor can no longer hold the pelvic organs in their normal positions. The location and nature of the muscle injury and fascial ruptures determine which pelvic organ falls downward. The symptoms experienced by a woman come from this mechanical misalignment, along with the functional derangements of the pelvic viscera determined by neurologic injury and these prolapses. Identifying both the location of the abnormality and the symptoms experienced by the individual woman are critical to successful treatment. Surgical treatment is aimed at resuspending the prolapsed pelvic organs and rearranging the damaged muscles to place them in the most favorable locations for function. This field is characterized by a great diversity of operative procedures used to correct individual elements of defective pelvic organ support. The particular operation that is chosen depends on the anatomic and functional abnormalities that a woman experiences, as well as the training of the surgeon. Individual surgeons become familiar with a subset of these techniques that allows them to address the problems they face. No two surgeons treat these problems in exactly the same way, and a good surgeon rarely will do two operations that are exactly alike.

SUMMARY POINTS Pelvic organ prolapse is a remarkably common and distressing condition. Vaginal delivery and advancing age are the most common factors causing pelvic organ prolapse. Pelvic organ prolapse arises due to a combination of neuromuscular and connective tissue injury and has variable symptoms associated with its development. Evaluating the status of each element of pelvic organ support is critical prior to instituting treatment. Surgical treatment of pelvic organ prolapse must be tailored to each woman's condition, so that all areas of pelvic support loss are corrected. SUGGESTED READINGS Epidemiology of Surgically Managed Pelvic Organ Prolapse and Urinary Incontinence Mant J, Painter R, Vessey M. Epidemiology of genital prolapse: observations from the Oxford Family Planning Association study. Br J Obstet Gynaecol 1997;104:579–585. Olsen AL, Smith VJ, Bergstrom JO, et al. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 1997;89:501–506. Timonen S, Nuoranne E, Meyer B. Genital prolapse: etiological factors. Ann Chir Gynaecol Fenn 1968;57:363–370.

Pelvic Floor and the Nature of the Genital Prolapse Emge LA, Durfee RB. Pelvic organ prolapse: four thousand years of treatment. Clin Obstet Gynecol 1966;9:997. Richardson AC, Lyons JB, Williams NL. A new look at pelvic relaxation. Am J Obstet Gynecol 1976;126:568–573.

Basic Anatomy and Pathophysiology of the Pelvic Floor Campbell RM. The anatomy and histology of the sacrouterine ligaments. Am J Obstet Gynecol 1950;59:1. DeLancey JOL. Structural anatomy of the posterior compartment as it relates to rectocele. Am J Obstet Gynecol 1999;180:815–823. DeLancey JOL. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol 1994;170:1713–1720. DeLancey JOL. Anatomic aspects of vaginal eversion after hysterectomy. Am J Obstet Gynecol 1992;166:1717.

Kuhn RJP, Hollyock VE. Observations on the anatomy of the rectovaginal pouch and septum. Obstet Gynecol 1982;59:445. Range RL, Woodburne RT. The gross and microscopic anatomy of the transverse cervical ligaments. Am J Obstet Gynecol 1964;90:460.

Issues Relating to Incontinence DeLancey JOL. Structural aspects of the extrinsic continence mechanism. Obstet Gynecol 1988;72:296. McGuire EJ. Urodynamic findings in patients after failure of stress incontinence operations. Prog Clin Biol Res 1981;78:351. Richardson AC, Edmonds PB, Williams NL. Treatment of stress urinary incontinence due to paravaginal fascial defect. Obstet Gynecol 1981;57:357–362. Smith ARB, Hosker GL, Warrell DW. The role of pudendal nerve damage in the aetiology of genuine stress incontinence in women. Br J Obstet Gynaecol 1989;96:29.

Causes of Pelvic Floor Damage Bump RC, Norton PA. Epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin North Am 1998;25:723–746.

Diagnosis and Classification, Evaluating Individual Elements of Support and Pelvic Organ Prolapse Classification Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–17. Weidner AC, Bump RC. Terminology of pelvic organ prolapse. Curr Opin Obstet Gynecol 1997;9:309–312.

Symptoms Barber MD, Visco AG, Wyman JF, et al. Continence Program for Women Research Group. Sexual function in women with urinary incontinence and pelvic organ prolapse. Obstet Gynecol 2002;99:281–289. Ellerkmann RM, Cundiff GW, Melick CF, et al. Correlation of symptoms with location and severity of pelvic organ prolapse. Am J Obstet Gynecol 2001:185:1332–1337.

Surgical Procedures Addison WA, Livengood CH, Sutton GP, et al. Abdominal sacral colpopexy with Mersilene mesh in the retroperitoneal position in the management of posthysterectomy vaginal vault prolapse and enterocele. Am J Obstet Gynecol 1985;153:140. Addison WA, Timmons MC, Wall LL, et al. Failed abdominal sacral colpopexy: observations and recommendations. Obstet Gynecol 1989;74:480. Burch JC. Cooper's ligament urethrovesical suspension for stress incontinence. Am J Obstet Gynecol 1968;100:764. Kammerer-Doak DN, Rogers RG, Johnson Maybach J, et al. Vasopressin as an etiologic factor for infection in gynecologic surgery: a randomized double-blind placebo-controlled trial. Am J Obstet Gynecol 2001;185:1344–1347. Heaney NS. Report of 565 vaginal hysterectomies performed for benign pelvic disease. Am J Obstet Gynecol 1934;28:751. Heaney NS. Vaginal hysterectomy: its indications and technique. Am J Surg 1940;48:284. Lee RA. Atlas of gynecologic surgery. Philadelphia: WB Saunders Co, 1992. Morley GW, DeLancey JOL. Sacrospinous ligament fixation for eversion of the vagina. Am J Obstet Gynecol 1988;158:872. Nichols DH, Randall CL. Vaginal surgery, fourth ed. Baltimore: Williams & Wilkins, 1996. Nichols DH. Sacrospinous fixation for massive eversion of the vagina. Am J Obstet Gynecol 1982;142:901. Sheth SS. The place of oophorectomy at vaginal hysterectomy. Br J Obstet Gynaecol 1991;98:662. Symmonds RE, Williams TJ, Lee RA, et al. Post-hysterectomy enterocele and vaginal vault prolapse. Am J Obstet Gynecol 1981;140:852.

Chapter 46 Laparoscopic Surgery Danforth’s Obstetrics and Gynecology

Chapter 46 Joseph S. Sanfilippo and Lisa M. Roberts

Laparoscopic Surgery

LAPAROSCOPY Patient Positioning Equipment Physiologic Effects of the Pneumoperitoneum Anatomy of the Anterior Abdominal Wall Obtaining Intraabdominal Access Laparoscopic Procedures Complications SUMMARY POINTS SUGGESTED READINGS Introduction Laparoscopy Complications

In one sense, the historical perspective of endoscopic surgery begins in 1847, with Sir James Y. Simpson of Edinburgh, Scotland, introducing chloroform narcosis. This advance was followed by the first endoscopic evaluation of the abdominal cavity, using a dog model in 1901, by Kelling. Ten years later, Jacobaeus reported the first laparoscopic procedure in the human. Cold illumination fiberoptics followed, by Semm in 1963, and then the surgical area rapidly evolved. The first laparoscopic appendectomy was performed in 1980 and the first cholecystectomy in 1985. This relatively new aspect of endoscopic surgery has far-reaching and ever increasing innovative aspects. Hysteroscopy has altered the approach to the myomatous uterus. Laparoscopy has blossomed to include radical hysterectomies and node sampling. This chapter provides the technical advances in a segment of gynecologic surgery that appears to be in a continuous state of evolution and ever more far-reaching new instrumentation.

LAPAROSCOPY Patient Positioning Proper patient positioning is paramount. The surgeon should position the patient so as to prevent any excessive pressure on the lower extremities. Nerve injuries can result from improper placement of the lower extremities in stirrups. Lithotomy position with access to steep Trendelenburg is appropriate. The physician should discuss with anesthesia personnel decisions regarding whether the arms should be extended or “tucked” along the sides of the patient. The latter requires particular attention to prevent any trauma to the hands and fingers when the table is maneuvered. With the patient in the low lithotomy position and the legs supported in stirrups, the buttocks should protrude slightly from the lower edge of the table. The lateral aspect of the knee should be protected with padding in the stirrup to prevent peroneal nerve injury. The knees should be kept in slight flexion to minimize stretching of the sciatic nerves and to provide increased stability in the Trendelenburg position. With respect to positioning of the arms, care should be taken not to stretch or traumatize the brachial plexus. It has been advocated that an angle of 145 degrees between the abdomen and the lower extremity (thigh) is ideal, providing the surgeon with adequate space for instrumentation. The typical patient positioning for endoscopic surgery is noted in Figure 46.1.

FIG. 46.1. Patient positioning for laparoscopic surgery.

Equipment An array of instruments designed to facilitate operative laparoscopic procedures continues to evolve. This section describes a number of instruments. Viewing System The video equipment should include a three-chip camera, a processor, a 300-watt xenon light source with fiberoptic cable, a high-resolution monitor, and a video recorder. The three-chip camera provides a sharper, brighter image with higher resolution than the older single-chip cameras ( Fig. 46.2). Three-chip cameras can deliver over 600 lines of resolution; therefore, the video monitor's horizontal resolution should be greater to maximize image quality. Fiber light cables should be inspected for broken fibers, because damaged fibers will result in suboptimal light delivery.

FIG. 46.2. Autoclavable Goldtip videolaparoscope with camera processor and xenon light source (Olympus, Melville, NY). Table An operating room table that allows 30 degrees of flexion (Trendelenburg position) is ideal for visualization of the deep pelvis. Shoulder braces (Stierlen-Maquet Shoulder Braces, Siemens Medical Systems, Englewood, CO) placed at the acromioclavicular joints and the arms placed at the patient's sides will minimize nerve injuries. Insufflator For most procedures, the pneumoperitoneum may be maintained with an insufflator that flows at a rate of 2 to 7 L/min. High-flow insufflators that achieve up to 30 L/min are available and will maintain the pneumoperitoneum during procedures that allow escape of large amounts of CO 2 , such as during tissue morcellation. Carbon Dioxide Warmer There is evidence that warmed CO 2 is less irritating to the peritoneal surface and, therefore, may cause less adhesion formation. CO 2 warmers are commercially available. Laparoscopes Laparoscopes are telescopes that vary in size from 2 to 10 mm. Zero-degree, 30-degree, 45-degree, and 70-degree viewing angles are available. Diagnostic laparoscopes do not contain an operating channel. Operative laparoscopes contain a channel through which instruments or laser may be passed, but they have a smaller number of optic fibers ( Fig. 46.3 and Fig. 6.4).

FIG. 46.3. Five-millimeter laparoscopes (Karl Storz Endoscopy-America, Culver City, CA).

FIG. 46.4. Ten-millimeter laparoscope (Karl Storz Endoscopy-America, Culver City, CA). Trocars Trocars come in a variety of sizes, materials, and designs. Sizes range from 3 to 15 mm. Reusable metal trocars have the advantage of cost efficiency, but they may actually be more dangerous if the tips are not kept sharp ( Fig. 46.5, Fig. 46.6 and Fig. 46.7). Disposable trocars are more expensive but are always sharp, which may decrease the risk of injury during insertion ( Fig. 46.8). Some trocars are designed to stretch the fascia so that fascial closure is not necessary. Others have been designed to allow visualization of tissue layer separation with the laparoscope during insertion. Advocates assert that this type of trocar decreases bowel and vascular injuries. There are no objective data supportive of a reduction in risk for any trocar. The trocar chosen should be based on the surgeon's preference. Hasson-type trocars have a blunt end for placement in open laparoscopy ( Fig. 46.9 and Fig. 46.10).

FIG. 46.5. Five- and ten-millimeter trocars (Karl Storz Endoscopy-America, Culver City, CA).

FIG. 46.6. Autoclavable 5-mm slim trocar with duckbill (Olympus, Melville NY).

FIG. 46.7. Five- and ten-millimeter reusable trocars.

FIG. 46.8. Ten-millimeter disposable trocar (Olympus, Melville, NY).

FIG. 46.9. Ten-millimeter blunt trocar for open laparoscopy (Marlow Surgical Technologies, Inc., Willoughby, OH).

FIG. 46.10. Ten-millimeter blunt trocar for open laparoscopy (Apple Medical, Bolton, MA). Operating Instruments A standard laparoscopy set should include the following: Trocars Bipolar forceps ( Fig. 46.11 and Fig. 46.12)

FIG. 46.11. Five-millimeter bipolar forceps (Olympus, Melville, NY).

FIG. 46.12. Five-millimeter bipolar forceps (Olympus, Melville, NY). Atraumatic grasping forceps ( Fig. 46.13, Fig. 46.14 and Fig. 46.15)

FIG. 46.13. Atraumatic tissue grasper (Applied Medical, Rancho Santa Margarita, CA).

FIG. 46.14. Atraumatic tissue grasper.

FIG. 46.15. Atraumatic tissue grasper. Biopsy forceps ( Fig. 46.16)

FIG. 46.16. Biopsy forceps. Blunt probe Blunt sawtooth scissors ( Fig. 46.17)

FIG. 46.17. Blunt sawtooth scissors (Richard Wolf Medical Instruments, Vernon Hills, IL).

Pointed Metzenbaum scissors ( Fig. 46.18)

FIG. 46.18. Metzenbaum scissors (Richard Wolf Medical Instruments, Vernon Hills, IL). Cyst aspiration needle ( Fig. 46.19)

FIG. 46.19. Cyst aspiration needle. Suction and irrigator Uterine manipulator ( Fig. 46.20 and Fig. 46.21)

FIG. 46.20. RUMI uterine manipulator (Cooper Surgical, Shelton, CT).

FIG. 46.21. Hulka uterine manipulator tenaculum (Richard Wolf Medical Instruments, Vernon Hills, IL) and Cohen Uterine Cannula (Karl Storz Endoscopy-America, Culver City, CA). An advanced laparoscopy set may also include: Needle holders ( Fig. 46.22, Fig. 46.23 and Fig. 46.24)

FIG. 46.22. Five-millimeter needle holder (Olympus, Melville, NY).

FIG. 46.23. Five-millimeter needle holder (Olympus, Melville, NY).

FIG. 46.24. Five-millimeter needle holder (Karl Storz Endoscopy-America, Culver City, CA).

Knot pusher ( Fig. 46.25)

FIG. 46.25. Knot pusher for extracorporeal suturing (Marlow Surgical, Roebling, NJ). Vaginal delineator ( Fig. 46.26)

FIG. 46.26. Koh Cup Vaginal Fornices Delineator (Cooper Surgical, Shelton, CT). Babcock clamp ( Fig. 46.27)

FIG. 46.27. Babcock atraumatic grasper. lAlis clamp Adson forceps ( Fig. 46.28)

FIG. 46.28. Adson forceps. Corkscrew Single-tooth tenaculum Atraumatic bowel grasper Monopolar spatula ( Fig. 46.29)

FIG. 46.29. Monopolar spatula for cutting and coagulation. Endoscopic specimen retrieval bags ( Fig. 46.30 and Fig. 46.31)

FIG. 46.30. Endo Catch instrument with a specimen bag (U.S. Surgical, Norwalk, CT).

FIG. 46.31. Specimen retrieval bag (Cook Ob/Gyn, Spencer, IN). Endoscopic suture ligatures ( Fig. 46.32)

FIG. 46.32. Surgitie endoscopic suture ligature (U.S. Surgical, Norwalk, CT). Microbipolar forceps Morcellator Rectal probe Staplers Endoscopic staplers that simultaneously ligate and divide tissue are available. In gynecologic procedures, they are used most commonly on the infundibulopelvic ligament, round ligament, fallopian tubes, and uteroovarian ligaments in laparoscopically assisted vaginal hysterectomies. They also may be used when excising endometriosis from the bowel ( Fig. 46.33).

FIG. 46.33. Stapling device (U.S. Surgical, Norwalk, CT). Harmonic Scalpel The active blade of the Harmonic Scalpel (Ethicon Endosurgery, Cincinnati, OH) vibrates at a rate of 55,000 cycles per second, resulting in coagulation and tissue separation. The scalpel is available as a 5- or 10-mm “hook,” “ball” dissector, spatula, or opposing jaws. Although electrical energy is not used, the tissue still becomes heated and a 2-mm lateral energy spread, which can cause thermal injury, may be seen. Electrosurgery Unipolar electrosurgical instruments are used to cut and coagulate tissue. Cutting occurs when there is sufficient voltage (at least 200 volts) between the electrode and the tissue to produce an electric arc. This arc concentrates the current to points along the tissue, resulting in a cutting effect. In contrast to cutting current, coagulation is produced through instrument contact with the tissue. Contact allows heating of the tissue followed by irreversible cellular damage, evaporation of intracellular water, and contraction of blood vessels and surrounding tissue. Electrosurgical burns may occur due to insulation failure, direct coupling (activated electrode makes unintended contact with another metal object in the area of the surgical field), or capacitive coupling (induction of stray current to a surrounding conductor through the intact insulation of an active electrode). Bipolar electrosurgical instruments contain the electrical current between an active and return electrode, usually the two blades of forceps. The flow of alternating current is passed between the two electrodes rather than passing through the patient to a grounding pad. This eliminates the risk of capacitive coupling and stray current. Laser Lasers offer an alternative method of cutting and vaporizing tissue. The CO 2 laser with a depth of penetration of 1 mm allows the surgeon some security when working around bowel, ureters, and large blood vessels. In addition, a CO 2 laser will not traverse through water, thus irrigation fluid may be used as a backstop. The Coherent 5000L (Coherent Laser, Palo Alto, CA) laser offers high-power density over a short period of time, minimizing thermal damage to surrounding tissue. This laser uses a 13C isotope of CO 2 with an 11.1-mm–wavelength beam. The wavelength of the CO 2 purge gas in the operating channel of the laparoscope is 10.6 mm. Because the wavelengths are different, there is no absorption of the laser beam by the purge gas, keeping the power density 10 times more than at similar settings with other standard laser beams with 10.6-mm wavelengths. The 1.5-mm spot size is maintained at all power settings, offering precision, minimal surrounding tissue damage, and less charring. Fiber lasers (potassium-titanyl-phosphate [KTP], argon, and Nd:YAG [YAG]) are introduced through small channels of the operating laparoscope or through ancillary trocars. The KTP and argon lasers may be used for cutting and coagulating. As the tip of the fiber approaches the tissue, the power density increases and the laser is used for cutting. As the tip is moved away from the tissue, the spot size increases and power density decreases rapidly. The clinical applications for the YAG laser are more limited. The YAG laser coagulates well, but it doesn't cut well unless a sapphire tip is used to increase the power density. Handoscopy Hand-assisted laparoscopy, or “handoscopy,” has become popular, mainly in the field of solid organ and bowel surgery. To date, the only gynecologic procedure where this device has been used is rectal resection for deep fibrotic endometriosis. The main advantage of handoscopy is that it allows the surgeon to regain the tactile feel of tissues lost in standard laparoscopies. Physiologic Effects of the Pneumoperitoneum The respiratory, hemodynamic, and renal consequences of the pneumoperitoneum are substantial but usually are well tolerated by the healthy patient. These changes may be severely problematic in the patient whose cardiopulmonary or renal system is compromised. Respiratory Changes Ventilation during laparoscopic surgery is altered by the increased intraabdominal pressure (IAP) of the pneumoperitoneum. The airway pressure required for adequate ventilation consequently is increased. Thoracic compliance, diaphragmatic excursion, and functional residual capacity are all decreased. Studies have shown that the increase in IAP alone reduces compliance by 30%; this may be decreased by another 20% when the patient is placed in a Trendelenburg position. In addition to mechanical changes, the CO 2 pneumoperitoneum produces changes in acid–base balance. Because carbon dioxide is absorbed rapidly from the peritoneal cavity, the minute volume must be increased as indicated by the end-tidal CO 2 to prevent respiratory acidosis ( Table 46.1).

TABLE 46.1. Ventilatory effects of carbon dioxide pneumoperitoneum Hemodynamic Changes Circulatory changes seen during laparoscopic surgery are the result of the mechanical compressive effects of the increased IAP and its associated hormonal changes ( Table 46.2). Catecholamines, angiotensin, and vasopressin are all increased, resulting in increased systemic vascular resistance. An increase in systemic vascular resistance results in an increase in mean arterial pressure and a decrease in cardiac index. The increased IAP also causes increased intrathoracic pressure, with resultant increases in central venous pressure, pulmonary capillary wedge pressure, and pulmonary vascular resistance.

TABLE 46.2. Hemodynamic effects of carbon dioxide pneumoperitoneum Renal Blood Flow Changes Animal studies have shown an increase in renal vascular resistance and a decrease in renal perfusion and glomerular filtration rate secondary to increased IAP. A subsequent decrease in urine output may be observed and may continue to remain low 1 hour after release of the pneumoperitoneum ( Table 46.3).

TABLE 46.3. Renal effects of carbon dioxide pneumoperitoneum Compromised Patient The multiple cardiopulmonary and renal changes observed during CO 2 laparoscopy are well tolerated by the healthy patient. Extreme caution must be taken when performing gaseous laparoscopy in patients with compromised respiratory function, ischemic heart disease, congestive heart failure, and renal dysfunction. These compromised states may even be considered relative contraindications to gaseous laparoscopy. The lowest IAP that allows adequate visualization should be used to minimize increases in intrathoracic pressure. An alternative consideration would be to use an abdominal wall lift method (elevating the ventral abdominal wall by an inflatable balloon or fan retractor), eliminating the consequences of CO 2 in the peritoneum. The use of alternative gases such as argon, nitrous oxide, and helium for pneumoperitoneum has been considered. Carbon dioxide's advantages over these gases include higher solubility and, therefore, possibly fewer deleterious effects in event of gas embolism, ability to use cautery in its presence, and it is inexpensive. It is the surgeon's responsibility to take a careful history and perform appropriate preoperative diagnostic testing when determining a patient's candidacy for laparoscopic surgery. All of the advantages of laparoscopic surgery are inconsequential if the procedure is performed on an inappropriate patient. Anatomy of the Anterior Abdominal Wall Knowledge of the anterior abdominal wall vascular anatomy will reduce vascular complications associated with trocar placement. Of particular concern are the superior and inferior epigastric vessels. The superior epigastric artery, one of the terminal branches of the internal thoracic artery, enters first the rectus sheath and then the rectus muscle coursing near its lateral border. This artery and its adjacent vein often can be visualized by transillumination of the abdominal wall with the laparoscope. Visualization of the ventral abdominal wall laparoscopically often will locate the deep inferior epigastric vessels. The artery, a branch of the external iliac, and its accompanying vein course along the abdominal wall peritoneum just lateral to the rectus muscle until midway between the symphysis pubis and umbilicus, where it blends into the body of the rectus muscle. These vessels may be seen medial to the insertion of the round ligament at the deep inguinal ring. Therefore, placement of the trocar lateral to the deep inguinal ring and lateral border of the rectus muscle will avoid injury to these vessels. If placement of the trocar is too far laterally, branches of the superficial circumflex iliac vessels may be injured. Again, transillumination of the anterior abdominal wall using the laparoscope will assist in avoiding these vessels. As a general guideline, the superficial and inferior epigastric vessels are located approximately 5.5 cm from the midline. The superficial circumflex iliac vessels are approximately 7 cm from the midline. Theoretically, a “safe area” would be 8 cm above the symphysis pubis and 8 cm from the midline. If transillumination is not effective due to a thick abdominal wall, the surgeon may consider insertion of a spinal needle through the abdominal wall at the selected trocar insertion site. If no bleeding is observed after removal of the needle, the location is likely safe for trocar placement. Obtaining Intraabdominal Access For most laparoscopic surgeons, obtaining access to the peritoneal cavity is the most anxiety-provoking step of the entire endoscopic procedure. There are a number of ways to obtain intraabdominal access for trocar placement and establishment of a pneumoperitoneum. None of the available methods precludes the possibility of complications from trocar or needle insertion. In general, these methods of entry can be classified as either open or closed. An open entry uses the technique first described by Harry Hasson. An incision into the peritoneal cavity through all anterior abdominal wall layers at the umbilicus is made carefully. A blunt-tipped trocar is placed through the incision into the abdominal cavity. Insufflation is conducted through this trocar sleeve. This technique may reduce vascular injuries due to sharp trocar or needle placement, but the possibility of bowel injury still exits if there are abdominal wall adhesions. The closed entry may be performed with or without the use of a Veress needle to insufflate CO 2 . Placement of the sharp trocar through the umbilicus without prior insufflation has lost favor due to the higher risk of bowel and vascular injuries. Insufflation of CO 2 through a Veress needle is the most common method of entry, although some would argue the open technique should become the method of choice. One technique of access using a Veress needle is performed in the following manner: 1. A vertical incision through the skin corresponding to the size of the trocar is made at the base of the umbilicus. Note that this incision ideally is made at the base of the umbilicus and not infraumbilically as is classically taught. This is because the distance between skin and peritoneum on the anterior abdominal wall is shortest at the base of the umbilicus. In addition, the peritoneum is firmly attached, which will prevent tracking through the subcutaneous tissue and subsequent retroperitoneal insufflation. 2. The Veress needle is inspected for sharpness and a functioning spring mechanism to extend the protective sheath in the absence of pressure. 3. The anterior abdominal wall inferior to the umbilicus is grasped with the nondominant hand and the umbilicus is moved in the caudad direction, further displacing it below the bifurcation of the aorta. Alternatively, the incision may be elevated by grasping the edges of the umbilical incision with two Alice clamps and lifting. 4. The tip of the Veress needle is held in the dominant hand between the thumb and forefinger while the ulnar palm rests on the patient's abdomen. The needle is inserted carefully at a 90-degree angle, through the base of the umbilicus, millimeter by millimeter, until a click is heard and resistance is no longer felt, identifying intraabdominal placement. 5. A saline-filled 10-cc syringe is attached to the Veress needle and aspirated, inspecting for blood or bowel contents. If only bubbles are visible, saline is injected and observed to fall from the trough on the needle into the peritoneal cavity. 6. The syringe is removed, the insufflation tubing (with CO 2 turned on low flow) is attached, and the initial IAP is observed. If the pressure is greater than 10 mm Hg, the needle should be removed quickly to avoid retroperitoneal insufflation, which further displaces the peritoneum from its attachment to the anterior abdominal wall. A second needle placement attempt is then made. 7. Once intraperitoneal placement is confirmed, a CO 2 pneumoperitoneum is obtained. Insufflation up to an IAP of 20 to 25 mm Hg, as initially described by Reich and others, remains one accepted method. This temporary increase in IAP increases the distance between abdominal viscera and the anterior abdominal wall (in the absence of adhesions) during primary sharp trocar placement through the umbilicus. It is also extremely beneficial for surgeons with small hands, who may find it difficult to grasp and further elevate the abdominal wall if the initial pressure is less than 20 mm Hg. 8. The end of the sharp trocar is held in the palm of the dominant hand, with the forefinger extended along the shaft as close to the sharp tip as possible. The tip is inserted through the umbilical incision until the fascia at the base of the umbilicus is felt. Insertion is carried through at a near-90-degree angle until the tip is felt to pass through the peritoneum. The trocar is then directed toward the pelvis to minimize risk of vascular or bowel injury. 9. After placement of the primary trocar, the intraabdominal placement is confirmed visually and the IAP is reduced to 12 to 15 mm Hg. Alternative methods for obtaining pneumoperitoneum have been described and are useful in cases in which adhesions at or near the umbilicus are suspected. The most common method is insertion of the Veress needle in the left upper quadrant, followed by placement of a 5-mm trocar and laparoscope. Others have performed insufflation after inserting a Veress needle transfundally or through the posterior cul-de-sac into the intraabdominal cavity. There are no prospective randomized studies comparing the safety and efficacy of open-entry with closed-entry techniques. The incidence of injuries is so low that a very large study population would be required to show any statistical difference. Randomized studies comparing the Veress needle with the direct trocar insertion technique favor the Veress needle. The vast majority of surgeons agree that the method used should be that with which the surgeon has the most experience and is, therefore, most comfortable. Laparoscopic Procedures Ectopic Pregnancy Through earlier diagnosis, ultrasonography has reduced significantly the morbidity and mortality associated with ectopic pregnancies. In 98% of cases after the fifth week of pregnancy (when HCG levels are greater than 1,000), transvaginal ultrasonography can reliably visualize a normal gestational sac. The absence of an intrauterine sac above this level should alert the physician of a high likelihood of ectopic pregnancy. With the advent of methotrexate use, the total number of ectopic pregnancies treated surgically has decreased. Laparoscopy remains a standard for surgical management in the stable patient. While obtaining

informed consent, the surgeon must discuss management alternatives. Should the pregnancy alone be removed leaving the fallopian tube in place, or should a salpingectomy be performed? Assuming the pregnancy can be aborted without significant bleeding from the fallopian tube, this decision should be based on the status of the contralateral ovary and tube. If the contralateral tube and ovary are normal in appearance, without adhesions, then a salpingectomy or salpingostomy may be performed with a subsequent intrauterine pregnancy rate of 85% and a repeat ectopic rate of 90%. If the contralateral tube is impaired, then the subsequent intrauterine pregnancy rate is 46% and repeat ectopic rate 52%. In this situation, the affected fallopian tube should be removed. Technique for Salpingostomy or Salpingectomy Laparoscopic access is obtained as previously described. In addition to the umbilical port, two 5-mm lower abdominal ports are indicated, left and right. A suction-irrigator is used to aspirate the hemoperitoneum, if present. The pregnancy is identified within the fallopian tube, usually within the ampullary portion. The tube is stabilized with a Babcock or atraumatic grasper. There is considerable debate regarding the use of dilute vasopressin prior to making the incision on the antimesenteric side of the tube overlying the pregnancy. Clinicians must understand the use of vasopressin has the potential for delayed bleeding postoperatively. An incision parallel to the axis of the tube is now made on the antimesenteric side. This may be performed using a spoon, knife, or hook monopolar electrocautery, set at 50 to 80 watts to minimize bleeding. The suction-irrigator is used to “aquadissect” the attachments between the pregnancy and tube, thus aborting the pregnancy into the pelvis. Alternatively, the pregnancy may be grasped using atraumatic graspers and pulled from its attachment. It is then removed from the abdomen through the operating channel of a right-angle laparoscope or a 10-mm or larger port. If significant bleeding is encountered or the pregnancy does not dissect freely, a salpingectomy may need to be performed. Again, the fallopian tube is grasped with an atraumatic grasper and lifted. The tube is transected approximately 1 cm from its cornual end. Transection is performed after complete desiccation using bipolar cautery via Kleppinger forceps or may be performed after suture ligation (a window is created within the mesosalpinx inferior to the ligation site). The mesosalpinx, immediately inferior to the fallopian tube, is then desiccated using bipolar cautery and transected, freeing the tube and pregnancy from its anatomic attachments. Adnexal Mass Laparoscopic management of adnexal masses is gaining acceptance but still conjures considerable debate. The patient's age, transvaginal sonographic findings, and CA-125 level (if indicated based on patient's age) will determine if a laparoscopic approach rather than laparotomy is acceptable. Ultrasonographic criteria for benign and malignant tumors are presented in Table 46.4 and Table 46.5. CA-125 levels are most useful in postmenopausal women. A level of 35 units in postmenopausal women carries a sensitivity of 81% and specificity of 91% for malignancy. If the level is greater than 50, then the sensitivity remains essentially the same and the specificity increases to 97%. In premenopausal women, sensitivity remains 60% as the level increases, but specificity increases from 73% at a level of 35 to 95% at a level of 100.

TABLE 46.4. Ultrasonographic criteria for malignant ovarian tumors

TABLE 46.5. Ultrasonographic criteria for benign ovarian tumors A primary concern for those advocating laparotomy over laparoscopy for management of suspicious adnexal masses is the potential increased risk of rupture during laparoscopic surgery. Preoperative or intraoperative rupture changes a stage Ia ovarian carcinoma to a stage Ic. Studies have shown that intraoperative iatrogenic rupture of the tumor capsule does not adversely affect survival, but survival is negatively influenced by spontaneous or preoperative rupture. The American College of Obstetricians and Gynecologists guidelines for intraoperative management of adnexal masses are presented in Table 46.6. As a general rule, if malignancy is highly suspected, then a laparotomy and complete surgical staging procedure should be performed. If a laparoscopy is performed initially and malignancy subsequently is identified histologically, the staging procedure should be completed within 1 week.

TABLE 46.6. ACOG guidelines for laparoscopic intraoperative management of adnexal masses Technique for Ovarian Cystectomy After obtaining intraperitoneal access and placement of trocars, the uteroovarian ligament is grasped to stabilize the ovary. The ovarian cortex overlying the cyst is incised very superficially so as not to rupture the cyst. This may be performed using a CO 2 laser set at 5 to 10 watts, a monopolar electrode (knife, hook, spoon) set at 30 to 50 watts of cutting current, or scissors. The edge of the cut ovarian cortex is grasped with a biopsy forceps and elevated. The suction-irrigator is placed gently at the opening between cortex and cyst wall. Aquadissection is performed using Ringer's lactate. A blunt grasper is introduced within the space and the cyst is shelled out from its attachment to the cortex. The incision is made longer as necessary. The cyst is placed within an impermeable sac introduced through the umbilical port or through a culdotomy incision. For larger cysts, a zip-lock type sandwich bag may be gas sterilized and placed through the 10to 12-mm umbilical port. The sac is then removed either through the umbilicus or transvaginally. A spinal needle may be used to aspirate the contents of the cyst within the sac prior to removal to decrease its volume. Technique for Oophorectomy The first step in an oophorectomy procedure is always identification of the adjacent ureter. The ureter is most easily located at the pelvic brim near the bifurcation of the common iliac. Its retroperitoneal course may then be followed and location relative to the infundibulopelvic ligament noted. Once certain of the ureteral location, the infundibulopelvic ligament is grasped gently, elevated, and brought medially using a Babcock or other atraumatic grasper. The ligament is then desiccated at its attachment to the ovary using bipolar cautery. It is then transected. Alternatively, an opening through the peritoneum inferior to the ligament may be created using scissors. Suture ligatures may then be placed and tied. After transection of the infundibulopelvic ligament, the uteroovarian ligament is desiccated and transected. Desiccation of the mesovarium is followed by transection, completely freeing the ovary. The ovary is then placed in an impermeable sac and removed as described previously. Tubal Sterilization Surgical occlusion of the fallopian tubes is the most common method of contraception in developing countries. The techniques have evolved from a laparotomy to minilaparotomy to a laparoscopic approach including excision, electrosurgical desiccation using bipolar cautery, and application of clips or rings. A Pomeroy tubal ligation is easily performed laparoscopically by grasping the fallopian tube at its isthmic portion and placing a chromic endoloop suture around a segment of tube. The knuckle of tube may then be excised. Two clips are approved by the U.S. Food and Drug Administration for tubal occlusion, the Hulka-Clemens clip and the Filshie clip. Both clips are placed properly at the proximal isthmus after the tube has been placed on stretch. The clip should be applied at 90 degrees to the long axis of the tube and advanced until the hinge reaches the tube, incorporating the mesosalpinx at its tip. The falope ring is placed on the ampullary portion of the fallopian tube by grasping the tube with the applicator and drawing the tube inside the applicator. A 1- to 2-cm segment of tube is drawn into the ring. Although sterilization failure rates are lowest with unipolar cautery, most are performed using bipolar cautery. Bipolar desiccation is performed correctly with an in-line ammeter (indicating complete desiccation), current set at 25 watts, and desiccation of three contiguous areas of the isthmus incorporating the blood supply from the mesosalpinx. The results of the U.S. Collaborative Review of Sterilization study found that all methods of tubal occlusion are very effective, although failure rates were higher than expected. The Hulka-Clemens clip and bipolar methods carry the highest failure rates, likely due to improper application technique. Myomectomy Women with symptomatic leiomyomas who wish to retain their reproductive potential or who wish to retain their uteri are candidates for myomectomy. Most of these procedures are still performed abdominally, because laparoscopic myomectomy is one of the more difficult laparoscopic procedures. There is controversy as to whether a laparoscopic closure of the myometrium is comparable to a closure achieved in an open procedure. In general, a laparoscopic approach should not be selected if leiomyomas are larger than 5 to 8 cm, multiple, or embedded deep within the myometrium. If a laparoscopic approach is chosen, the patient should be informed of the potential increased risk of spontaneous uterine rupture prior to labor and of recurrence (33% after 27 months). Most myomectomies can be performed with three trocar sites. One 10-mm trocar is placed umbilically and 5-mm ones in the lower left and right abdomen. A vertical incision is made in the myometrium overlying the myoma using a spoon electrode (or other electrosurgical instrument) set at 80 watts of cutting current to minimize bleeding. The incision is continued through the myometrium to the surface of the myoma, identifying the plane between the pseudocapsule and fibroid. The myoma is grasped with a 5-mm single-toothed tenaculum or corkscrew, and traction is applied. The plane between the fibroid and pseudocapsule is entered and adhesions released with a combination of blunt and sharp dissection. The suction-irrigator works well here for blunt dissection. The adhesive attachments are desiccated using bipolar forceps prior to their transection to minimize bleeding. These adhesions may also be vaporized using the CO 2 laser. The process is continued until the base of the myoma is reached, where the remaining attachments are desiccated using bipolar cautery. The myoma is freed and placed in the posterior cul-de-sac for later retrieval. Hemostasis of the myometrium is achieved. A multilayer closure is now performed using single interrupted sutures placed and tied laparoscopically. Alternatively, if the operator is not skilled in laparoscopic suturing, or if there is a concern regarding obtaining a maximally tight closure, a small laparotomy incision may be made and the suturing performed abdominally. A uterine manipulator placed within the uterus can help to elevate the uterine incision to the laparotomy incision and make suturing through a smaller incision easier. The myoma is now removed from the abdomen. If a small laparotomy incision has been made, the myoma may be removed through this incision. Morcellation using a scalpel or scissors may be necessary if the myoma is larger than the incision. If the entire procedure has been performed laparoscopically, the myoma may be removed using a morcellator introduced through the umbilical incision and visualized with a 5-mm laparoscope placed in one of the lower trocar sleeves. Alternatively, a culdotomy incision can be made and the myoma removed vaginally. The culdotomy incision may then be closed vaginally or

laparoscopically. Hysterectomy The advanced surgical skills required to perform a laparoscopic hysterectomy and the economic factors involved have limited the acceptance of this procedure in gynecologic practice. The laparoscopic hysterectomy or laparoscopically assisted vaginal hysterectomy was never intended to replace the vaginal hysterectomy. The primary role of this procedure is to reduce the number of abdominal hysterectomies performed. Seventy percent of hysterectomies in the United States are performed abdominally. A review of the data from the Health Insurance Commission Medicare in Australia showed a decrease in the incidence of abdominal hysterectomy from 70% in 1991-1992 to 57% in 1994-1995 after the introduction of laparoscopically assisted hysterectomies in private hospitals. The American College of Obstetricians and Gynecologists has agreed on the indications listed in Table 46.7 for laparoscopically assisted vaginal hysterectomy (there has been no comment on total laparoscopic hysterectomies, but the same indications inherently would apply).

TABLE 46.7. Indications for laparoscopically assisted vaginal hysterectomy Several randomized controlled trials comparing laparoscopically assisted vaginal hysterectomies with abdominal hysterectomies reported that the former are associated with significantly less postoperative pain, shorter hospital stays, faster return to work and normal activities, but possibly longer operative times. Technique for Total Laparoscopic Hysterectomy A uterine manipulator (e.g., RUMI system), a vaginal extender (e.g., Koh Cup Vaginal Fornices Delineator), and a pneumooccluder (Colpo-Pneumo Occluder; all products from Cooper Surgical, Shelton, CT) or similar instruments are put in place at the beginning of the procedure. A 10-mm reusable trocar is introduced through the umbilicus after pneumoperitoneum is obtained as described previously. Five-millimeter trocars are introduced into the lower left and right abdomen under direct visualization. The course of the ureters is identified retroperitoneally. If the ovaries are to be removed, the infundibulopelvic ligament is isolated, desiccated, and transected as described previously for oophorectomy. If ovaries are to remain, then the round ligament, fallopian tube, and uteroovarian ligaments are desiccated using bipolar forceps and transected approximately 1 cm from their uterine attachments. The anterior leaves of the broad ligament are incised bilaterally to meet in the midline, releasing the bladder peritoneal attachment to the lower uterine segment. The harmonic scalpel with the hook attachment works well to minimize oozing. The bladder is gently pushed inferiorly over the vaginal metal colpodelineator. The posterior broad ligaments are incised downward using scissors to the insertion of the uterosacral ligaments, allowing lateral retraction of the ureters. The uterine arteries are identified within the broad ligament, and the surrounding loose, areolar tissue is skeletonized in a blunt fashion. The uterine arteries are then desiccated completely, as indicated by the ammeter, by placing the bipolar forceps at right angles to the lower uterine segment on either side, above the level of the vaginal metal colpodelineator. The uterus should take on a bluish hue if complete desiccation has been achieved. The desiccated arteries are transected, freeing the cardinal ligament attachment. The vaginal pneumooccluder balloon is then insufflated prior to the colpotomy incision. The vaginal metal colpodelineator is emphasized by pushing the uterus cranially. The harmonic scalpel with hook electrode is used to make a circumferential incision over the metal ring, releasing the cervicovaginal attachment. The occluder balloon is deflated, and the uterus is brought into the vagina and now serves as the pneumooccluder. Hemostasis of the vaginal cuff is achieved. The cuff is sutured and closed laparoscopically, incorporating the uterosacral ligaments for pelvic support. The uterus is then removed from the vagina. Endometriosis Surgical treatment of endometriosis usually is performed for pain or infertility. There are no randomized controlled trials comparing excision, vaporization, fulguration, and desiccation of endometriotic lesions. Although each method of destruction has its proponents, the authors prefer excision of deep fibrotic endometriosis and vaporization of superficial implants. The only definitive cure for endometriosis is surgical resection or complete destruction of the endometriotic lesion. Surgical management does not include bilateral salpingo-oophorectomy. Removal of the ovaries will not necessarily ensure relief of symptoms. Estrogen, in addition to that produced by the ovaries, also is produced by peripheral conversion of androgens in the presence of aromatase. Aberrant aromatase expression has been identified in some endometriotic implants, causing local production of estradiol within them. This may explain the reason for failures or “recurrences” after bilateral salpingo-oophorectomy or only superficial destruction of the implant. Ovarian endometriomas should be treated by cystectomy rather than drainage and coagulation, as evidenced by the results of a randomized clinical trial. Often, these ovaries are adherent to the pelvic sidewall, adjacent bowel, or the opposing ovary. During adhesiolysis, the endometrioma often is ruptured, spilling its thick, chocolate-like fluid into the cul-de-sac. The thick fluid should be aspirated and opening to the endometrioma identified. If necessary, the opening is enlarged until the thick endometrioma cyst wall is identified. Using a biopsy forceps (single-toothed grasper), the cyst wall is grasped and peeled from its attachment to the ovarian cortex. Once hemostasis is achieved, the cortex usually does not require suture closure. Complications Information regarding entry access injuries is provided in a review reflecting the Physician Insurers Association of America (PIAA). They noted that entry access injuries occurred in between 5 per 10,000 and 3 per 1,000 patients undergoing laparoscopic surgery. Bowel, rectal, and retroperitoneal vascular injuries accounted for 76% of all injuries incurred in the process of establishing a primary port. The PIAA reported that approximately 50% of bowel injuries were unrecognized for 24 hours or longer following the laparoscopic surgical procedure. A delay in recognition, age over 59 years, and major vascular injuries were independent factors associated with death. In a Canadian Task Force study, the overall complication rate for all laparoscopic procedures was 0.65%; for operative laparoscopy it was 0.80%. The American Association of Gynecologic Laparoscopists has noted similar results (1.45 complications per 1,000 procedures). In a review of 32,205 gynecologic laparoscopic procedures in a Finnish study, 130 major complications were noted. The total complication rate was reported as 4 per 1,000 and, more specifically, 0.6 per 1,000 diagnostic laparoscopies, 0.5 per 1,000 sterilizations, and 12.6 per 1,000 operative laparoscopies. Intestinal injuries were reported in 0.7 per 1,000 procedures, incisional hernias in 0.3, urinary tract injuries in 2.5, major vascular injuries in 0.1. Overall, laparoscopically assisted vaginal hysterectomy was associated with the highest complication rate at 2.3%, with ureteral injuries occurring in 1% of women undergoing the procedure. Gas and air emboli have been reported with both laparoscopic and hysteroscopic procedures. Oxygen saturation monitoring, as well as end-tidal carbon dioxide and, more recently, capnography and mass spectrometry, are designed to provide early diagnosis. If an embolus occurs, it frequently is accompanied by cardiac arrhythmia and cardiac arrest. It is beyond the scope of this chapter to provide detailed information regarding cardiopulmonary resuscitation. Ureteral injury, as noted above, has been reported in association with minimally invasive surgery. Often times it is not recognized at the time of the initial procedure. The diagnosis should be considered when a patient exhibits postoperative fever, hematuria, flank pain, or evidence of peritonitis. Initial treatment with ureteral stenting on occasion is inadequate, resulting in subsequent ureteroneocystostomy. With respect to bladder injury, if one suspects such, distension of the bladder with methylene blue can identify a defect. If there is suspicion of ureteral injury, intravenous administration of indigo carmine frequently proves helpful in identifying the defect. Other minimally invasive surgery–associated complications have included spleen laceration. The injury was reported in four patients, and predisposition included splenomegaly, with injury occurring primarily during trocar insertion. This was associated with a laparoscopic salpingoplasty. However, this complication is extremely rare. Death, although extremely rare following minimally invasive surgical procedures, has been reported. It is more commonly associated with unrecognized trauma to the intestinal tract with ensuing sequelae. It also has been reported in association with major vascular trauma.

SUMMARY POINTS Endoscopic surgery has replaced open surgery for many commonly performed gynecologic procedures such as tubal sterilization, excision of unruptured tubal pregnancies, ablation of peritoneal endometriosis, and simple ovarian cystectomy. Endoscopically assisted vaginal hysterectomy with removal of the adnexal structures demonstrates the potential for this technology. Endoscopic procedures have several advantages. They can be performed on an ambulatory basis, require shorter recovery times, cause less postoperative pain, and improve the cosmetic impact of incisions. Endoscopic procedures require specialized equipment and a high degree of technical skill. SUGGESTED READINGS Introduction Dicker RC, Scally MJ, Greenspan JR. Hysterectomy among women of reproductive age: trends in the United States. JAMA 1982;248:323–338.

Laparoscopy

Chandler J, Corson S, Way L. Three sprecta of laparoscopic entry access injuries. J Am Coll Surg 2001;192:478–491. Isaacson K. Endometrial ablation: Is one technique best? Contemp Obstet Gynecol 2001:31–38. Molloy D, Crosdale S. National trends in gynaecological endoscopic surgery. Aust N Z J Obstet Gynaecol 1996;36:27–31. Morice P, Vaila J, Pautier P, et al. Port-site metastasis after laparoscopic surgery for gynecologic cancer. J Reprod Med 2000;45:873–874. Sanfilippo JS, Levine R, eds. Operative gynecologic laparoscopy. New York: Springer Verlag, 1996.

Complications Chang M, Shiau C, Chang C, et al. Spleen laceration, a rare complication of laparoscopy. J Am Assoc Gynecol Laparosc 2000;7:269–272. Cooh B, Kwon D, Park K, et al. Late presentation of ureter injury after laparoscopic surgery. Obstet Gynecol 2000;95:370–379. Harkki-Siren P, Sjoberg J, Kurki T. Major complications of laparoscopy: a follow-up Finnish study. Obstet Gynecol 1999;94:94–98. Hershlag A, Loy R, Lavy G, et al. Femoral neuropathy after laparoscopy-a case report. J Reprod Med 1990;35:575. Reich H, Ribeiro SC, Rasmussen C, et al. High-pressure trocar insertion technique. JSLS 1999;3:45–48. Sadik S, Onoglu S, Mendilicogu I, et al. Urinary tract injury during advance gynecologic laparoscopy. Am Assoc Gynecol Laparosc 2000;74:569–572. Speights S, Moore R, Miklos J. Frequency of lower urinary injury at laparoscopic Burch and paravaginal repair. J Am Assoc Gynecol Laparosc 2000;7:515–518. Wang P, Lee W, Waun C, et al. Major complications of operative and diagnosis laparoscopy for gynecologic disease. J Am Assoc Gynecol Laparosc 2001;8:68–73.

Chapter 47 Hysteroscopic Surgery Danforth’s Obstetrics and Gynecology

Chapter 47 R. Stanford Williams

Hysteroscopic Surgery

INTRODUCTION INSTRUMENTATION Distension Media General Technique ENDOMETRIAL ABLATION HYSTEROSCOPIC SEPTUM RESECTION HYSTEROSCOPIC MYOMECTOMY HYSTEROSCOPIC POLYPECTOMY ASHERMAN SYNDROME PROXIMAL TUBAL CANNULATION SUMMARY POINTS SUGGESTED READINGS Instrumentation, Distension Media, and General Technique Endometrial Ablation Hysteroscopic Septum Resection Hysteroscopic Myomectomy Hysteroscopic Polypectomy Asherman Syndrome Proximal Tubal Cannulation Complications

INTRODUCTION Although hysteroscopy has been described since the early 1800s, widespread use by practicing gynecologists did not occur until the 1980s. With improvements in optics, video systems, and distension media, there has been an increased acceptance of hysteroscopy as the gold standard in the evaluation of the uterine cavity and treatment of intracavitary pathology. Hysteroscopy is used most commonly for evaluation of abnormal uterine bleeding, but it also is used frequently for evaluation of the endometrial cavity in patients with recurrent pregnancy loss and infertility. Many studies have shown that blind dilation and curettage may miss up to 60% of endometrial pathology, such as endometrial polyps and submucous leiomyoma, although with increasing expertise in ultrasonographic techniques such as the saline sonohysterogram, endometrial pathology often can be suspected prior to performing definitive operative hysteroscopy. Alternatively, many practitioners perform office hysteroscopy with small-diameter hysteroscopes, which do not require significant dilation of the cervix, can provide direct visualization of the endometrial cavity, and facilitate a directed biopsy of suspected endometrial lesions. Operative hysteroscopy requires larger diameter instrumentation and is best performed under anesthesia in the outpatient operating room because of the need for significant cervical dilation and more extensive instrumentation. Hysteroscopic surgery allows a variety of intrauterine surgical procedures such as myomectomy, polypectomy, resection of a uterine septum, endometrial ablation, correction of intrauterine synechiae, and cannulation of proximal tubal occlusion.

INSTRUMENTATION Hysteroscopy can be used both as a primary diagnostic tool and as a more definitive operative technique. Diagnostic hysteroscopy can be performed either in an office setting or in the outpatient operating room and requires only small 3.6-mm to 5-mm hysteroscopes. These hysteroscopes may be either flexible or rigid. A small channel often is provided for a small biopsy instrument, but these are rather delicate, and significant surgical procedures will require larger hysteroscopes. The flexible hysteroscope is used often for office hysteroscopy because it can be inserted through a narrow cervical os and can negotiate the cervical canal, using a deflecting lever to guide the instrument. The flexible hysteroscope is available in two diameters, 3.6 mm and 4.9 mm, both with a zero-degree optical view. The larger diameter endoscope also provides a 2.2-mm diameter instrument channel, which allows for directed biopsies of endometrial pathology. The larger instrument provides better optics than the smaller flexible hysteroscope but is still somewhat inferior optically to a 4-mm rigid scope. Most small diagnostic hysteroscopes have only a single channel for installation of a distending medium. In an office setting, many physicians employ CO 2 as the distension medium, necessitating the use of a hysteroscopic insufflator to control intrauterine pressure. If blood or mucus is present within the endometrial cavity, a liquid medium will be required. If a low-viscosity liquid distension medium is used, then a continuous flow sheath, which allows for the continuous simultaneous inflow and outflow of the medium, thus flushing any blood or mucus from the cavity, will give the optimal view. Rigid hysteroscopes are available with 0-, 12-, 15-, 30- and 70-degree optical view telescopes. Thirty-degree scopes are used primarily for diagnostic procedures in order to see easily the entire endometrial cavity by rotating the telescope 360 degrees. Operative procedures most commonly require the 12- or 15-degree telescopes, so that the operational component of the instrumentation can be visualized fully. The operative hysteroscope uses a 4-mm rigid telescope within a 7- to 8-mm operative sheath, which can be configured to either provide a port for the insertion of accessory instruments (scissors, biopsy forceps, catheters) or use a resectoscope with a working element for resectoscope electrodes. Older models of operative hysteroscopes included only a single channel for installation of distension medium. Adequate egress of fluid usually is not possible with these hysteroscopes, because the only available exits are the fallopian tubes or around the hysteroscope through a patulous or over-dilated cervix. Use of these single-channel hysteroscopes should be limited to high-viscosity distension media. Most commonly, operative hysteroscopy is performed with a continuous flow sheath allowing for input of the distension medium through a middle channel and egress of the medium through an outer evacuation channel. This provides for constant washing of the uterine cavity, removing blood and debris and allowing the use of a low-viscosity distension medium. Infusion of low-viscosity fluids can be accomplished either by the force of gravity or with infusion pumps. Suction tubing attached to the outflow port can be directed to wall suction or allowed to drain by gravity. Low-viscosity–fluid pumps have been designed to operate in pressure ranges of 0 to 80 mm Hg. They can deliver fluid at a rate necessary to maintain a preset pressure with a maximum flow rate of 300 mL/minute, although the upper limit of flow through the outflow ports for most hysteroscopes is 250 mL/minute. Outflow usually is adjusted to be significantly lower than the maximum to allow adequate visualization free from blood and debris, maintain adequate uterine distension, and minimize the amount of fluid needed to complete the procedure. Alternatively, inflow of the distension medium can be controlled by gravity. The height of the bag of infusion medium above the patient controls the maximum intrauterine pressure. Every 1 foot of height above the patient that the bag of distension medium is placed will deliver approximately 25 mm Hg of pressure to the endometrial cavity. This system is then regulated by the amount of outflow, to maximize visualization while maintaining adequate uterine pressure. Because pressure of inflow is constant, changes in intrauterine pressure, and thus distension, are affected by alterations of the rate of the outflow. Use of standard suction containers to collect the outflow will make fluid measurement and calculation of any fluid deficit straightforward. Modern operative hysteroscopy requires a video camera and video monitor for adequate operative visualization. A halogen or xenon light source providing 150 to 300 watts of incandescent light is used and attached to the hysteroscope by a fiber optic light cable. The light cable should be inspected frequently to ensure that a significant number of the internal fibers have not broken and that the light cable is capable of delivering an adequate amount of light through the hysteroscope. Most chip cameras have the capability to adjust gain and can be integrated with the light source for automatic light balance. For documentation of findings and recording of procedures, video capture units may be used. Most commonly, VCRs are used to videotape pertinent portions of the procedure. Also available are video capture units for taking still pictures or storing digital images on a computer hard disk or CD. Energy for operative hysteroscopy can be delivered either with the Nd:YAG laser or with electrosurgical generators delivering either unipolar or bipolar energy. Other

lasers, such as the CO 2 laser, are not used in hysteroscopy because of their failure to penetrate fluid and because of the generation of smoke if CO 2 is used as the distension medium. The Nd:YAG laser can be delivered through a flexible quartz fiber passed through the instrument channel of the operating hysteroscope, and its wavelength penetrates through the liquid distension medium used in hysteroscopy. The extent of tissue necrosis can be up to a depth of 4 to 5 mm. Varying the distance of the fiber tip and incident angle can regulate the extent of thermal damage. It is rendered ineffective at distances greater than 2 cm or as the incident angle deviates more than 90 degrees. This laser is used often to perform endometrial ablation, using power outputs of 50 watts by dragging the fiber over endometrial surfaces. Electrosurgery through a standard resectoscope usually utilizes a monopolar technique, with the electrical probe serving as the source electrode and the return plate on the patient as the return electrode. The resectoscope can be used with either cut or coagulation output settings on the electrosurgical generator, or a blend of the two. When used in the cutting mode a high-frequency sine wave is delivered, which creates extremely high current density, instantly superheating cellular water to vaporization, causing cellular architecture to explode, resulting in tissue cutting. In the coagulation mode, delivery of high-frequency energy is interrupted by periods of modulation. This alternation of frequency and interruption results in wider zones of lateral tissue coagulation and damage, resulting in coagulation and sealing of blood vessels. A variety of electrode tips are available including a cutting loop for excision of tissue, a roller ball or bar for coagulation and ablation, and a knife electrode for incision. A new family of bipolar electrical generators (Versapoint, Gynecare of Ethicon, Somerville, New Jersey) has been developed. Electrodes have been designed in several configurations producing variable tissue effects. A ball tip can be used for vaporization with limited tissue desiccation, a spring tip for vaporizing larger amounts of tissue, and a twizzle tip for resecting and morcellating tissue. These tips have both active and return electrodes and require an electrolyte-containing medium such as saline. In contrast, if one is using a monopolar resectoscope, a nonelectrolyte distension medium such as glycine must be used. Distension Media Four basic types of distension media are used for hysteroscopy. The first type, CO 2 , is used primarily for diagnostic hysteroscopy in an office setting. Secondly, a high-viscosity medium such as Hyskon is used primarily with inflow only–type hysteroscopes. The third and fourth types are both low-viscosity solutions that are used with continuous flow hysteroscopes, electrolyte solutions and nonelectrolyte solutions. The choice between an electrolyte and nonelectrolyte solution will depend on the use of monopolar versus bipolar electrocautery. For diagnostic procedures in the office, many physicians choose CO 2 as the distension medium. CO 2 use requires a hysteroflator that delivers the gas at preset intrauterine pressures and has regulated flow rates. CO 2 may be used with either a small diagnostic rigid hysteroscope or a flexible hysteroscope and does not require a return channel for continuous flow, because the CO 2 gas will escape from the cervix or through patent fallopian tubes into the peritoneal cavity where it is absorbed. Starting pressures for CO 2 are usually between 50 and 75 mm Hg. If adequate distension is not achieved, it may be necessary to increase the intrauterine pressure to a maximum of 100 mg Hg or a maximal flow of 100 mL/minute. Higher pressures or flow rates may produce a gas embolus and fatalities have been reported. CO 2 will give an ideal view of a uterus that is not bleeding, because light reflection is identical to that of room air. However, any blood or mucus within the endometrial cavity will require changing to a liquid medium. Thirty-two percent high-molecular-weight Dextran-70, Hyskon, was used commonly as a liquid distension medium for operative hysteroscopy for many years prior to the evolution of hysteroscopes that accommodate continuous flow. Its nonmiscibility with blood allowed its use when either blood or mucus was present in the endometrial cavity or bleeding is anticipated. Hyskon is compatible with either the Nd:YAG laser or electric cautery devices. When using Hyskon as the distension medium, its delivery requires significant constant pressure to overcome the resistance of a high-viscosity fluid flowing through a standard diagnostic sheath. The major disadvantage of Hyskon is the difficulty in cleaning the solution from the instruments and stopcocks. If the instrumentation is not thoroughly cleaned, the dextran crystallizes and results in clouding of the hysteroscope lens and freezing of stopcocks. Rarely, patients may have an anaphylactic reaction to the dextran. Intravascular absorption of Hyskon will also result in a ten-fold increase of intravascular volume, with accompanying cardiovascular overload and pulmonary edema. Careful monitoring of the amount of Hyskon intravasated during the procedure is mandatory, and absorption of 100 to 200 mL of Hyskon should warrant terminating the procedure. Because the molecular weight of Hyskon exceeds that which can pass into the circulation from the peritoneal cavity, spill through the fallopian tubes is inconsequential. Operative hysteroscopy most commonly now uses low-viscosity solutions, which can be either electrolyte solutions or nonelectrolyte solutions. If monopolar resectoscopes are used, this equipment requires nonelectrolyte solutions, so that the flow of energy will be directed from the electrode tip into the tissue and not allowed to “short circuit” through an electrolyte-containing medium throughout the entire uterus. Nonelectrolyte solutions that have been used commonly for operative hysteroscopy include 1.5% glycine, sorbitol, 5% mannitol, and dextrose in water. Significant intravasation of distension medium may occur with resectoscope use. As tissue is resected, venous channels within the endometrium and myometrium are opened, and the pressure of the distension medium will result in the absorption of these solutions. The primary complications associated with nonelectrolyte low-viscosity solutions include fluid overload and hyponatremia. Fluid overload may result in pulmonary edema, and severe hyponatremia may result in neurologic sequela such as confusion, seizures, and even death. Intraoperative monitoring of inflow and outflow must be performed every 5 to 10 minutes throughout the procedure, and a discrepancy between 500 and 1,000 mL with nonelectrolyte solutions should warrant termination of the procedure. Glycine use has also been reported to cause hyperammonemia because of its conversion from glycine to ammonia by the liver. Electrolyte solutions, such as normal saline or lactated Ringer solution, are used with bipolar electrical devices or for continuous flow diagnostic hysteroscopy. Because bipolar devices contain both the active and return electrodes at the electrode tip, electrolytes are needed to complete the electrical circuit. The primary complication associated with electrolyte solutions is fluid overload, and a discrepancy of 1,500 to 2,000 mL during the procedure warrants termination of the procedure. During operative hysteroscopy with significant operating time and use of large amounts of distension medium, the anesthesiologist should keep intravascular fluid replacement to a minimum to avoid fluid overload. Anesthesia personnel should also monitor the patient carefully for electrolyte abnormalities when using nonelectrolyte solutions and anaphylactic shock when using Hyskon. General Technique Hysteroscopy can be difficult to perform during the luteal phase because of the abundance of endometrial tissue. Performing hysteroscopy during the early to middle follicular phase should ensure adequate visualization of the uterine cavity. Alternatively, the endometrium can be suppressed with 2 to 4 weeks of progestin therapy, or hysteroscopy may be performed at any time in a patient taking oral contraceptives because of the dominant atrophy effect of progestin. Gonadotropin-releasing hormone (GnRH) analogs have most commonly been used to prepare the endometrium for endometrial ablation. At least 4 weeks of preoperative treatment are required for GnRH analogs such as leuprolide acetate (Depo-Lupron), because these medications are initially agonists and will actually increase estrogen output for the first 7 to 10 days before subsequent down-regulation of the pituitary ovarian axis and subsequent endometrial atrophy. The cervix should be dilated no larger than the outer diameter of the hysteroscope that will be used. With many larger operative hysteroscopes, this will require dilation of the cervix to at least the diameter of a 20-French Hank dilator or a 9/10-French Hegar dilator. Care should be taken to avoid cervical lacerations and uterine perforation during cervical dilation. With insertion of the hysteroscope, the cervical canal can be visualized and the hysteroscope guided into the endometrial cavity under direct vision. If over-dilation of the cervix has occurred and the distension medium cannot be kept within the endometrial cavity, an additional tenaculum may be placed on the posterior lip of the cervix, or a special four-pronged tenaculum can be used to compress the cervix around the hysteroscope. The cervical canal and internal os will appear off-center within the field of view when using offset-angle lenses. When the angle of the lens is oriented to look downward, the internal os will appear at the 12 o'clock position. If the telescope is inverted and the lens pointed upward, the os will appear in the 6 o'clock position. The latter position is useful for viewing a retroverted uterus. The surgeon always should maintain the camera position in a straight-up-and-down configuration so that the view on the screen corresponds anatomically to the patient's position. As the hysteroscope is rotated to visualize the entire endometrial cavity, one hand should be kept on the camera to prevent its rotation, or the view on the monitor will be oriented improperly. With insertion of the hysteroscope, the cervical canal should be visualized and the endometrial cavity entered carefully through the internal os. If adequate visualization is prevented by blood and mucus, continuous flow of the distension medium should be maintained for 30 to 60 seconds to wash out blood and debris. If the field of view still appears red, the hysteroscope should be pulled back 1 to 2 centimeters, because it is a common mistake to insert the hysteroscope too far and the lens may be abutting the uterine fundus. Visualization can also be compromised by inadequate distension of the uterine cavity because of insufficient intrauterine pressure. The fluid delivery devices should be checked to ensure adequate pressure, or if a gravity system is being used, the height of the bag above the patient should be extended. Careful adjustment of the outflow should be made to clear any ongoing bleeding and debris without decreasing intrauterine pressure or using extremely large volumes of distension medium. The entire endometrial cavity should be inspected carefully, including the identification of both tubal ostia, the fundus, and the anterior and posterior portions of the

uterine wall. Video documentation of the intraoperative findings is useful. If operative techniques are to be used during the hysteroscopy, inflow and outflow should be measured carefully and reported to the surgeon every 5 to 10 minutes. If significant bleeding or a long operative time is anticipated, vascular constriction through the paracervical injection of a dilute pitressin solution may be used. Care should be taken that pitressin is not injected intravascularly, because reports of fatalities with inadvertent intravenous administration have been reported.

ENDOMETRIAL ABLATION Approximately 35% of gynecologic complaints concern menorrhagia, and it is estimated that 60% of these women ultimately will be treated with hysterectomy. Endometrial ablation originally was developed as an alternative treatment for patients who are medically too unstable for the surgical stress of hysterectomy. Since its original development in the 1980s, however, patient selection criteria have now expanded and hysteroscopic endometrial ablation is viewed by many as an alternative to hysterectomy, even in healthy patients. Prior to consideration of hysteroscopic endometrial ablation, endometrial pathology needs to be excluded with either a combination of endometrial biopsy to rule out hyperplasia or carcinoma and transvaginal ultrasonography with saline infusion to rule out polyps or submucous myomas or, alternatively, office diagnostic hysteroscopy could be performed with directed biopsies. Because the goal of an endometrial ablation is the artificial creation of intrauterine synechiae, many experts feel that preoperative preparation of the endometrium with an GnRH agonist or continuous progestin will maximize the chance of adequate scar formation within the endometrial cavity postoperatively. Alternatively, some authors perform a mechanical preparation of the endometrium using a sharp curette or suction curettage immediately prior to endometrial ablation. This technique has the advantage of being able to immediately perform the procedure without a lengthy preoperative medical suppression of the endometrium. The disadvantage of a mechanical preparation, however, is the possibility of inadequate visualization during the procedure because of bleeding and the possibility of a compromised outcome secondary to inadequate destruction of the endometrial basalis. Endometrial ablation can be done either with the Nd:YAG laser or the hysteroscopic resectoscope, using either a wire lube or roller ball technique. The Nd:YAG laser system uses a 600-µ bare fiber with power settings of 55 to 70 watts. The Nd:YAG laser penetrates 4 to 5 mm into the endometrium, destroying not only the superficial endometrial layers but also the endometrial basalis and superficial myometrium. A touching technique was described originally, dragging the fiber against the endometrium, thus destroying the endometrium and superficial myometrium by vaporization and necrosis. Alternatively, a nontouch technique with the higher power setting has been described in which the fiber is held perpendicular to the endometrium without touching the surface, achieving a deep coagulation effect. Many practitioners use a combination of the two techniques, and results of the two techniques appear comparable. Due to the expense of laser ablation, electrosurgical ablation is performed more commonly. Most resectoscopes use monopolar electrodes, necessitating a nonelectrolyte low-viscosity medium. Hysteroscopic resection of the endometrium has been described using a loop electrode to excise strips of endometrium, although the incidence of uterine perforation is reportedly higher with this technique when compared with a roller ball or laser ablation. Most commonly, a roller ball technique is used. With a roller electrode, a blend 1 setting with 70 to 140 watts of power is used, depending on the width and thickness of the electrode. This current will allow consistent penetration and destruction of the uterine tissue. Care must be taken to ensure adequate ablation of the uterine cornu and to avoid destruction of the endocervical canal to prevent cervical stenosis. Extensive studies of the outcome following endometrial ablation have been published. Approximately 80% to 90% of women will have a significant reduction of their menorrhagia and will be satisfied with the outcome, including approximately 25% to 40% of women experiencing amenorrhea postoperatively. Complications of endometrial ablation include a 10% hysterectomy rate because of dissatisfaction with the outcome of surgery. In addition to the immediate postoperative complications previously discussed, there also have been reports of isolated endometrial carcinoma in areas of the uterus not adequately ablated and symptomatic hematometra developing in loculated areas of inadequate ablation. Although most women are sterile after this procedure, intrauterine pregnancies can occur and patients should be counseled to use adequate contraception.

HYSTEROSCOPIC SEPTUM RESECTION The septate uterus is the most common congenital uterine malformation associated with recurrent reproductive failure and obstetric complications. The appearances of a septate uterus and bicornuate uterus are extremely similar on hysterosalpingogram, and the distinction must be made by demonstrating a normal external fundal contour with a uterine septum, as opposed to the presence of two uterine horns in a bicornuate uterus. The uterine septum may be thin or broad and may be of varying length, from an exaggerated arcuate appearance to total division of the uterine corpus, possibly including the cervix ( Fig. 47.1). Indications for hysteroscopic resection of a uterine septum include repeated first- or second-trimester losses, a history of premature labor and delivery and, possibly, concurrent infertility.

FIG. 47.1. Hysteroscopic view of a uterine septum extending down to the lower uterine segment before division.

The hysteroscopic approach to the division of the uterine septum is performed almost exclusively today rather than the classic transabdominal resection or incision of the septum performed prior to the 1970s. The hysteroscopic approach allows for ambulatory treatment of this disorder and, because the uterine wall has not been invaded, cesarean delivery is not indicated postoperatively. After adequate documentation of the diagnosis of a septate uterus and the exclusion of concurrent causes of pregnancy wastage, the surgery should be scheduled during the early follicular phase to allow adequate visualization of the septum. The use of GnRH analogs or other drugs to induce endometrial suppression may result in the iatrogenic formation of postoperative intrauterine scar formation unless postoperative estrogen replacement is given. Semi-rigid scissors are most commonly used to divide uterine septa ( Fig. 47.2). The uterine septum is usually avascular and cautery is typically unnecessary. Under direct visualization, the septum is incised at its midpoint. As the incision is carried toward the fundus, the septal tissue normally retracts anteriorly and posteriorly into the myometrial wall, and resection of tissue usually is not needed. When the uterine cavity is symmetric, or when normal myometrial tissue is observed because of bleeding, the dissection should stop ( Fig. 47.3). It is also useful to perform concurrent laparoscopy to both visualize the normal fundal contour before dividing the septum, as well as to guide the direction and extent of the dissection hysteroscopically.

FIG. 47.2. Uterine septum is being divided by hysteroscopic rigid scissors.

FIG. 47.3. The uterine septum has been resected completely and a contour of the fundus is normal, with normal myometrial tissue visible.

The septum also can be divided with either fiberoptic lasers or cutting loop resectoscopes. The disadvantages of these techniques include additional instrumentation and further complications should uterine perforation occur. When perforation occurs with hysteroscopic scissors, bleeding is unlikely and damage to surrounding structures is rare. When uterine perforation has occurred using an electrical cutting loop or laser fiber, damage to the surrounding bowel is possible. There is also concern that these modalities may damage surrounding normal myometrial and endometrial tissue. Although some surgeons routinely administer high-dose estrogen therapy for 30 days postoperatively, randomized studies have shown no benefit when used in normally cycling women. Similarly, an intrauterine splinting device such as a pediatric Foley catheter has been advocated by some authors but usually is not needed because of the rapid reepithelialization of the endometrial cavity postoperatively. Normal reproductive outcome following the division of a uterine septum is reported to be between 70% and 85% term delivery rate in patients with prior recurrent spontaneous abortions. Pregnancy outcome is equal with all three methods of septum division.

HYSTEROSCOPIC MYOMECTOMY Uterine leiomyomas are the most common benign tumors of the uterus, and submucous myomas are common causes of menorrhagia and abnormal uterine bleeding. Submucous myomas also are associated with recurrent pregnancy loss and infertility. The evaluation of abnormal uterine bleeding and menorrhagia should include either a diagnostic hysteroscopy or a saline infusion sonohysterography to diagnose submucous myomas. Blind procedures, such as endometrial biopsy and dilation and curettage, will commonly miss these tumors. Most submucous myomas do not have a pedicle but project into the endometrial cavity with a broad base. Myomas that have greater than 50% of their volume projecting into the cavity and are less than 3 to 4 cm in size can be approached adequately for resection hysteroscopically. It is often useful to pretreat patients with a GnRH analog prior to a hysteroscopic myomectomy to shrink the myoma, as well as thin the endometrium. If the patient is anemic from menorrhagia, GnRH-agonist pretreatment also will allow her a period of amenorrhea and permit her hemoglobin level to normalize. The resectoscope with a loop electrode is used commonly to shave the myoma until it is flush with the endometrial cavity. Power settings of 100 to 120 watts are used. With unipolar devices, a nonelectrolyte solution must to be used as the distension medium. The loop should be advanced past the myoma, and resection of strips of tissue is accomplished by pulling the loop toward the operator ( Fig. 47.4). This technique will minimize the risk of uterine perforation. It is frequently necessary to remove the chips of tissue when they become so numerous as to obscure the surgical view. After removal of the resectoscope, a uterine polyp forceps can be used to blindly remove the chips. Even if all the tissue fragments are not removed, they will be expelled from the uterus with the next menstrual period.

FIG. 47.4. A submucous myoma is being resected with a resectoscope loop electrode.

It may be useful to reduce intraoperative bleeding and reduce intravasation of distension medium by injecting a dilute solution of pitressin, 20 units per 100 cc injectable saline, paracervically prior to the endometrial resection to achieve vascular constriction. Care should be taken not to inject this solution intravascularly, because deaths have been reported. A total of 5 to 10 cc is usually sufficient. Intraoperatively, small bleeding vessels may be coagulated with the loop electrode using 40 watts of coagulation current. Bipolar systems have been developed (Versapoint, Gynecare, a division of Ethicon), which cause vaporization of tissue. Because these devices use a bipolar electrode, an electrolyte solution is needed as the distension medium. This system requires its own generator and will not work with other electrical generators such as Valley Lab's product. The advantage of these systems is the prevention of multiple tissue fragment generation as the technique vaporizes tissue. If pathologic diagnosis is desired, the tips can be used to isolate and remove a portion of tissue to be sent to the pathology department. When performed for menorrhagia, success rates of hysteroscopic myomectomy have been reported in approximately 80% of patients. Conception rates have been reported between 43% and 63% following hysteroscopic myomectomy in previously infertile women.

HYSTEROSCOPIC POLYPECTOMY Endometrial polyps are diagnosed in approximately 20% of women with abnormal uterine bleeding. This diagnosis usually requires either direct visualization of the polyp with a diagnostic hysteroscope or the visualization of a polyp using saline infusion sonohysterography. Hysteroscopically, endometrial polyps generally are more often pedunculated than submucous myomas and have a softer, fleshier appearance ( Fig. 47.5).

FIG. 47.5. An endometrial polyp is seen hysteroscopically as a soft fleshy growth.

Depending on the size of the endometrial polyp, it may be removed by one of several techniques. With a small pedunculated polyp, grasping the polyp stalk with a hysteroscopic grasper and rotating of the grasper to remove the polyp can be performed easily ( Fig. 47.6). With larger polyps or polyps with a larger base, a cutting loop resectoscope should be used. By resecting the base of the polyp with the loop electrode, the polyp subsequently can be removed intact with grasping forceps. If a small portion of the base remains, it can be destroyed with the loop electrode. Care should be taken not to destroy adjacent normal endometrial tissue to prevent postoperative adhesion formation.

FIG. 47.6. A rigid grasper is placed through the hysteroscope to grasp the base of an endometrial polyp for removal.

ASHERMAN SYNDROME Intrauterine adhesions or synechiae, Asherman syndrome, may develop when opposing endometrial surfaces are damaged and heal as a coalescing adhesion. This typically occurs when inflammation or infection persists after spontaneous first-trimester abortion and when estrogen production is low. Patients who have had late postpartum curettages for retained placental fragments are at high risk for the development of Asherman syndrome ( Fig. 47.7). Postpartum, estrogens of placental origin are metabolized rapidly and ovarian suppression continues for several weeks. With the absence of estrogen, the endometrium will fail to proliferate and reepithelialize damaged areas of endometrium following a curettage. Any concurrent inflammation or infection will increase the risk of intrauterine synechiae. Patients who have had elective abortions or postabortal curettages for any reason are also at risk for the development of intrauterine adhesions. The patient with Asherman syndrome may have amenorrhea if the extent of the adhesions is severe, but more commonly it manifests with normal menses or mild hypomenorrhea with recurrent spontaneous abortion or with infertility. The diagnosis of Asherman syndrome can be made by hysterosalpingogram, saline infusion sonohysterography, or direct visualization with diagnostic hysteroscopy.

FIG. 47.7. The hysteroscopic appearance of a retained placental fragment 1 month postpartum.

Intrauterine adhesions should be lysed with hysteroscopic scissors under direct visualization. This procedure is straightforward with mild to moderate cases of Asherman syndrome but can be very complicated in severe cases with complete obliteration of the endometrial cavity. In difficult cases, concurrent laparoscopy can be useful to guide the hysteroscope. Pockets of normal endometrium are used as guides to the lysis of thick adhesion bands. Both cornu of the uterus ultimately should be visualized, and the uterine contour should be relatively normal at the end of the procedure. Postoperatively, most practitioners will place patients with Asherman syndrome on high-dose estrogen replacement, Premarin (conjugated estrogens) 2.5 mg twice daily, or its equivalent, for 30 days to help induce proliferation of any remaining endometrium to reepithelialize the previous site of adhesions. Stenting of the uterus with a pediatric Foley catheter balloon filled with approximately 3 cc of saline for 1 week postoperatively also will prevent the damaged endometrial surfaces from coming immediately into contact with one another. With this foreign body in place, antibiotics should be used to prevent infection.

PROXIMAL TUBAL CANNULATION Approximately 20% of tubal occlusion occurs in the intramural portion of the fallopian tube at the uterotubal junction. Pathologically, proximal occlusion may be caused by debris plugs or fibrosis from inflammation, intraluminal endometriosis, or salpingitis isthmica nodosa. The diagnosis of proximal tubal occlusion often is made by hysterosalpingography. However, a definitive diagnosis cannot be made with this technique, because uterine muscle spasm falsely suggesting proximal tubal occlusion may be a consequence of the procedure itself. If pathologic proximal tubal occlusion is suspected, selective injection of contrast medium into the fallopian tubes may be performed either fluoroscopically or via hysteroscopy with concurrent laparoscopy. The latter technique has the advantage of simultaneously inspecting the distal portion of fallopian tube for damage. If the patient has both a proximal and a distal occlusion, the tubes are not candidates for surgical correction, and in vitro fertilization should be recommended. If selective injection of one or both of the fallopian tubes does demonstrate proximal tubal occlusion, then hysteroscopic cannulation of the fallopian tube can be performed immediately. Hysteroscopically, a flexible guidewire, 0.3 to 0.8 mm in outer diameter, is guided into the uterine cornu. The softness of this guidewire usually allows it to penetrate the occlusion while staying within the tubal lumen. When the guidewire is seen in the proximal portion of the fallopian tube laparoscopically, a soft Teflon catheter of 1.3-mm outer diameter is pressed over the guidewire until this is also within the proximal portion of fallopian tube. The guidewire can then be removed and methylene blue injected through the Teflon catheter to confirm tubal patency. Approximately 80% of proximal tubal occlusions can be cannulated successfully. Complications include perforation of the guidewire through the proximal portion of fallopian tube, but this rarely results in troublesome bleeding or infection. Intrauterine pregnancy rates have been reported as approximately 30% to 40% following proximal tubal cannulation.

SUMMARY POINTS The development of good hysteroscopy skills is important to gynecologists so that they may use the diagnostic and operative advantages these techniques offer. Careful attention to the surgical technique and particularly the distension medium employed is important in order to perform hysteroscopy safely. Diagnostic hysteroscopy is the gold standard for evaluation of the endometrial cavity. Operative hysteroscopy offers an outpatient surgical approach to many uterine problems including submucous and intracavitary leiomyomata, endometrial polyps, intrauterine synechiae, uterine septa, and proximal tubal occlusion. Endometrial ablation offers women an alternative to hysterectomy for the management of abnormal uterine bleeding or menorrhagia. SUGGESTED READINGS Instrumentation, Distension Media, and General Technique Bain C, Parkin DE, Cooper KG. Is outpatient diagnostic hysteroscopy more useful than endometrial biopsy alone for the investigation of abnormal uterine bleeding in unselected premenopausal women? A randomised comparison. Br J Obstet Gynecol 2002;109:805–811. Bronz L. Hysteroscopy in the assessment of postmenopausal bleeding. Contrib Gynecol Obstet 2000;20:51–59. Fay TN, Khanern N, Hosking D. Out-patient hysteroscopy in asymptomatic postmenopausal women. Climacteric 1999;2:263–267. Hucke J, De Bruyne F, Balan P. Hysteroscopy in infertility-diagnosis and treatment including falloposcopy. Contrib Gynecol Obstet 2000;20:13–20.

Shirk GJ, ed. The video encyclopedia of endoscopic of endoscopic surgery for gynecologists. St Louis: Medical Video Productions, 1994. Wiser F, Temper C, Kurz C, et al. Hysteroscopy in 2001: a comprehensive review. Acta Obstet Gynecol Scand 2001;80:773–783.

Endometrial Ablation Bratschi HU. Hysteroscopic endometrial resection. Contrib Gynecol Obstet 2000;20:121–136. Kochli OR. Endometrial ablation in the year 2000-do we have more methods than indications? Contrib Gynecol Obstet 2000;20:91–120. P> Lethaby A, Hickey M. Endometrial destruction techniques for heavy menstrual bleeding (Cochrane Review). Cochrane Database Syst Rev 2002;2:CDO01501. Ravi B, Schiavello H, Chandra P, et al. Safety and efficacy of hysteroscopic endomyometrial resection-ablation for menorrhagia. J Reprod Med 2001;46:717–723.

Hysteroscopic Septum Resection Daly DC, Maier D, Soto-Albors C. Hysteroscopic metroplasty: six years' experience. Obstet Gynecol 1989;73:201–205. Hickok LR. Hysteroscopic treatment of the uterine septum: a clinician's experience. Am J Obstet Gynecol 2000;182:1414–1420. Nawroth F, Schmidt T, Freise C, et al. Is it possible to recommend an “optimal” postoperative management after hysteroscopic metroplasty? A retrospective study with 52 infertile patients showing a septate uterus. Acta Obstet Gynecol Scand 2002;81:55–57.

Hysteroscopic Myomectomy Brandner P, Neis KJ, Diebold P. Hysteroscopic resection of submucous myomas. Contrib Gynecol Obstet 2000;20:81–90. Subramanian S, Clark MA, Isaacson K. Outcome and resource use associated with myomectomy. Obstet Gynecol 2001;98:583–587. Tercanli S, Kochli OR, Hoesli I, et al. Differentiation and management of endometrium abnormalities and leiomyomas by hydrosonography. Contrib Gynecol Obstet 2000;20:69–80. Vercellini P, Zaina B, Yaylayan L, et al. Hysteroscopic myomectomy: long-term effects on menstrual pattern and infertility. Obstet Gynecol 1999;94:341–347.

Hysteroscopic Polypectomy Chiu YK, Bhal PS. Role of hysteroscopy in the detection and extraction of endometrial polyps: results of a prospective study. Am J Obstet Gynecol 2002;186:1104. Gimpelson RJ. Hysteroscopic treatment of the patient with intracavitary pathology (myomectomy/polypectomy). Obstet Gynecol Clin North Am 2000;27:327–337. Varasteh NN, Neuwirth RS, Levin B, et al. Pregnancy rates after hysteroscopic polypectomy and myomectomy in infertility. Obstet Gynecol 1999;94:168–171.

Asherman Syndrome Al-Inany H. Intrauterine adhesions. An update. Acta Obstet Gynecol Scand 2001;80:986–993. Capella-Allouc S, Morsad F, Rongieres-Bertrand C, et al. Hysteroscopic treatment of severe Asherman's syndrome and subsequent fertility. Hum Reprod 1999;14:1230–1233. Westendorp IC, Ankum WM, Mol BW, et al. Prevalence of Asherman's syndrome after removal of placental remnants or a repeat curettage for incomplete abortion. Hum Reprod 1998;13:3347–3350.

Proximal Tubal Cannulation Deaton JL, Gibson M, Riddick DH, et al. Diagnosis and treatment of cornual obstruction using a flexible hysteroscope. Fertil Steril 1990;53:232–236. Dumesic DA, Dhillon SS. A new approach to hysteroscopic cannulation of the fallopian tube. J Gynecol Surg 1991;7:7–9. Thompson KA, Kiltz RJ, Koci T, et al. Transcervical fallopian tube catheterization and recanalization for proximal tubal obstruction. Fertil Steril 1994;61:243–247. Woolcott R. Proximal tubal occlusion: a practical approach. Hum Reprod 1996;11:1831–1833.

Complications Bradley LD. Complications in hysteroscopy: prevention, treatment and legal risk. Curr Opin Obstet Gynecol 2002;14:409–415. Jansen FW, Vredevoogd CB, van Ulzen K, et al. Complications of hysteroscopy: a prospective multicenter study. Obstet Gynecol 2000;96:266–270. Konig M, Meyer A, Aydeniz B, et al. Hysteroscopic surgery: complications and their prevention. Contrib Gynecol Obstet 2000;20:161–170. Propst AM, Liberman RF, Harlow BL, et al. Complications of hysteroscopic surgery: predicting patients at risk. Obstet Gynecol 2000;96:517–520.

Chapter 48 Incontinence Danforth’s Obstetrics and Gynecology

Chapter 48 Dee E. Fenner

Incontinence

DEFINITIONS OF TERMS AND DIFFERENTIAL DIAGNOSIS OF URINARY INCONTINENCE EPIDEMIOLOGY Prevalence and Social Impact Risk Factors Gender and Age Race Childbirth Smoking, Exercise, and Obesity EVALUATION OF URINARY INCONTINENCE History Physical Examination Urinary Diary or Bladder Record Urinalysis and Postvoid Residual Volume Stress Test Pad Test Specialized Tests TREATMENT OPTIONS Behavioral Techniques Electrical Stimulation Therapy Pharmacotherapy Surgical Management SELECTING THE APPROPRIATE SURGICAL PROCEDURE FOR STRESS INCONTINENCE SUMMARY SUMMARY POINTS SUGGESTED READINGS Definition of Terms Epidemiology Evaluation of Urinary Incontinence Nonsurgical Management of Urinary Incontinence Surgical Management of Urinary Incontinence

Urinary incontinence is a medical condition affecting 15% to 50% of adult women depending on the age and risk factors of the population studied. As we live longer and enjoy more active lives, increasing numbers of women will be affected. Costing approximately $12.4 billion dollars to treat women alone in 1995, urinary incontinence ranks in the top ten in expenditures for treatment of illnesses. While once considered a “natural” consequence of aging, further investigations have shown that urinary incontinence is seen in women of all ages, with varying degrees of severity and different etiologies. Unfortunately, incontinence often goes undetected and only half of the women suffering from this condition seek treatment. This is unfortunate because there are many types of effective therapies available including bladder retraining, pelvic floor exercises, and surgery. As health care providers for women, we are obligated to ask the question “Do you have difficulty controlling your bladder?” Once the condition is identified, specific questions targeted at identifying the type of incontinence, the frequency of loss, and the social impact for the woman can be addressed. Many times with the use of voiding diaries, detailed physical examinations, and simple bladder testing, the cause of incontinence can be determined and treated.

DEFINITIONS OF TERMS AND DIFFERENTIAL DIAGNOSIS OF URINARY INCONTINENCE Maintenance of urinary incontinence requires that many anatomic, neurologic, and psychological functions and relationships be intact. Therefore, it is not surprising that many conditions may affect urinary continence ( Table 48.1, Table 48.2).

TABLE 48.1. Drug effects on lower urinary tract function

TABLE 48.2. Causes of urinary incontinence

In order to communicate with the patient and other practitioners that evaluate, diagnose, and treat the various causes of urinary incontinence, a standardization of terms and description of conditions is required. A committee on terminology as part of the International Continence Society (ICS) has provided standardized terms and definitions for the classifications of functions of the lower urinary tract. The ICS definitions of terms and conditions of urinary incontinence are listed in Table 48.2 and Table 48.3.

TABLE 48.3. International Continence Society terms and definitions

Patients with symptoms of urinary incontinence are generally not concerned about the strict definition of their condition but are rightly focused on the cause and

foremost the treatment of their incontinence. As clinicians, we may or may not agree with the patient on the severity of her symptoms. Generally, when the symptoms are mild, the patient and practitioner are nearer in agreement than when the patient perceives her symptoms as severe. Therefore, it is useful to use standardized criteria or a validated quality-of-life measure to determine the impact of the incontinence condition on the patient's ability to perform her daily activities. Several such questionnaires exist and can be beneficial in the initial assessment as well as determining the benefit of therapy. An example of such a validated questionnaire is seen in Table 48.4.

TABLE 48.4. Urogenital distress inventory (UDI-6)

A patient complaining of stress symptoms may actually have cough-induced detrusor overactivity or mixed incontinence. Therefore the clinician must not only listen to the symptoms of the patient, but also document some sign of urinary incontinence. Signs of urinary incontinence are objective measures of urine loss evaluated in conjunction with the bladder state (empty or full) and conditions of leakage. Signs include urine loss documented by a voiding diary noting physical activity or bladder urgency linked to incontinence, a pad test, a cough stress test, or urine leakage at the time of urodynamic testing.

EPIDEMIOLOGY Prevalence and Social Impact Urinary incontinence is a common problem among women. The estimated prevalence ranges from 3% to 40% depending on the population studied and how the condition is defined. Approximately 10% to 30% of community-dwelling women between the ages of 15 and 64 years of age and 25% of women older than 65 years experience urinary incontinence. Approximately 50% of nursing home patients have urinary incontinence. One in five women with urinary incontinence will also suffer from fecal incontinence or “dual incontinence.” Urinary incontinence is two to three times more likely in women than in men. This is especially true for women under the age of 60, when stress incontinence is more likely the dominant type. As men and women age, the likelihood of urge incontinence or voiding dysfunction increases in both sexes. How urinary incontinence is defined determines the prevalence of the condition in the population or the incidence of the condition at any given time. Clinicians generally use the ICS definition of urinary incontinence that is “the involuntary loss of urine sufficient to be a social or hygienic problem for the patient.” For some women this may mean leaking urine once a day, while for others urinary incontinence occurring only periodically with exercise may be unacceptable. Survey studies of community-dwelling women using “daily,” “weekly,” or “most of the time” criteria for diagnosing urinary incontinence report prevalence ranges of 3% to 14%. This amount of leakage corresponds with the frequency of leakage found in women seeking treatment. Higher estimates of incontinence are based on broader definitions, such as “any incontinence in the prior year” or “leakage more than twice a month.” Many etiologies and factors contribute to urinary incontinence and many of the factors are variable over time and affect the presence and severity of urine loss. The strength of the pelvic floor muscles and urethral sphincter, the level of physical activity or high-impact exercise, the amount of fluid intake, or the level of bacteria are just of few of the variables that relate to urinary incontinence. While little information is known about the natural progression of incontinence severity, for women surveyed over a 2-year span, there was an 11% remission rate or resumption of continence. Risk Factors Many risks factors, conditions, or causes have been linked to the development of urinary incontinence. Unfortunately, few longitudinal and well-designed studies have been performed to separate the cause and effect for many of the risk factors. Why certain conditions and events such as childbirth and aging affect some women and not others is still unknown. Bump has developed a model ( Fig. 48.1) combining many of the known risk factors for the development of urinary incontinence. Which of these factors are most important and how they interact for a particular patient is not known but are areas of current investigation.

FIG. 48.1. Factors related to the development of urinary incontinence. (From Bump RC, Norton PA. Epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin North Am 1998;25:723.)

Gender and Age Urinary incontinence is two to three times more likely to occur in women than in men. This is especially true regarding community-dwelling adults under 60 years of age. Compared to urge incontinence, stress incontinence is more common in younger women and rarely occurs in men. Since childbirth is considered a major risk factor for the development of stress urinary incontinence, it is not surprising that this condition occurs more often in women than in men. But as men and women age, they both have more symptoms of bladder overactivity, urge incontinence, and voiding dysfunction. These symptoms develop such that the gender gap narrows as men and women approach the age of 70. Brown and colleagues have shown that 12.5% of women under 80 years of age reported daily incontinence and that for every 5-year increment in age beginning at age 80, there was a 30% increase in the prevalence of incontinence. While increasing age is consistently seen as a risk factor for urinary incontinence, the idea that urinary incontinence is a “natural” part of aging is not true. Like other conditions, such as heart disease, hypertension, and arthritis, incontinence is more prevalent as we get older. Like these other conditions, not all older people suffer from incontinence and those who do should be offered treatment and care and not left to “suffer this malady of old age.” Why urinary incontinence increases with age is not fully understood. With aging, approximately 1% of striated muscle mass is lost each year. Since the sphincter urethra muscle, as well as pelvic floor muscles play important roles in the maintenance of incontinence, it is not surprising that incontinence increases with aging. How aging, menopause, and decreasing hormone levels relate to urinary incontinence and other pelvic floor dysfunctions is poorly understood. Studies evaluating the use of oral combined estrogen replacement therapy have not shown improvement in or prevention of the development of urinary incontinence. Yet clinicians and some patients note improvement in function of the lower urinary tract, especially in decreasing nocturia, when using transvaginal estrogen replacement therapy. Further studies on the effect of estrogen on the blood supply, smooth muscle of the urethra and vaginal submucosa, and collagen content of the anterior vaginal wall are needed to determine the usefulness, correct route and preparation, and dosage of hormone replacement therapy that could be beneficial. Race

Racial differences in the prevalence and type of urinary incontinence have been reported but remain poorly understood. Fultz and associates reported a urinary incontinence prevalence of 23% in Caucasian women and 16% in African-American women in a national survey of 4,040 community-dwelling individuals over the age of 70. Smaller studies evaluating younger women do not include large enough numbers of African-American, Hispanic, or Asian women to make any conclusions about the prevalence or type of incontinence. In women seeking treatment for incontinence, African-American women appear to have more urge incontinence and detrusor overactivity on urodynamic testing than pure stress incontinence. Whether these differences are related to treatment patterns or true biologic differences is yet undetermined. It does appear, however, that there are racial differences in urethral muscle mass, urethral closure pressures, and urethral hypermobility between continent nulliparous African-American women and Caucasian women. These findings suggest morphologic and physiologic reasons for racial differences in the type of urinary incontinence. Childbirth Vaginal childbirth has been seen as the leading inciting factor in the development of urinary incontinence and pelvic floor dysfunction. Yet vaginal delivery does not lead to urinary incontinence in all women and women who have never delivered vaginally or have even been pregnant can develop urinary incontinence and pelvic floor dysfunction. Strong associations have been made between vaginal delivery and stress urinary incontinence. Large population studies have also shown vaginal childbirth is also a risk factor for developing urge incontinence. Women who deliver by elective cesarean section without ever laboring have a reduced risk of developing urinary incontinence compared to women who deliver vaginally, but a cesarean section is not 100% protective. In addition, the protective benefits appear to decrease after three pregnancies and repeat cesarean sections. Episiotomies, once thought to be protective, have not been shown to decrease or increase the risk of urinary incontinence, either immediately after birth or after several years. The majority of the damage to the pelvic tissues appears to occur with the first delivery. Other factors associated with increased risk from vaginal delivery include forceps deliveries, prolonged second stage, large-for-gestational-age birth weight, and large head circumferences. Not all of these obstetric factors have been found to be statistically or clinically significant in the literature and their etiologic role in the development of urinary incontinence remains to be determined. Vaginal childbirth causes damage to the pelvic floor tissues by stretching, tearing, and pressing the soft tissues against the bony pelvis causing anoxia and tissue necrosis. The muscles, nerves, and connective tissues of the pelvis can all be affected and damaged. Studies have shown damage to all of the tissues as measured by prolonged pudendal nerve motor latencies, decreased levator muscle strength, and decreased urethral pressures. Fortunately, with healing and muscle strengthening with Kegel exercises, the majority of the damage can be repaired, and continence regained in the first 6 months after delivery. Long-term pelvic floor trauma secondary to vaginal delivery and aging appear to combine as the major risk factors for urinary incontinence and pelvic floor dysfunction. Smoking, Exercise, and Obesity Women who are cigarette smokers or ex-smokers are two to three times more likely to have urinary incontinence than nonsmokers. Whether this association is related to chronic cough and increased abdominal pressure or tissue damage from tobacco toxins is not known. Physiologically, exercise has many positive and negative effects that may alter the balance between continence and incontinence. Increased abdominal pressure with high-impact exercise may overcome the continence mechanism in some women briefly causing situational episodes of leakage, but does not appear to increase their long-term risks for developing stress incontinence. Exercise that strengthens the lower extremities also strengthens the pelvic floor and enhances the continence mechanism. For the many health benefits, including strengthening the pelvic floor, women should be encouraged to exercise or take daily walks. Obesity has been shown to be more common in women with both stress and urge incontinence than in continent women. In addition, morbidly obese women who lose weight have shown objective as well as subjective cure of their incontinence. Consideration should be given to surgically managing their morbid obesity over surgically attempting to correct their stress incontinence.

EVALUATION OF URINARY INCONTINENCE History The first step in evaluating a patient with urinary incontinence is to pose the question to the patient. Often a brief, directed review of bladder and bowel function can elicit symptoms. Open-ended questions such as “Could you please tell me about problems you are having controlling your bladder?” can be a useful initial approach. If the answer is affirmative for incontinence symptoms then an incontinence evaluation in initiated. The evaluation begins with more direct questioning concerning the patient's symptoms, timing, and severity of incontinence. Frequently asked questions listed in Table 48.5 can help distinguish the type and severity of a patient's incontinence. Questions aimed at type of protection or pad used and number of pads used in a 24-hour period can indicate the severity of the problem. In addition, questions regarding lifestyle changes such as decreasing fluid intake or avoiding physical activities reflect the impact of incontinence on the patient's quality of life.

TABLE 48.5. Frequently asked questions

The presence of symptoms associated with both the storage and the emptying phases of the bladder cycle should be determined. Specific questions should ask about irritative symptoms (urgency or frequency), incontinence (stress or urge), and voiding dysfunction (urinary retention or postvoid fullness). A detailed history of any prior treatment of the genital or urinary tract, including surgeries should be obtained. The surgical history should contain details of any prior procedures for incontinence, including indications, technique, and type of suture used. Any complications (i.e., prolonged catheter use after surgery) or the development of new symptoms (i.e., urinary frequency and urgency following surgery) should be noted. An obstetric history including route of delivery, episiotomy, lacerations, the use of forceps or vacuum, birth weights, and the length of the second stage of labor can determine the patient's risks associated with childbirth. A history of pelvic cancer therapy should include the dates and findings of any cystoscopic examinations. Other comorbid conditions affecting the lower urinary tract, including cigarette smoking, neurologic and endocrine diseases, as well as family history of urinary incontinence and pelvic organ prolapse may direct the clinician toward one type of urinary incontinence or need for further diagnostic evaluation. Any functional components such as difficult ambulation, declining cognition, or severe fecal impaction or constipation should be addressed prior to further evaluation or treatment. Medications currently used should be reviewed in detail as many medications have adverse effects on the lower urinary tract (see Table 48.1). Unfamiliar medications should be verified. Diuretics and a-antagonists, two classes of medication commonly used to treat hypertension, may cause incontinence in some women. When incontinence is temporarily related to onset of use or dosage increase, it may be reasonable to use alternative medications that have less effect on the lower urinary tract. Office evaluation should follow a stepwise progression ( Fig. 48.2). Initial evaluation should focus on the severity of urinary leakage and the patient's goals for resolution of the problem. Some women may simply wish to regain sufficient bladder control to participate in certain social or athletic activities, while others desire complete continence under all circumstances. The extent of the evaluation and the proposed therapy should match the patient or her caregiver's expectations and goals.

FIG. 48.2. Evaluation algorithm. (From Fantyl JA, Newman DK, Colling J, et al. Urinary incontinence in adults: acute and chronic management. Clinical Practice Guidelines Number 2, 1996 Update. Rockville, MD: Agency for Health Care Policy and Research, US Dept of Health and Human Services; 1996. AHCPR publication 96-0682.)

Physical Examination Beyond the standard speculum and bimanual examination, the specific goal of a urogynecologic physical examination is to assess vaginal supports, strength of the pelvic floor, the anatomic position of the anterior vaginal wall, and the stability of the urethrovesical junction (UVJ). The examination for pelvic organ prolapse and the use of the standardized Pelvic Organ Prolapse Quantitative System (POP-Q) can be found in Chapter 45. In addition to noting any prolapse of the vaginal walls, uterus, or vaginal apex, the resting location and mobility of the UVJ with cough or strain is noted. In most cases, the urethral hypermobility can be documented by having the patient strain and the clinician noting prolapse of the distal anterior vaginal wall. If uncertain, and deemed important by the clinician, then a sterile Q-tip may be placed in the urethra or transperineal ultrasound used to document mobility. Not all women with a hypermobile urethra have stress incontinence and not all women with stress incontinence have a hypermobile urethra. African-American women appear to have more urethral hypermobility than Caucasian women, yet they have less stress incontinence because of higher urethral pressures. The importance of noting a hypermobile urethra is useful when considering certain treatment options. Both surgical and nonsurgical techniques that work by stabilizing or supporting the UVJ work best when the urethra is hypermobile allowing for repositioning or stabilization. Neurologic evaluation of the incontinent female includes assessing intact perineal sensation that primarily represents the S2 dermatome. The bulbocavernosus reflex or anal wink is used to assess the sacral reflexes. A Q-tip swab is gently used to touch the skin overlying the bulbocavernosus muscles in the labia majora or perianally. A positive reflex indicates an intact pudendal nerve and L5-S5 motor neurons causing spontaneous contraction of the external anal sphincter. As with other reflexes, the sign is positive or negative. Hyperreflexia may be seen in upper motor lesions. Unlike the corresponding cremaster reflex in men, approximately 10% of normal, neurologically intact women will not demonstrate this reflex or the response is too weak to see. If, however, in the context of urinary or fecal incontinence or with other neuromuscular abnormalities of the buttocks or lower extremities, the absence of the bulbocavernosus reflex should alert the clinician to possible pathology in the sacral or cauda equina regions of the spinal cord. A cough stress test should be performed, noting the position of the patient as supine or standing and whether her bladder is full or empty. Documentation of the sign of stress incontinence is adequate for diagnosis in most patients and avoids performing further, more complex and expensive testing. Other conditions that can be detected during physical exam include hypoestrogenism, levator atrophy, and fecal impaction, which can all affect bladder control. If comorbid conditions such as edema or declining cognitive function are contributing to the patient's incontinence, then an expanded physical examination including mental status assessment, manual dexterity, and ability to ambulate is warranted. Urinary Diary or Bladder Record While not always available at the time of the initial history and physical, the urinary diary is an essential part of the of incontinence evaluation ( Fig. 48.3). This tool allows the patient to record voided volumes, leakage episodes including amounts and associated activities, and the type and volume of fluid intake. Although a 3- to 7-day diary is optimal, evaluation of the events of a typical 24-hour period can be very helpful clinically. By simply keeping the diary, patients can notice how the amount and types of fluid they drink, as well as when they drink impact incontinence. In addition, the clinician can assess the severity of the leakage and begin to formulate an opinion of the type and etiology of the patient's incontinence and need for further evaluation. Upon review of the diary, the practitioner can suggest simple behavioral interventions such as fluid intake and medication adjustments.

FIG. 48.3. Voiding diary.

Urinalysis and Postvoid Residual Volume Sterile urine and a normal postvoid residual volume should be documented in all women. This assessment can be performed using a small-gauge catheter within 20 minutes after voiding. While there is no strict definition of a normal postvoid residual, the volume should generally be less than 100 mL. The catheterized urine specimen can then be used to rule out infection. While dipstick methods are available to detect bacteriuria and pyuria, the diagnostic accuracy is variable and dependent on the prevalence of urinary tract infections in that population. Therefore if the dipstick is positive or the patient's symptoms are consistent with a urinary tract infection, obtaining a urine culture is recommended. Alternatively, an approximation of postvoid residual volume may be obtained with the use of an ultrasound specifically designed for determining bladder volume and using a clean catch urine sample to rule out infection. Infection should be treated and sterile urine should be demonstrated before further incontinence evaluation. If the residual volume is elevated, the test should be repeated on a separate visit. Persistently high residual volume should be evaluated. Conditions affecting bladder emptying such as peripheral neuropathies or pelvic organ prolapse outside the hymen are common etiologies for elevated postvoid residuals. Unlike men who frequently suffer urinary retention due to obstruction by an enlarged prostate gland, women rarely have obstruction unless they have undergone prior surgery for urinary incontinence that is obstructing urine flow. Stress Test The stress test is used to observe urinary leakage associated with straining or coughing, and is best performed with a full bladder in the standing position. Often a series of coughs is required to demonstrate incontinence. The patient should be positioned over a pad or paper towel, wrapped with a sheet or gown, and reassured that the physician wants to see urine leakage. Many times patients are embarrassed and reluctant to leak urine in the office setting in front of ther people. Putting the patient at ease and respecting her privacy helps to obtain a positive test and preserves patient dignity. A positive test confirms the sign of stress incontinence. A negative test does not completely rule out stress incontinence, but when performed with a full bladder and in the standing position the diagnosis of stress incontinence is less likely. If the patient also suffers from pelvic organ prolapse, then the prolapse should be repositioned to a “normal” anatomic position with a pessary or large vaginal swabs during the stress test. Pad Test A pad test is an objective test used to document the presence and amount of urinary incontinence. The test requires that the patient wear a preweighed pad while she performs activities usually associated with her incontinence. A 1-hour pad test is typically conducted. Generally, the test begins with the patient emptying her bladder and then drinking a known amount of water, typically 500 cc. Alternatively, the bladder can be filled with a set amount or to maximum capacity. The pad is weighed after 1 hour of activity. While pad testing has been found to be reproducible, in terms of objective documentation of incontinence and ability to evaluate pre- and post-treatment incontinence, its use in routine clinical evaluation is limited. A specific use for the pad test is when the clinician is unsure if the symptoms of incontinence

are due to urine loss or secondary to vaginal secretions or sweating. By giving the patient phenazopyridine hydrochloride (Pyridium) to turn her urine orange, a pad test can confirm urine loss. After taking a careful history, performing a directed pelvic examination including a postvoid residual, ruling out infection, and performing a stress test, an accurate diagnosis can be made on the type of urinary incontinence for approximately 80% of women. Initial management can begin as outlined in Table 48.6.

TABLE 48.6. Management options after basic evaluation

Specialized Tests The need and use of specialized tests for the diagnosis and management of patients with urinary incontinence is far from standardized in clinical practice. While some clinicians feel advanced testing is useful, the reproducibility of certain tests such as multichannel urodynamics, leak point pressures, and voiding studies have led many clinicians and expert panels to use specialized testing in selected cases. The Agency of Health Care Policy and Research under the jurisdiction of the United States Department of Health and Human Services has recommended the use of specialized testing according to the guidelines listed in Table 48.7. Clinicians caring for patients with urinary incontinence recognize the difficulties in trying to determine the exact cause of leakage in all patients. When therapies can be used which effectively treat both urge and stress incontinence or there is acceptable cost and morbidity for the individual patient, determining the exact etiology is less critical. However, prior to surgical interventions, because of cost and morbidity, cystometry is recommended to document stress incontinence.

TABLE 48.7. Criteria for Further Evaluation of Urinary Incontinence a

If a patient presents with moderate to severe pelvic organ prolapse outside the hymen and has no subjective urinary incontinence, she should be tested for occult or potential stress incontinence. Studies have shown that with the prolapse reduced and straightening of the urethra, incontinence can “occur” in 30% to 50% of patients with prolapse. If urinary incontinence is found, then an antiincontinence surgery should be performed at the time of surgical correction for the patient's pelvic organ prolapse. Which antiincontinence surgery is best for occult incontinence is not known. Consultation should be considered for cystometric testing and surgical management for patients who have not responded to initial treatments or have failed prior incontinence procedures. Cystometry Cystometry is the test used to measure the pressure–volume relationship during bladder filling. The main purpose of cystometry is to diagnose or rule out detrusor contractions. A catheter attached to a vertical fluid column, a pressure transducer, or a microtip catheter measures the bladder pressure. Single channel or simple cystometry implies that only the bladder pressure is being monitored during the study. A multichannel or complex cystometry includes a second catheter for monitoring intraabdominal pressure. During a multichannel study, true detrusor pressures are calculated by subtracting the intraabdominal pressure from the bladder pressure. While filling the bladder, the compliance (bladder pressure divided by the infused volume), the first desire to void, maximum bladder capacity, and the presence of detrusor contractions are documented. Stress incontinence is diagnosed with the patient coughing or by performing a Valsalva maneuver and noting urine loss in the absence of a detrusor contraction. Simple Cystometry or Single Channel Cystometry Performing simple cystometry can easily be done in the office of most gynecologists using a red rubber catheter, a three-way stopcock, intravenous tubing, and sterile water or saline. The patient is placed supine or if possible seated at a 45-degree angle. The sterile catheter is inserted and a postvoid residual volume is obtained. By attaching a three-way stopcock to the end of the catheter, a tube for filling and a tube for measuring detrusor pressures can be used. By taping the pressure tubing vertically to a 100-cm measuring stick taped to an intravenous pole, and then having the zero mark at the level of the bladder, the bladder is filled in 50- to 100-cc increments. After filling, the stopcock is turned off to the filling catheter. The pressure in the tubing represents the pressure in the bladder, including the intraabdominal pressure. The patient is instructed not to talk or perform any maneuvers that will increase her intraabdominal pressure. A rise in the fluid column should then represent a rise in bladder pressure secondary to abnormal compliance or a bladder contraction ( Fig. 48.4).

FIG. 48.4. Simple office cystometrogram. The zero mark of a meter stick is aligned with the upper margin of the patient's symphysis pubis, and a Foley catheter is in place for infusion of fluid. (Adapted from the American College of Obstetricians and Gynecologists, Bent AE, ed. Urogynecologic evaluation, endoscopy, and urodynamic testing in the symptomatic female. Washington, DC: ACOG Audiovisual Library, 1990.) An alternative method for simple cystometry uses a specialized filling catheter with a pressure transducer that relays the bladder pressure to a recording device. Several types are commercially available. Multichannel Urodynamics Approximately 10% of patients require multichannel urodynamic studies for diagnosis and treatment of urinary incontinence. Like all diagnostic tests, the clinician must understand the information gained from the procedure, the accuracy of the measurements, and how the information obtained will alter diagnosis, prognosis, or management of the condition. Multichannel urodynamics, once considered the gold standard for the diagnosis of detrusor overactivity and

stress incontinence, have been questioned regarding their reproducibility and cost-effectiveness in the routine evaluation of urinary incontinence. As outlined in Table 48.7, if initial assessment and treatment are not successful, further evaluation with multichannel urodynamic studies is indicated. Table 48.8 lists the common terms and definitions for urodynamic studies. Multichannel urodynamics are performed using a specialized machine or unit that records the bladder and intraabdominal pressures, subtracts the intraabdominal pressure from the bladder pressure to determine true detrusor pressure, and provides a report. Most units also have the ability to record urethral pressures and electromyography of the pelvic floor or urethra during the multichannel study. When reporting a multichannel cytometric study, the filling medium, rate of bladder fill, and position of the patient during the study should be noted. A typical tracing for a patient with detrusor overactivity is seen in Fig. 48.5. Note the rise in detrusor pressure without a rise in the abdominal pressure that is seen with a cough.

TABLE 48.8. Cystometric definitions

FIG. 48.5. Cystometric tracing. Uroflow Studies and Voiding Cystogram Uroflowmetry measures the flow rate visually, electronically, or with the use of a disposable unit and is measured in mL per second. An electronically generated flow curve is considered helpful in identifying abnormal flow patterns such as Valsalva voiding (increasing vesicle pressure by pushing rather than having a detrusor contraction). Uroflowmetry does not help diagnose the type of urinary incontinence and is not a routine part of the evaluation of urinary incontinence. A voiding cystometrogram is a uroflow performed with a catheter in the bladder to record bladder pressures. For a patient with urinary retention, a voiding cystogram may help distinguish obstruction from detrusor weakness. A voiding cystometrogram may be useful in the evaluation of patients with elevated postvoid residuals, neurologic conditions such as diabetes that may affect bladder emptying or inability to void after extensive pelvic surgeries, or antiincontinence procedures. Urethral Pressure Profilometry Urethral pressure profilometry (UPP), a urethral function test, measures resting and dynamic (with cough) pressures in the urethra. Clinically, UPP is used by some clinicians to assess the patient for a weak sphincter or intrinsic sphincter deficiency (ISD). Patients with ISD have been found to have less success with antiincontinence surgeries aimed at repositioning or elevation of the bladder neck and therefore the diagnosis of ISD would determine which surgical procedure was used. Sphincteric function can also be assessed by measuring abdominal or vesical pressure needed to overcome urethral resistance or abdominal or vesical leak point pressure. Clinically, the usefulness of the UPP versus abdominal leak point pressure has not been adequately studied. Electromyography Both surface and needle electrodes have been used to record urethral sphincter and pelvic floor muscle activity during urodynamic studies. Technically difficult to reproduce, except in a few urodynamic laboratories which have expertise in electromyography (EMG), the clinical usefulness is questionable. The goal of EMG is to evaluate relaxation of the pelvic floor and urethra at the time of voiding. Many patients, while seated in the urodynamics chair with catheters in place and persons in the room, will have difficulty relaxing their pelvic floor and may give a false-positive test for EMG activity and poor relaxation during voiding. Pathologically, patients with detrusor sphincter dyssynergia (DSD) have a detrusor contraction to void but the urethra and periurethral striated muscle contract preventing or decreasing urine flow. DSD is caused by an upper motor neuron injury, most commonly multiple sclerosis. Cystourethroscopy Cystourethroscopy is not recommended in the basic evaluation of urinary incontinence. For patients with bladder pain, sterile microscopic or gross hematuria, new-onset irritative symptoms without infection, or suspected intravesical foreign body, a cystoscopy is warranted to rule out neoplasia.

TREATMENT OPTIONS Three major categories of treatment for urinary incontinence include behavioral, pharmacologic, and surgical. Treatment options are often used in combination, such as behavior modification with timed voiding in conjunction with pelvic floor exercises for strengthening the pelvic floor. Realistic expectations for success and any risks of the therapy must be reviewed with the patient before initiating treatment. In choosing treatment for one condition, the potential for exacerbating another condition must be considered. For example, surgery performed to cure minor stress incontinence may dramatically worsen urge incontinence. Behavioral Techniques When taught or provided by knowledgeable health care providers, behavioral techniques decrease the frequency of urinary incontinence in most individuals, have no reported side effects, and do not limit future treatment options. Behavioral therapies are most successful when persons can actively participate in their training program, but they can be provided effectively by caregivers for patients with cognitive and motor deficits. Bladder Retraining Behavioral modification is defined as the analysis and alteration of the relationship between the patient's symptoms and her environment for the treatment of maladaptive voiding patterns. Because most women either decrease their fluid intake or urinate more frequently in order to avoid leakage, behavior modification is helpful in virtually every patient with urinary incontinence. By inquiring about self-directed modifications the patient has made in her diet, fluid intake, or activity level, the practitioner can gain insight not only into the patient's diagnosis, but save time and have a more direct impact by not recommending modifications that the patient has already tried. Simple techniques may be discussed after review of the urinary diary, including adjustments in the type and volume of fluid intake. Older patients frequently have nocturnal diuresis. The diaries of these patients reflect several episodes of nighttime voiding, possibly with urinary loss while en route to the toilet. However, the important diary clue is a large bladder volume—sometimes larger than the patient's first morning voided volume. Rather than adding a nighttime bladder medication, a late afternoon supine rest or leg elevation can help the diuresis. Consuming large volumes of fluid intake, often in an attempt to aid weight loss, is a problem and should be addressed. Women should be counseled to drink and titrate their fluids according to urine color and number of voids. Generally the urine should be a light yellow. Women should aim for six to seven voids in a 24-hour time period. Inadequate hydration may be found in women who restrict intake as a method of controlling their incontinence symptoms. For some women, this may contribute to urgency and frequency as the concentrated urine is more irritative to the bladder than dilute urine. Also, caffeinated beverages increase urinary urgency and frequency and may exacerbate incontinence episodes. Timed voiding is effective for approximately 60% of women with detrusor overactivity. The patient is instructed to void at a specific interval by the clock during waking hours. By examining the voiding diary, a voiding interval that is slightly longer than the current interval that results in leakage is used as a starting time. For example, if a patient is voiding every 30 minutes to avoid leakage, she is encouraged to try and hold her urine for 1 hour and to use the clock to monitor her frequency. Over a period of 6 to 8 weeks, the intervals are gradually lengthened until the patient is able to urinate without leakage at 3- to 4-hour intervals. An alternative technique for women who are going to the bathroom every 30 minutes to 1 hour to avoid incontinence is to add 15 minutes of waiting time to when they feel the first desire to void. By asking the woman to sit and squeeze her pelvic floor when she feels the urge to void and to try and wait 15 minutes before voiding, she can again increase her times between voids to 3- to 4-hour intervals within 6 to 8 weeks. These techniques have significant advantages, including the lack of systemic adverse effects and the minimal cost. It is an ideal initial intervention for many women. Written materials are helpful for optimal understanding and compliance. For women with cognitive impairment or decreased mobility, timed-voiding prompted by a caregiver can also be effective in decreasing the number of incontinent episodes and the amount of leakage. A bedside commode can facilitate voiding and help the patient avoid falls. These behavior modification techniques are often done in conjunction with pelvic floor exercises and have been found to be equal to or more effective than pharmacotherapy for some women with urinary incontinence. Pelvic Floor Muscle Training Pelvic floor muscle training is defined as “repetitive selective voluntary contraction and relaxation of specific pelvic floor muscles.” Most

women are familiar with the concept of pelvic floor exercises often termed “Kegels” after the gynecologist who first introduced the concept in the 1950s. Women with weakened pelvic floor muscles may benefit from muscle-training protocols. These protocols are helpful for patients with detrusor overactivity or stress incontinence. Muscle rehabilitation aims not only to strengthen but to activate the muscle contraction at the appropriate time. Miller and colleagues have shown that with proper instruction as to the timing of the pelvic floor contraction with a cough or urge to void, women can have immediate success in decreasing their incontinence before any strengthening of the pelvic floor muscles takes place. Like exercises aimed at strengthening other striated muscles, measurable improvements in strength occur after 4 to 6 weeks of training. Patients frequently have difficulty identifying muscles used in pelvic floor contractions and are unaware if they are contracting the right muscles and how to perform the exercises. Assessment as part of the exam for pelvic floor dysfunction can help the patient isolate the right muscles. The examiner places an index finger with light pressure on the posterior fourchette and asks the patient to Kegel, or squeeze her pelvic floor as if she is trying to hold urine or not pass flatus. Another method is to have the patient try to stop her urine flow. This method should only be used as a test and not for performing the exercises. If the patient is able to stop or slow the urine flow then she is squeezing the correct muscles and a strengthening program can be initiated. Women with profound weakness or apparent paralysis generally require a supervised exercise program and biofeedback or electrical stimulation therapy. Biofeedback uses a pressure balloon, surface electrodes that record pelvic floor muscle activity, or another measuring device that, visually or by sound, alerts the patient that she is squeezing the proper muscles. Over time, the patient is able to isolate and contract the muscles without the auditory or visual cue and a formal exercise program can be initiated. Strength can be gained through a variety of exercise programs. There is little scientific evidence to suggest the superiority of any specific muscle-training regimen; however, it is likely that general principles of muscle strengthening apply. Physical therapists and nurses may be helpful in initiating and supervising the rehabilitation program. Simple written or verbal instructions may be sufficient to teach patients how to perform pelvic muscle exercises properly. Performing these exercises supine decreases the gravitational forces and may make the exercise easier for the patient to perform. With improved strength, she may advance to exercising in the upright position. Patients should be instructed to perform 10 to 20 ten-second pelvic floor contractions three or more times per day. A minimum of 30 contractions per day for at least 6 weeks is usually required to achieve a detectable beneficial effect. Too vigorous a program should be discouraged as muscle soreness could result in dyspareunia and abandonment of the exercise program. Older women may need a longer training period. These exercises should be performed indefinitely to prevent recurrence of incontinence. An adjunct to biofeedback and standard pelvic floor exercises is the use of weighted vaginal cones. The cones, weighted 20 to 100 g, are used as part of a structured, progressive, resistive exercise program. Women insert the plastic coated cone into the vagina much like a tampon, and attempt to keep the cone from falling out by squeezing the pelvic floor. The exercise is performed for 15 to 20 minutes once a day. Once a weight has been successfully held for 3 days, the next higher weight is used. For women with stress incontinence, weighted cones have been found to be equal to other pelvic floor exercise programs where patients are monitored closely by nurses or physical therapists. Cones may be preferred by some women because they are done at home, do not require nursing or physical therapy visits, and they are relatively inexpensive. Regardless of the exercise method used, as the muscles are strengthened, the patient must also learn to use her muscles correctly to inhibit urinary urge and reduce stress incontinence episodes. She must learn to contract these muscles firmly in advance of a cough, sneeze, or similar anticipated stress. Strength without timing or timing without strength will result in clinically unsatisfactory results. Once the muscles are strengthened and the patient has learned to squeeze at the appropriate time, the extent of exercise needed to maintain the effect is not fully known. Like many exercise programs, the patient's dedication to the activity likely will diminish with time. Electrical Stimulation Therapy Transvaginal electrical stimulation produces a contraction of the levator ani, external urethral sphincter, and anal sphincters, accompanied by a reflex inhibition of the detrusor and is dependent on a preserved reflex arc through the sacral micturition center in order to work. Electrical stimulation with a transvaginal probe reduces detrusor overactivity and stress urinary incontinence in approximately 50% to 70% of affected women. Two randomized trials comparing electrical stimulation to placebo or sham device have been performed. Sand and colleagues, using high frequency probes, reported objective “cure” or improvement in 48% of patients with stress urinary incontinence versus 13% improvement in the control group. Brubaker and co-workers reported on 121 women with urge, stress, and mixed incontinence. Participants used either a sham device or underwent electrical stimulation using a low-frequency, 20-Hz, 2-second to 4-second work–rest cycle for 20 minutes twice daily for 8 weeks. Detrusor overactivity was cured in 49% of women receiving electrical stimulation, but there was no statistically significant change in the percentage with detrusor overactivity in the sham device group. There was no statistical improvement in the women with stress urinary incontinence. Typically, a transvaginal probe is used once or twice daily for 15 to 20 minutes. Symptom relief generally occurs within 6 to 8 weeks and continues with reduced stimulation (e.g., three times weekly). Like timed voiding and muscle training, this treatment is free of systemic adverse effects. The main drawback is the cost of rental or purchase, which may be out-of-pocket expense for many patients. Many third-party providers and Health Care Financing Administration (HCFA) have approved the use of electrical stimulation therapy for urinary incontinence making the treatment more affordable. Several devices approved by the U.S. Food and Drug Administration are available, and treatment can be home-based or, less commonly, office-based. Women with pacemakers should avoid this therapy unless other options are not available. Implantable electrostimulators, such as InterStim, are now being used for patients with refractory urge incontinence. Before considering InterStim, all other treatment options should have been used properly, including transvaginal electrostimulation. InterStim requires insertion of a wire electrode, through an incision in the back and placement of a permanent stimulator, to the nerve root of S2, S3, or S4. The nerve root is stimulated and theoretically aberrant afferent signals to the spinal cord are blocked, decreasing bladder overactivity. Improvement of urgency and frequency symptoms ranges from 60% to 83%. Vaginal and Urethral Devices An increasing selection of vaginal and urethral devices is available and appears to be the fastest-growing type of incontinence therapy. Pessaries modified for use in incontinence provide additional suburethral pressure and are safe and reasonably effective. For some women, the placement of a vaginal tampon may provide adequate urethral support to prevent leakage during physical activities such as playing golf or tennis. Both external and internal urethral barriers are becoming increasingly available. While not optimal for all patients, these devices can provide effective treatment for patients who leak only under known circumstances or perhaps to improve their quality of life while awaiting surgery. External barriers range from patches that fit over the urethral meatus to small suction cups that limit urine loss. Internal urethral barriers are available, but they have the distinct disadvantage of causing urinary tract infections and hematuria. Although effective in preventing urine loss, the devices are less than ideal because of these adverse effects. In addition, many women are not comfortable with the idea of placing something into or over their urethra or are not physically able to do so. The efficacy, cost, comfort, and adverse effects including urinary tract infection and skin irritation for each of these barriers must be considered. More experience is needed before recommendations can be made. Pharmacotherapy Detrusor Overactivity or Urge Incontinence No ideal medication for the treatment of incontinence is available, primarily because of the unwanted side effects. In general, drug treatment is used for detrusor overactivity and is aimed at reducing inappropriate detrusor contractions ( Table 48.9).

TABLE 48.9. Pharmacologic therapy for detrusor instability Drugs for detrusor overactivity are typically anticholinergic agents. Thus, the adverse effects include xerostomia or dry mouth, blurred vision, drowsiness, and constipation. Narrow-angle glaucoma is an absolute contraindication to the use of this class of medications, and they should be used cautiously in women with cardiovascular disease. While other anticholinergic medications have been used clinically, oxybutynin chloride, propantheline, and tolterodine are the only medications in this class with scientific evidence of efficacy for treating detrusor overactivity. Oxybutynin chloride is available in generic formulations and in a longer-acting, once-a-day nongeneric formulation. Tolterodine, available in a once- or twice-a-day dosing form, may have fewer adverse effects when compared to generic oxybutynin chloride. Because of the high side-effect profiles, the medications should be started at very low doses and increased according to the patient's symptoms of incontinence and adverse reactions. The medications can be self-titrated over a period of 6 to 8 weeks with written instructions. Since constipation is a common comorbid condition in the elderly, recommending a stool softener when initiating therapy should be considered. Long-term compliance with incontinence medications is poor and usually ranges from 40% to 50%. This low compliance should be considered when selecting this mode of therapy and follow-up should be scheduled. The effectiveness of pharmacotherapy is poor in women whose incontinence is related to insensible loss (leakage that occurs without the patient's knowledge). Stress Incontinence Medications that stimulate a-receptors are used to increase urethral tone in an effort to reduce stress incontinence episodes. These medications stimulate a-receptors throughout the body (e.g., in the cardiovascular system) and may cause hypertension. Phenylpropanolamine and pseudoephedrine are the first-line pharmacologic therapy for women with stress incontinence. Women with mixed urinary incontinence may benefit from imipramine hydrochloride, a tricyclic antidepressant. This medication offers a combination of anticholinergic medication for detrusor suppression and a-receptor sympathomimetic activity to increase urethral pressure. Estrogen The role of oral or vaginal estrogens in the management of overactive bladder or stress incontinence remains unclear. Prospective studies have shown little effect of oral combined estrogen/progestin therapy on lower urinary tract function, including prevention or improvement in urinary incontinence. Meta-analysis of prior studies has shown some improvement in nocturia secondary to estrogen therapy in postmenopausal women. While many clinicians feel strongly that preoperative transvaginal estrogen therapy improves surgical outcome and tissue healing, further evaluation is needed.

Surgical Management Surgical management of urinary incontinence has been reported since the early 1900s. With an aging United States population, there will be greater demand for surgical and nonsurgical treatment of incontinence. Estimates indicate the demand for services will increase 10.9% for women between the ages of 30 to 58 years and 80.7% for those between 60 to 89 years by 2020. Although hundreds of operations have been described, today there are basically three types of procedures used: a suburethral plication, a shelf or repositioning type of procedure, or a pubovaginal sling. The large number of reported procedures indicates that no single type of operation will cure all patients. Table 48.10 presents the studies that have included objective testing and highlights the handful of studies that have been prospective, randomized, controlled trials. Based on the clinical findings in each individual patient, the surgeon must select the procedure or combination of procedures that will most likely alleviate the condition. Selection of the antiincontinence procedure is determined by many factors including need for other concomitant abdominal or vaginal procedures to correct pelvic organ prolapse, history of prior repairs, urethral hypermobility, and overall health and surgical risks for the patient.

TABLE 48.10. Urinary incontinence procedures: objective cure rates

Significant limitations in reviewing data on the surgical treatment of stress incontinence have been encountered. There is no standardized pre- or postoperative evaluation for the surgical treatment of stress incontinence. Variable lengths of follow-up, inconsistent outcome criteria, and reporting biases of operating surgeons make comparison of many clinical series nearly impossible. The 1996 revision of the Agency for Health Care Policy and Research Urinary Incontinence Clinical Practice Guidelines summarized that “the surgical literature is deficient in standards for describing the patient population, the type of incontinence, the method for accurate diagnosis, the techniques of the surgical procedure, or the outcome in different domains.” In 1997, the American Urologic Association completed a comprehensive, evidence-based review of the surgical literature on the treatment of stress incontinence. Minimal criteria for follow-up interval and reporting outcomes resulted in the elimination of over half the reported clinical series from the database. There were no standardized pre- and postoperative evaluations and uniformity in the terminology describing the surgical procedures. Limited, poorly defined outcome measures (e.g., “dry”, subjective vs. objective variables, “improved”—1 pad/day vs. 3 pads/day) with little accuracy in morbidity rates made comparison of the surgical literature impossible. Although our understanding of the pathophysiology of stress incontinence is incomplete, most accept the current concepts of urethral hypermobility and intrinsic sphincter dysfunction as causes, whether alone or in combination. Einhorning believed that the proximal urethra descended beyond the pelvic diaphragm and thus beyond the sphere of influence of the abdominal cavity, and restoration of continence required the urethra be drawn back within the abdominal cavity. Reexamination of the anatomy of the female pelvis from both a structural and functional perspective has led to the current concept that poor support of the proximal urethra as a result of relaxation or detachment of the anterior vaginal wall is responsible for the hypermobility associated with stress incontinence. It is recognized that this is an incomplete explanation since the majority of women with hypermobility do not have stress incontinence. Thus, there must also be some abnormality in the function of the urethra in women with stress incontinence. Whether this dysfunction is a primary neurologic defect or muscular defect is not yet determined, but there is strong evidence that denervation plays an important role. Snooks and co-workers demonstrated that the denervation effects of vaginal birth and subsequent pudendal neuropathy are significantly associated with GSI. A subset of women develop stress incontinence despite excellent support of the UVJ. The term “intrinsic sphincter deficiency” (ISD) was created to describe this condition. ISD is a weakness or defect in the intrinsic urethral sphincter, resulting in stress incontinence, and is generally considered a more severe type of stress incontinence which emphasizes decreased urethral sphincter function rather than urethra position or mobility. Whether there is an actual physiologic difference in these “types” of stress incontinence or their underlying pathophysiology is unknown. This distinction may be artificial, looking at the same problem of neuromuscular dysfunction along one continuous spectrum. Traditionally, from a surgical standpoint, the distinction between the two types of stress incontinence has been used to determine the type of surgical procedure performed—colposuspension or pubovaginal sling. Hypermobile stress incontinence has generally been managed with the “urethral hammock” support using a colposuspension such as a Marshall-Marchetti-Krantz (MMK), Burch, or paravaginal repair. The surgical procedure's goal is to elevate and stabilize the UVJ. Patients with ISD have been found to have lower success rates following retropubic urethropexies when compared to patients with urethral hypermobility and a better functioning urethra. Thus pubovaginal sling procedures employed to treat ISD are designed to restore compression and coaptation of the damaged urethral sphincteric mechanism. While some surgeons continue to surgically treat these “types” of stress incontinence differently, many surgeons, both urologists and gynecologists, are increasingly using a pubovaginal sling for all patients with stress urinary incontinence. Because of the reported durability, the pubovaginal sling, in increasing numbers of practices, is being performed for patients with hypermobile stress incontinence without ISD. Cross and associates performed pubovaginal slings on such patients with a cure rate of 93% reported at 22 months. In 1996, a review by Black found four prospective studies done to compare colposuspension with sling procedures. None reported a difference in cure rate although all had substantially less than 50% power to detect a significant difference and all defined surgical success differently. Only one study analyzed complication rates and found a statistically significant higher risk of complications after a sling operation including elevated postvoid residual urine, entry into the bladder, and uterine prolapse. One large retrospective trial suggested that colposuspension may be more effective than sling procedures (95% vs. 79%). Marinkovic and colleagues, in a retrospective review of 18 women having a Burch colposuspension and 18 women undergoing a pubovaginal sling, found that the Burch group had twice the number of postoperative complications, with comparable success. In contrast, Enzelsberger showed significant increases in postoperative voiding dysfunction in patients after pubovaginal sling (13%) that was not observed in Burch patients. Thus, there is no convincing evidence that the effectiveness of colposuspension and sling procedures differ. The differences may be in the complication rates for the two procedures. However, less than 150 patients have ever been included in prospective studies concerning these two surgeries. Two large, prospective trials comparing Burch colposuspension and pubovaginal sling are underway and should help determine the best procedure. Suburethral Plication/Anterior Colporrhaphy While most surgeons have abandoned the suburethral plication as an antiincontinence procedure, for a select group of patients, the Kelly-Kennedy plication is appropriate. Dr. Howard Kelly from Johns Hopkins University was the first to describe this operation in 1913. The technique has changed very little from his original description. Through an anterior vaginal incision, the vesicovaginal space is dissected bilaterally to the pubic rami. A series of plicating mattress sutures of either permanent or delayed-absorbable material are then placed beneath the UVJ and then subsequently, in any significant pubocervical fascial defect, below the urethra and bladder. Although Dr. Kelly reported a more than 90% subjective success rate with his anterior vaginal repair, subsequent success rates in the literature have ranged from 31% to 69% with 1- to 5-year follow-up. One noteworthy exception to these figures, however, is two studies from Dr. R. Peter Beck of Canada. In his original series of 105 patients reported on in 1982, a 90% subjective cure rate was described. In a follow-up paper published in 1991, similar outcome numbers were provided for over 500 patients. Dr. Beck attributed much of his success to the use of delayed absorbable polyglycolic suture instead of chromic cat gut suture. Anterior vaginal repair remains indicated in the treatment of a cystocele caused by a central fascial defect. Concomitant suburethral plication may be a reasonable choice in the prevention of potential or occult stress incontinence for patients having vaginal reconstruction for pelvic organ prolapse. The most appropriate indication for a suburethral plication is in conjunction with an obliterative procedure, such as when complete or LaForte colpocleisis is performed. These patients, generally older with poorly functioning detrusor muscles or detrusor overactivity, are at risk for urinary retention or worsening of incontinence from detrusor contractions if a more durable antiincontinence surgery is performed. A suburethral plication may be the best antiincontinence surgery for these patients. Their incontinence may not be “cured”, but the low morbidity balances the higher risks of urinary retention or overactive bladder. Success in these cases is defined by improved quality of life from prolapse surgery and hopefully a reduction in urinary incontinence and avoidance of self-catheterization. Complications include hemorrhage, vaginal foreshortening, and suture misplacement in the bladder, urethra, or ureter. Thus, follow-up cystoscopic evaluation is recommended to assess for potential injury to the lower urinary tract. Retropubic Urethropexy Retropubic urethropexy (RPU) is used by many surgeons as their primary repair for stress incontinence. The first RPU, described in 1949, the Marshall-Marchetti-Krantz (MMK) procedure has had success rates ranging from 70% to 90%. After entry to the retropubic space is achieved, the urethra is identified along its course to the bladder neck. One to three pairs of sutures, usually nonabsorbable, are placed on each side and close to the urethra through the pubocervical fascia. These are then fixed to the fibrocartilage of the symphysis pubis. Because the angle of elevation is more acute and centrally placed than with a Burch procedure, the potential exists for the patient to have an increased risk of bladder neck obstruction. Voiding dysfunction occurs in 5% to 20% of cases. The rate of subsequent osteitis pubis is approximately 2%. In 1961, Dr. John Burch of Nashville, Tennessee described a new fixation point for the RPU sutures, namely Cooper's ligament. Through a Pfannenstiel incision, the retropubic space is entered. With the surgeon's nondominant hand in the patient's vagina, the Foley catheter is pulled down to the UVJ. While gently retracting the bladder medially, the paravaginal tissues are identified and, if necessary, gently cleaned off at the UVJ and midurethra extending approximately 2 cm lateral to the urethra. Seeing the white surface of the vaginal submucosal layer in contrast to the yellow bladder helps identify

the vagina. A full thickness figure-of-eight of permanent suture is placed at the UVJ and midurethra on both sides. Cooper's ligament is identified, and both arms of the sutures are brought through the ligament, with the midurethral sutures most caudad ( Fig. 48.6). The sutures are tied while elevating the vagina so that the urethra is repositioned parallel to the floor. Overcorrection should be avoided; suture bridges are the norm. Cystoscopy, transurethrally or through the bladder dome, teloscopy, is then performed to assess the integrity of the lower urinary tract. Consideration for placement of a suprapubic catheter is reasonable, especially if the patient is predisposed to postoperative voiding dysfunction.

FIG. 48.6. Burch retropubic urethropexy. Sutures are placed lateral to the urethrovesical junction and lateral to the midurethra and brought to the Cooper ligament bilaterally. (Adapted from Tanagho EA. Colpocystourethropexy: the way we do it. J Urol 1976;116:751.) Objective cure rates for the Burch RPU range from 70% to 90%. Complications include ureteral damage (usually from overcorrection or kinking, not direct trauma) and bleeding from the adjacent large veins of Santorini. Enterocele formation following colposuspension has been reported in 3% to 17% of patients, leading most surgeons to perform a uterosacral ligament plication or cul-de-sac obliteration for patients predisposed to pelvic organ prolapse. The prevalence of postoperative detrusor overactivity is 5% to 18%. Postoperative management is similar to that for other abdominal surgical procedures. It is usually possible to allow patients to have oral intake soon after the operation, because the peritoneal cavity was not entered. The bladder is drained continuously for the first 1 or 2 postoperative days. If used, the suprapubic catheter is occluded and patients are asked to indicate when they feel that the bladder is full. They are then allowed to void, and the amount of urine remaining inside the bladder is measured by the nursing personnel. If transurethral catheter drainage has been used, the bladder can be filled retrograde until the patient feels full. The catheter is then removed and the patient is asked to void. The voided amount is measured using a hat placed in the toilet. By subtracting the voided amount from the amount instilled in the bladder, the postvoid residual can be determined. The catheter should be replaced or the suprapubic catheter should be left in until the patient is able to void adequately with small amounts of residual urine (10%) of having a mutation for BRCA1 ( Table 50.2).

TABLE 50.2. American Society of Clinical Oncology (ASCO) guidelines for recommending genetic testing for families with high probablility (>10%) of having BRCA1 mutation Identification of patients with a high-risk family history should be referred for genetic counseling and testing. The decision to undergo genetic testing is a complex one, as it can affect an individual's personal, psychological, social, financial, and ethical well-being. Women who have a negative genetic test should still be considered at risk on the basis of age, environment, or other genetic factors or unknown mutations.

HISTORY AND PHYSICAL EXAMINATION Obtaining a thorough history, including a family history and information on menstrual status, pregnancies and lactation, hormone use, prior breast surgeries and trauma, is essential. In addition, ascertaining whether the patient performs breast self-examinations, as well as the presence and characterization of nipple discharge or a breast mass is important. Bilateral breast examination is best performed right after menstruation and prior to ovulation. At this time, breast engorgement and tenderness is less likely to be present. A multipositional breast examination should be performed, including examination in the upright and supine positions ( Fig. 50.1 and Fig. 50.2). Breast retraction and subtle changes in the skin and nipple may be missed if the patient is examined in only one position. Examination should be performed with hands at sides, elevated above the head, and finally, with the arms tensed at the waist (contracting the pectoralis muscles). Attention is directed toward the supraclavicular area and axilla. Digital palpation is performed beneath the lateral pectoralis muscles into the axilla itself.

FIG. 50.1. A: Inspection of patient with arms at sides. B: Inspection of patient with both arms raised. C: Inspection of patient with hands at waist, pectoral muscles contracted.

FIG. 50.2. A: Palpation with patient upright, with support of ipsilateral elbow; axillary nodes and also supraclavicular nodes examined. B: Palpation of breast with patient in supine position.

The second phase of the breast examination is conducted with the patient in the supine position. Digital palpation is carried out using the index and middle fingers and applying varying amounts of pressure with the flats or pads of the fingers. A thorough examination systemically covers the entire breast and chest wall. The examination can be done in a clockwise direction or by rows (stripwise). It is important to carefully examine beneath the nipple–areolar complex and within the axilla. An inflammatory appearance of the breast should raise suspicion of an inflammatory carcinoma. The classic appearance of inflammatory breast cancer includes a red, swollen breast with skin edema (“peau d'orange”). The breast is generally not tender. If the inflammation persists following a short course of antibiotics to rule out cellulitis, biopsy of the breast and skin is warranted. Inflammatory breast cancer is often a clinical diagnosis, and a benign skin biopsy should not dissuade the clinician from undertaking further evaluation and treatment. Any asymmetric skin changes or changes of the nipple–areolar complex should arouse suspicion. Paget disease of the nipple is the presence of intraductal or invasive cancer involving the nipple and should be excluded by a nipple biopsy of the abnormal area following a mammogram. It is important to instruct patients in the technique of breast self-examination. Physician-directed discussion on breast self-examination is the most effective approach. Physicians have the opportunity to reinforce what is normal versus abnormal to patients during the examination. If no abnormal findings are noted on examination, it is critical to document negative findings. The date of the last mammogram, discussion of cancer screening, and plans for follow-up should also be recorded. Hormones (HRT or OCP) should not be renewed without a documented annual breast examination or mammography if indicated. A great deal of litigation results from failure to diagnose breast cancer. The Physician Insurers Association of America's breast cancer claims study, conducted in 1988, determined that 75% of successful malpractice lawsuits involved primary care physicians with practices in family medicine, internal medicine, or obstetrics and gynecology. It is important that the medical

chart include careful documentation, since approximately one third of the cases reported in the Physician Insurers Association of America's study resulted from inadequate documentation.

MAMMOGRAPHY The primary goal of mammography is to screen asymptomatic women to help in detection of breast cancer at an early stage. In general, a routine screening mammogram consists of a mediolateral oblique (MLO) view and a craniocaudal (CC) view of each breast. With modern low-dose screening, the dose is less than 0.1 rad per study (for comparison, a chest x-ray delivers 0.025 rad per study). The effectiveness of screening also varies depending on the density of the breast. Breast composition may be one of four patterns of increasing density: 1. 2. 3. 4.

almost entirely fat scattered fibroglandular densities heterogeneously dense extremely dense.

The greater the breast density, the lower the sensitivity of the mammogram. Because some palpable cancers are invisible on mammography, a negative study cannot always exclude cancer. It is important to note that the false-negative rate for mammograms is 10% to 15% and that a normal mammogram does not eliminate the need for further evaluation of a dominant mass in the breast. If the clinical examination is suspicious, a negative mammogram should not delay further investigation. Mammographic screening in women 40 years or older has reduced mortality by 20% to 30%. The efficacy of screening mammography in decreasing breast cancer mortality has been demonstrated in numerous studies. In the 1960s, the Health Insurance Plan of Greater New York performed a study of physical examination and mammography in a study group of 30,756 women and a control group of 30,239 women between the ages of 40 to 64 years. At 10-year follow-up, the study group had a 30% decrease in breast cancer mortality compared with the control group. A total of eight large randomized trials on mammographic screening have been conducted. Six of the eight trials revealed a statistically significant reduction in mortality with mammographic screening. The reduction in mortality was not as evident among women between the ages of 40 and 49 compared with women over 50 years of age. The relative mortality reduction appears later in women between the ages of 40 and 49 at randomization compared with women 50 years of or older. It is also likely that the small numbers of women between 40 and 49 years of age in the existing randomized trials may have contributed to this difference. In a meta-analysis of eight randomized, controlled trials of mammographic screening, a statistically significant 18% reduction in mortality in women aged 40 to 49 was noted. Combined data from five Swedish trials yielded a statistically significant mortality decrease of 29% ( Table 50.3).

TABLE 50.3. Randomized population-based mammography trials

A re-analysis of the meta-analysis excluded six of the eight studies because of issues related to randomization methods used and other factors in these trials. This re-analysis questioned the risk reduction offered by mammography and resulted in much controversy. The risk reduction associated with mammography continues to be an area of debate within the medical community. Screening Interval For several years, there has been a significant debate about the appropriate age at which to commence mammographic screening. In 1997, the American Cancer Society (ACS) and the National Cancer Institute (NCI) modified the guidelines for mammographic screening for women between the ages of 40 and 49, recommending regular mammograms for women in this age group. The recommended intervals differ: the ACS recommends a yearly mammogram starting at age 40, while the NCI recommends a mammogram every 1 or 2 years. The ACOG recommendations on mammography are similar to the NCI guidelines. Annual screening mammography may commence earlier than age 40 in a few special circumstances ( Table 50.4).

TABLE 50.4. Screening guidelines for women under age 40

BI-RADS In the past, a lack of uniformity in mammography terminology and reporting often led to confusion as to the malignant nature of a lesion. In 1994, the Mammography Quality Standards Act was passed by Congress and is administered by the Food and Drug Administration (FDA). It requires that mammography facilities monitor the results of their breast cancer–detection programs, including the number of recommended biopsies, and the size, number, and stage of cancers detected. The American College of Radiology (ACR) Breast Imaging Reporting and Data System uses a terminology and lexicon system called BI-RADS for reporting abnormalities seen on mammography ( Table 50.5). This standardized reporting system—the Breast Imaging Reporting and Data System—was developed in 1995. Each category leads to a fixed assessment and specific management recommendations.

TABLE 50.5. American College of Radiology (ACR) BI-RADS assessment categories

In addition, associated findings such as skin or nipple retraction, skin thickening, skin lesions, axillary adenopathy, and the presence of architectural distortion should also be reported. The predictors of malignancy for the BI-RADS categories are 0% to 2% for category 3 and approximately 98% or greater for category 5.

Category 4 is less predictable. Liberman and colleagues and Orel and colleagues have placed the risk of malignancy for this category around 30%. The ACR is working on a new edition of the BI-RADS classification system that, in particular, will attempt to provide data on category 4 in terms of risk of malignancy. According to the ACR, in the forthcoming edition of BI-RADS, category 4 will be divided into three subdivisions—low, medium, and high—in an effort to better guide clinicians and to get meaningful data about this category (personal communication, July 2002). Diagnostic Mammography Abnormalities found on mammographic screening may need further evaluation with additional mammography views or other imaging modalities such as ultrasound or magnetic resonance imaging (MRI). In some screening programs, the mammograms are reviewed by the radiologist as they are performed, and if additional views are needed, they are performed on the same day. In other programs, if additional studies are required, the patient is called back for them at a later date. In several studies, the frequency of “call-backs” has ranged from 5% to 11%. Mammographic Lesions A “mass” is defined as a space-occupying lesion seen in two different projections. If a possible mass is seen on only one view, it is called a “density” until its three-dimensionality is confirmed. A description of the shape and the margins of the lesion are also necessary. The highest frequency of carcinoma is noted in masses that have an irregular shape or spiculated borders. These lesions are associated with pleomorphic calcifications that appear discontinuous and linear in distribution. This discontinuous linear pattern suggests irregular filling of a duct with abnormal cells. Microcalcifications The BI-RADS lexicon describes calcification morphology (shape) and distribution. Calcifications may be scattered or clustered, coarse or fine, old or new. Comparison with prior mammograms is often necessary ( Table 50.6).

TABLE 50.6. Morphology of microcalcifications and associated lesions

Breast Ultrasound and MRI Breast ultrasonography can be used to distinguish between solid and cystic masses in the breast. It can be used to evaluate a focal mass identified on a mammogram or a palpable mass. It is also used as an adjuvant for biopsy. Because of its low specificity, it is not thought to be a good modality for screening. It cannot replace mammography, as it has no ability to detect microcalcifications. Ultrasound can complement mammography in young women with dense breasts (which limits the accuracy of the mammogram). Presently, MRI has no role in breast cancer screening. MRI has a high sensitivity in the diagnosis of breast cancer, ranging from 86% to 100%, but a low specificity 37% to 97%. Because of this low specificity it is of limited value in screening. It is an expensive test that requires intravenous contrast and the technology for performing biopsy under MRI guidance is not widely available. Current uses include evaluation of breast implants for rupture, evaluation of pectoralis involvement with extensive breast cancer, and evaluation of post-lumpectomy bed fibrosis. Research into the use of MRI for screening of patients with dense breasts is under investigation. Future uses may include evaluation of occult breast cancers and evaluation of multifocal disease in those patients who are considering breast conservation.

DIAGNOSTIC EVALUATION Palpable Mass The workup of a patient with a dominant mass should include a bilateral mammogram. In addition to gaining valuable information about the characteristics of the mass, a secondary purpose in this setting is to screen the normal surrounding breast and the contralateral breast for nonpalpable mammographic abnormalities (densities or calcifications). Evaluation of a palpable mass is important to determine whether the mass is cancerous even if the mammogram is negative. Fine-Needle Aspiration or Biopsy Fine-needle aspiration (FNA) can be extremely useful in providing a cytologic analysis of a palpable breast mass. Many palpable thickenings and all dominant masses should be considered for FNA as it can differentiate between solid and cystic masses. In addition, FNA can diagnose and treat simple cysts and provide cellular material for cytologic analysis. The FNA should be performed after radiologic examination because the resultant hematoma could mask an underlying abnormality. The breast is prepped with alcohol; with the physician facing the patient, the lesion is stabilized with the physician's opposite hand. Usually, a 21-gauge or 25-gauge needle on a 10-cc syringe is used. Approximately 3 cc of air is aspirated into the syringe to facilitate expulsion of the contents onto the slide following the procedure. The needle is introduced into the lesion, and suction is applied on the syringe ( Fig. 50.3A, Fig. 50.3B and Fig. 50.3C). If the mass is cystic, the fluid is completely evacuated and the lesion should completely disappear. The syringe is withdrawn and the fluid is discarded if it is serous and nonbloody. The patient should return in 4 to 6 weeks for reexamination.

FIG. 50.3. Aspiration biopsy. A: Mass. B: Stabilizing the lesion. C: Aspirating.

If the lesion encountered is not cystic or suspected to be solid, an FNA biopsy can be performed in the same manner. After insertion into the lesion, multiple passes (10–15) through the lesion with changes in direction allow extensive sampling and create a “feel” for the mass (carcinomas are usually hard and gritty). The goal of sampling is to obtain material in the hub of the needle, not to fill the syringe. Care should be taken to release the suction before withdrawing the needle to prevent aspiration into the syringe. The sample is then ejected onto a glass slide, gently smeared with another slide and placed in sterile jars containing 95% ethanol for transport to the cytology lab. Alternatively, it can be placed in a specimen jar containing cytofixative. The needle should be removed from the syringe, the medium

aspirated into the syringe, the needle replaced, and the medium then ejected into the jar. An FNA requires a cytopathologist experienced in breast pathology. The false-negative rate can range from 3% to 35% depending on the expertise of the aspirator and the cytopathologist, the size of the lesion, the location within the breast and the cellular composition of the lesion. Negative findings of an FNA in the presence of a suspicious mass should not preclude further diagnostic evaluation. A diagnosis of atypical cells following an FNA warrants a surgical biopsy. Any mass remaining after aspiration of a cyst should be excised. Similarly, a cyst that recurs in the same location after one or two aspirations should be excised. The false-positive rate of an FNA is less than 1%, but in the United States most surgeons will not perform definitive surgery (i.e., a mastectomy or axillary dissection) without a prior surgical biopsy, core-needle biopsy, or frozen-section diagnosis at the time of surgery. An FNA that is positive for adenocarcinoma could, however, provide a preliminary diagnosis and guide subsequent management. Patients with palpable solid masses can have a biopsy of the mass in the office with use of a Tru-cut 14-gauge biopsy device. The breast is prepped sterilely and a local anesthetic is used to infiltrate the skin. A small nick is made in the skin with a scalpel to accommodate the biopsy instrument. A core biopsy of the solid mass is obtained. The instrument has a “firing” range and therefore should be kept parallel to the chest wall to avoid penetrating trauma. The specimen is placed in formalin and sent to pathology. It is believed that if the specimen “floats” in the solution, it is likely nondiagnostic fat. Tumor specimens will have a grayish appearance and will typically “sink” in the solution. Needle-Localization and Excision Needle localization is a technique that allows surgical excision of a lesion that is nonpalpable. The technique uses a hook-wire system to target the lesion, and image guidance can be provided by mammogram, ultrasound, and in some cases MRI. In mammography-guided needle localization, coordinates of the lesion are obtained by placing the breast in an alphanumeric grid. The needle is inserted and, when adequate placement is noted, the hook wire is deployed and the needle removed. Two mammographic views are then obtained. The mammography films are available intraoperatively and show the relationship between the lesion and localizing hook. Excision with needle localization allows the surgeon to minimize the amount of breast tissue removed by following the needle to the targeted lesion. After removal, a specimen radiograph is obtained to ensure that successful removal of the lesion has been performed. Radiologists and surgeons experienced in needle localization and excisions report only 0.2% to 0.3% of lesions missed with this approach. The specimen radiograph helps to ascertain the lesion was not missed. Image-Guided Percutaneous Breast Biopsy With the current advancements available in breast imaging, percutaneous image-guided breast biopsy is increasingly being used as an alternative to surgical biopsy. Percutaneous biopsy methods differ with respect to the method of imaging guidance and the tissue-acquisition device used. The use of image-guided percutaneous biopsy has advantages over surgical excision for the diagnosis of breast lesions. It is less invasive and because less tissue is removed, it will result in less scarring on subsequent mammograms. Regardless of whether the diagnosis is benign or malignant, the patients who have percutaneous biopsies will undergo fewer operations. In addition, in cases of malignancy, the discussion and surgical treatment plan can be streamlined. The choice of which image-guided modality to use depends on the lesion. Stereotactic biopsy is best for calcifications. If a lesion is seen on ultrasound, it is best to use that modality, since it is easier to use and has been reported to be less costly. Stereotactic Biopsy Stereotactic biopsy uses specialized mammography equipment to calculate the location of a lesion in three dimensions. Stereotactic biopsy can be performed with the patient prone on a dedicated table or with the patient sitting in an upright unit. An automated core needle or directional vacuum-assisted biopsy probe is used to obtain the tissue specimens. Multiple tissue specimens are obtained for pathologic analysis. Many reports in the medical literature state the procedure has a sensitivity of 70% to 100% and a specificity of 85% to 100%. The greatest success is noted in reports using 14-gauge core needles, as well as in those with increased numbers of specimens obtained. In mass lesions, it is likely that five core samples may be adequate for accurate diagnosis; however, ten or more core specimens may be required in cases of calcifications. Ultrasound-Guided Biopsy The use of ultrasound imaging for percutaneous biopsy of lesions seen on ultrasound has certain advantages. For example, it requires no specialized equipment, no radiation exposure, and has the ability to sample areas that may be inaccessible with stereotactic biopsy (such as the axilla). A 14-gauge automated needle is used, and real-time imaging allows accurate positioning. Multiple tissue core samples are sent for pathologic analysis. Tissue-Acquisition Devices Available tissue-acquisition devices include fine needles, automated core needles, directional vacuum-assisted probes, and biopsy cannulas. Excellent results have been obtained using the 14-gauge automated needle for biopsy of masses under ultrasound or stereotactic guidance. Most centers use larger tissue-acquisition devices instead of fine needles, because of accuracy of tissue diagnosis when a larger volume of tissue is obtained. Compared with the automated needle, the vacuum device acquires larger samples of tissue, has a higher frequency of retrieval of calcifications, and may provide more accurate lesion characterization. Accurate placement of a localizing clip through the biopsy probe is necessary to facilitate subsequent localization if needed. Surgical Excision/Breast Biopsy The ACOG has stopped short of recommending that open biopsy be performed by every obstetrician and gynecologist. A biopsy can be performed on an outpatient basis under local anesthesia in the majority of patients. It is important to choose the appropriate incision and location ( Fig. 50.4). Unless the lesion is close to the nipple or suspected to be a fibroadenoma, the incision should be made in close proximity to the mass and not circumareolar. The surgeon should keep in mind the possibility of subsequent mastectomy when placing the incision. Many times, the biopsy is part of the treatment. The specimen should be adequately oriented for margin analysis by the pathologist and also sent for the appropriate markers such as estrogen-receptor (ER) and progesterone-receptor (PR) status and HER2/ neu. Orientation of the specimen is important, as a reexcision of a close or involved margin may need to be performed.

FIG. 50.4. Optimal orientation of biopsy incisions.

The incision should be closed with fine suture material, with a subcuticular closure. Hemostasis needs to be ascertained prior to closing and is usually achieved with electrocautery. Weck clips can be placed in the cavity bed if a diagnosis of breast cancer is known and breast conservation is planned. No particular immobilization is required but a good support bra is recommended to minimize hematoma and induration.

BENIGN BREAST CONDITIONS

Fibrocystic Changes Fibrocystic change is the most common benign breast condition in women. It is a result of fluctuating hormone levels, and most common in premenopausal women between the ages of 20 and 50. It is often associated with pain and tenderness (mastodynia) and tends to be bilateral. Most women will report symptoms during the premenstrual phase of the cycle. Pain is due to breast stromal edema, ductal dilation, and associated inflammatory response. An increase in breast size is also frequently reported. The differential diagnosis for breast pain includes other conditions affecting the anterior chest wall such as intercostal neuralgia, myalgia, and chronic costochondritis. Women with large pendulous breasts will have associated stretching of Cooper ligaments and associated breast pain. Etiologic factors are still inconclusive. The ingestion of foods and medications containing methylxanthines has been implicated through an inhibition of 3'5'-cyclic adenosine monophosphate (cAMP) phosphodiesterase and 3'5'-cyclic guanosine monophosphate (cGMP) phosphodiesterase. This inhibition will lead to accumulation of increased amounts of cAMP and cGMP. High levels of cAMP and cGMP have been detected in patients with fibrocystic change. In some studies, reduction of dietary methylxanthines has been associated with symptomatic subjective reduction in pain, tenderness, and palpable nodularity. Other studies, however, have failed to show an effect from decreased consumption of dietary methylxanthines. Fibrocystic change is not a risk factor for cancer in the majority of women. Histologically, there are two changes noted with fibrocystic change: nonproliferative changes and proliferative changes. The nonproliferative changes include cystic changes with formation of microcysts (2 mm or less in size), macrocysts, and fibrosis. Proliferative Changes Proliferative changes include hyperplasia and adenosis. Hyperplasia is proliferation of ductal epithelium, which results in layering of the cells. Atypia may be associated with this proliferation. If atypia is noted, this confers a five fold increase in breast cancer risk for the patient. Hyperplasia with atypia is the only fibrocystic change associated with an increased risk factor for breast cancer. If atypia is noted on a core biopsy, a surgical excision of the area is recommended, as it is thought that there is a 50% chance of finding a coexistent carcinoma. Adenosis is also a proliferative lesion, caused by changes in the acini in the distal mammary lobule. Sclerosing adenosis refers to the dense, fibrotic tissue surrounding these small ducts. These lesions may present as a palpable mass in women in their 30s and 40s. A papilloma can result from this ductal proliferation. Papillomas are papillary lesions with a branching fibrovascular core surrounded by epithelium. These lesions are associated with serosanguineous nipple discharge in 25% to 50% of presentations. Ninety percent of the time, there is a small palpable mass adjacent to the areola. Intraductal papillomas are rarely associated with carcinoma, but require surgical excision to rule out the possibility of misdiagnosis of a malignancy. Management of fibrocystic changes includes regular physical examinations, appropriate imaging, and supportive measures. Recommendations for use of a good support bra may be helpful, especially in physically active women. Dietary restrictions of methylxanthines may produce subjective improvement in 65% of patients. The use of vitamins A and E has been reported in some studies to be helpful. Diuretic therapy during the premenstrual period has been reported to provide temporary relief, and requires cyclical use. Fluid retention is a result of cyclical hormonal stimulation. Oral contraceptives suppress symptoms of fibrocystic changes in the majority of patients (70%–90%). Symptoms often recur after discontinuation. Other medications such as danazol (17a-norethisterone) in doses of 100 to 400 mg per day should be reserved for patients in whom other agents have been ineffective. Their side-effect profile can lead to poor compliance. A 3- to 6-month course can provide significant reduction in symptoms, and its effect can last several months after its discontinuation. Fibroadenoma Fibroadenomas are benign fibroepithelial tumors and are the second most common benign lesion of the breast. They are the most common lesion found in women under the age of 25. They will persist during the menstrual years of a woman's life, but regression after menopause has been reported. Patients typically present with a mobile, smooth, painless, palpable mass. Ultrasound examination, along with physical examination, can help in making the diagnosis. Mammographically, fibroadenomas may appear as round, oval, or lobulated masses with circumscribed margins. In older women, they can have a rim of coarse calcifications. FNA will reveal benign ductal epithelial cells and elongated dense stromal cells. Microscopically fibrous tissue composes most of the fibroadenoma. Carcinoma arising in fibroadenomas is rare. Fibroadenomas can be followed without the need for complete surgical excision. This can be achieved with physical examination or ultrasound examination if they are not palpable. However, surgical excision should be performed if: the mass continues to enlarge the results of FNA or core biopsy are inconclusive or yield atypia the patient desires surgical excision. Phyllodes Tumor Phyllodes tumors are uncommon, slow-growing fibroepithelial tumors. Previously referred to as cystosarcoma phyllodes, this name contributed to confusion in understanding this entity. Although very similar to a fibroadenoma, the stromal component is hypercellular with increased pleomorphism and mitotic activity. Phyllodes tumors can occur in women of any age, but more commonly occur in premenopausal women. Malignant behavior in phyllodes tumors is rare in premenopausal women. Malignant phyllodes tumors are noted when there is a combination of increased mitotic activity, invasive borders, or marked pleomorphism. Incomplete excision is a major determinant for local recurrence. Treatment is total surgical excision with a wide margin of healthy tissue. Superficial Thrombophlebitis Superficial thrombophlebitis is also known as Mondor disease of the breast. It is an uncommon benign inflammatory process. It can occur spontaneously, but usually is associated with breast trauma, breast surgery, or pregnancy. It is a thrombophlebitis of the thoracoepigastric vein, which drains the upper-outer quadrant of the breast. Patients present with acute pain and a linear, tender fibrotic band with skin retraction over the distribution of the thoracoepigastric vein. Treatment is conservative, with analgesics and application of heat. The condition resolves in 1 to 3 weeks. Skin retraction superficial to the area of inflammation can remain if the inflammation is extensive. Biopsy is not necessary. Mastitis Mastitis usually occurs in relation to lactation. It can occur in nonpuerperal periods in association with galac-torrhea. Skin organisms, Staphylococcus aureus, and Streptococcus species may cause infection of the nipple and breast ducts. Presence of milk in the ducts is an excellent medium for infection. Women with mastitis may continue to breast-feed. Antibiotic therapy with dicloxacillin sodium (250 mg q.i.d.) or penicillin G is indicated. If there is no response, an abscess that may require surgical drainage must be excluded. Inflammatory carcinomas can mimic mastitis, and if no resolution of infection is noted despite continued antibiotics, a skin biopsy may be indicated. Galactoceles are milk-filled cysts. They are usually tender and present after the abrupt termination of breast-feeding. Aspiration of the cyst is often necessary for symptomatic relief. If reaccumulation occurs, however, surgical excision may be required to avoid infection. Duct Ectasia Duct ectasia is a condition usually occurring in perimenopausal or postmenopausal women. Patients present with a tender, hard erythematous mass adjacent to the

areola in association with burning, itching, or a sensation of pulling in the nipple area. A thick greenish-black discharge may be present. Histologic evaluation of the area shows dilate, distended terminal collecting ducts obstructed with inspissated lipid-containing epithelial cells and phagocytic histiocytes. This process tends to occur in a segmental fashion extending from the involved nipple area to adjacent ducts. Occasionally, as a result of this infection, a small abscess forms at the base of the nipple. Treatment is excisional biopsy. Younger women can present with inflammation of the ducts in the region of the nipple, which may produce fissures and fistulae with connection from the nipple ducts to the skin at the edges of the areola. Prior periductal mastitis leads to the squamous epithelium of the terminal dilated portion of the collecting ducts to undergo squamous metaplasia. Keratin is formed in the duct, accumulates, and can cause an abscess at the base of the nipple. Excision of the area usually is necessary. Fat Necrosis Fat necrosis is a relatively uncommon benign condition occurring as a response to breast trauma. Patients present with a hard mass that can mimic a carcinoma. The irregular mass is palpable and may involve skin retraction. Multiple calcifications can be seen on mammography. The histology is active chronic inflammatory cells, with lymphocytes and histiocytes predominating. In the later stages, a collagenous scar is noted, with “oil cysts” or free lipid material released by lipocyte necrosis. Fat necrosis does not increase the risk of carcinoma, and its clinical importance is in the differential diagnosis of a carcinoma. Nipple Discharge Nipple discharge has been reported in 10% to 15% of women with benign breast disease and in 2.5% to 3.0% of those with carcinoma. The discharge is classified according to its appearance as milky, green, bloody, serous, cloudy, or purulent. The drainage should be classified according to whether it is unilateral, bilateral, or spontaneous, or recurrent. This information is obtained at the time of a thorough history and physical examination. For example, if the drainage first appeared in the patient's bra or nightgown on awakening, this finding is significant. The presence of a mass should also be investigated. The risk of cancer is increased when the discharge is unilateral from a single duct, occurs in a postmenopausal patient, or when a mass is present. Unilateral, Spontaneous Nipple Discharge In cases of unilateral, spontaneous nipple discharge, several causes are included in the differential. The most common cause of nipple discharge is mammary-duct ectasia, which produces a multicolored (green, yellow, white, brown, gray, or reddish brown) nipple discharge. The reddish-brown discharge is often mistaken for a blood discharge. It is thought to be due to an increase in glandular secretions, with the production of an irritating lipid fluid that can produce a nipple discharge. Guaiac of the discharge can help to diagnose whether it is bloody. The next most common cause of a multicolored, sticky nipple discharge is nonpuerperal mastitis. The persistent type involves inflammation in deeper portions of the breast; the transient types are associated with periareolar inflammation. If the inflammation develops into an inflammatory mass, surgical excision and drainage are necessary. Medical management with local care, avoidance of all nipple manipulation, and nonsteroidal antiinflammatory agents and an antistaphylococcal antibiotic is often successful when infection is suspected. Bloody nipple discharge warrants surgical evaluation. Intraductal papillomas are the most common cause of bloody nipple discharge. During the breast examination, physicians should look for an associated periareolar mass. The examination consists of gently and carefully palpating the subareolar region to identify the pressure point that produces the discharge. It is important to reproduce the discharge and demonstrate the breast quadrant from which it emanates. All significant nipple discharges warrant referral for tissue biopsy. Although a mass is usually present when the discharge is due to cancer, there is no palpable mass in 13% of cancers with nipple secretions. Bloody discharge occurring in the third trimester of pregnancy however may be regarded as physiologic and does not require intervention unless persistent for several months after delivery. There are no contraindications to breast-feeding in these patients. In addition, physicians should not rely solely on the cytology of the discharge because there is an 18% false-negative rate and a 2.6% false-positive rate with standard cytology alone. Galactography (injecting radiopaque contrast into the discharging duct and then performing mammography) offers better visualization of small intraductal papillomas but cannot differentiate between benign and malignant lesions. A surgical procedure is still necessary. Mammography has a 9.5% false-negative rate and a 1.6% false-positive rate for detecting cancer in patients with a nipple discharge.

BREAST CANCER Natural History The most common site of origin of breast cancer is the upper-outer quadrant (38.5%), central area (29%), upper-inner quadrant (14.2%), lower-outer quadrant (8.8%), and the lower-inner quadrant (5%). These percentages correlate with the amount of tissue that is present in these quadrants. Metachronous bilateral carcinoma of the breast has been observed in 5% to 8% of patients. Metastasis to the ipsilateral axilla is the most common route of spread. Metastasis to the internal mammary nodes is more frequent with inner-quadrant lesions and is more likely to occur when involvement of the axillary nodes is also present. Pathology Ductal Carcinoma in Situ Ductal carcinoma in situ (DCIS) is an abnormal proliferation of malignant epithelial cells within the mammary ductal–lobular system without invasion into the surrounding stroma. It is classified as a heterogenous group of lesions with different growth patterns and cytologic features. Classification of DCIS has traditionally been based on architectural pattern. The most common types are comedo, cribriform, micropapillary, papillary, and solid. Paget's Disease Paget's disease is involvement of the nipple with intraductal carcinoma. In absence of a palpable mass, invasive carcinoma occurs in less than 40% of cases. The malignant cells are large and pale-staining and are seen in the basal layer and upper portions of the epidermis. Diagnosis is made through a nipple biopsy. Lobular Carcinoma in Situ Foote and Stewart initially described lobular carcinoma in situ (LCIS) in 1941 as a noninvasive lesion arising from the lobules and terminal ducts of the breast. LCIS is characterized by a solid proliferation of small cells with round to oval nuclei that distort the involved spaces in the terminal duct–lobular units. Three important features of LCIS are: 1. It is usually an incidental microscopic finding that is not detected clinically or by gross pathologic examination. 2. It is multicentric, and the associated cancer may be ductal or lobular. 3. The risk for subsequent cancer is the same for both breasts. It is unfortunate that Foote and Stewart chose the name they did, as it has led to a great deal of confusion over the past several decades. LCIS is a marker for breast cancer risk and is not a malignant finding. Invasive Duct Carcinoma Invasive duct carcinoma is the most common group of malignant mammary tumors and comprises 65% to 80% of all mammary carcinomas. Included in this group are special subtypes: Tubular, medullary, metaplastic, mucinous (colloid) papillary, and adenoid cystic carcinoma. Each subtype constitutes only 1% to 2% of all invasive breast cancers, except medullary carcinoma, which constitutes 7%, and the rare adenoid cystic carcinomas, at less than 0.1%. Many of these subtypes, such as tubular and medullary carcinomas, carry an excellent prognosis. Metaplastic carcinomas, however, often have an aggressive behavior. These tumors are characterized by the presence of homologous (epithelial) or heterologous (mesenchymal) elements. Two types have been described: Squamous and pseudosarcomatous metaplasia. Invasive duct carcinoma not otherwise specified (NOS) is a generic term that includes tumors that may express more than one element of the specific forms of duct carcinoma. Infiltrating Lobular Carcinoma Infiltrating lobular carcinoma has been reported to constitute 10% to 14% of invasive carcinomas. These carcinomas are characterized by uniform cells with small, round nuclei and limited cytoplasm. The presence of intracytoplasmic mucin vacuoles often gives the cells the appearance of signet-ring cells. The cells tend to grow circumferentially around ducts and lobules with a linear arrangement. This pattern is referred to as “Indian-file” or targetoid growth. There is often an associated desmoplastic stromal reaction. Inflammatory Carcinoma Inflammatory carcinoma is characterized by cutaneous findings present with an underlying invasive carcinoma. Usually the invasive tumor is a poorly differentiated infiltrating duct carcinoma. Upon microscopic evaluation, skin involvement often reveals tumor emboli in dermal lymphatics with an associated lymphocytic reaction in the dermis. Metastases from Extramammary Tumors The most common primary site of an occult extramammary tumor is the lung. Other primary sites include the ovaries, uterus, kidneys, and stomach. In those previously diagnosed, melanoma, prostate, cervix, uterus, and urinary bladder are the most common sites. Metastatic ovarian cancer may simulate papillary or mucinous

carcinoma of the breast. A workup and history are often helpful in difficult cases. Often, identification of an in situ component helps to provide definitive evidence of a mammary origin. Biologic Markers/Prognostic Factors Axillary Lymph Node Status The most important prognostic factor is nodal status. The presence of metastasis, as well as the number of lymph nodes involved, is significant and correlates with local failure and distant metastases. It is predictive of overall survival. Tumor Size Tumor size correlates with the incidence of lymph node metastases. The size of the tumor is also important, even in the absence of lymph node involvement. Patients with tumors less than 1 cm in size, or with good histologic types measuring less than 3 cm, do very well. Histologic Grade Histologic grade also correlates with breast cancer outcome. Poorly differentiated tumors have been associated with more aggressive behavior. Estrogen Receptors/Progesterone Receptors The hormone receptors can be measured by immunohistochemical (IHC) studies using monoclonal antibodies directed against the receptors. Positivity correlates with response to antihormonal agents as well as better prognosis. HER2/neu HER2/ neu is an oncogene whose protein product may function as a growth factor receptor. It can be detected by IHC demonstration of the protein product or by gene amplifications. Overexpression or amplification has been shown to correlate with a poor prognosis; however, the studies differ with regard to the method of detection used, as well as the interpretation of results. HER2/ neu has been used as a predictor of response to certain chemotherapeutic agents. Particularly, increased-response, doxorubicin-based therapy has been reported in the treatment of patients with positive nodes and overexpression of HER2/ neu. Herceptin (trastuzumab) is a humanized anti-HER2 antibody against the extracellular domain of the 2-neu oncoprotein. Its use in the metastatic setting has been reported to demonstrate an increase in response rate and prolongation of disease-free and overall survival. p53 A tumor-suppressor gene, p53 has a protein product that is a nuclear transcription factor with many functions, including regulation of the cell cycle and apoptosis. Most clinical studies have used IHC to study protein expression. Accumulation of p53 protein has been reported to correlate with reduced survival in some studies.

STAGING OF BREAST CANCER Staging of Breast Cancer Using the Tumor–Node–Metastasis (TNM) System The American Joint Committee on Cancer (AJCC) determines staging of breast cancer. The AJCC staging system is a clinical and pathologic staging system based on the tumor–node–metastasis (TNM) system. The new updated AJCC staging system (2002) incorporates sentinel node staging. It distinguishes micrometastasis from isolated tumor cells on the basis of size and histologic evidence of malignant activity. In the current AJCC staging system, supraclavicular lymph node metastasis is now classified as N3 disease, rather than M1 disease, as in the old system ( Table 50.7, Table 50.8).

TABLE 50.7. American Joint Committee on Cancer Staging for breast cancer

TABLE 50.8. Stage by tumor, node, metastasis (TNM)

TREATMENT OF BREAST CANCER Mastectomy William Halsted performed a radical mastectomy in 1894. The guiding principle at this time was centered on the belief that cancer originates in the breast and spreads in a stepwise fashion first to the regional lymph nodes, and then to distant sites. Removal of all the breast tissue, pectoral muscles, and axillary contents was the standard surgical treatment. Over the next several decades, two simultaneous trends existed. One involved less-radical surgery, which included removal of the breast and axillary contents, but preserved the pectoral muscles and more skin. This was described by Patey and Dyson in 1948, and is the modern-day modified radical mastectomy. The other trend involved more extensive surgery, the extended radical mastectomy, which included en-bloc removal of the internal mammary chain at the time of radical mastectomy. Subsequent randomized trials failed to show a survival advantage with the extended radical mastectomy when compared with the radical mastectomy. Breast-Conservation Therapy The shift toward less radical surgery at the time of mastectomy occurred for several reasons. As earlier diagnosis of breast cancer with smaller tumors and less involvement of pectoral muscles occurred, the need for radical procedures decreased. In addition, even with radical mastectomy not all patients were cured, and although regional recurrences were low, patients died of distant disease. The morbidity of the radical mastectomy was well documented, including lymphedema, immobility of the shoulder, and disfigurement. A shift to breast conservation followed the same trends. The initial trials with radiation therapy using radium implants at the Princess Margaret Hospital in Toronto had promising results. This led to randomization studies comparing breast-conservation therapy with mastectomy. Table 50.9 lists randomized trials that compared radical and modified radical mastectomy in stage I and stage II carcinoma of the breast.

TABLE 50.9. Conservation surgery and irradiation in Stage I and II carcinoma of the breast: Results of randomized comparisons with radical or modified mastectomy

In breast-conserving surgery, a wide local excision is performed with excision of the tumor and a 1- to 2-cm rim of normal tissue. This excision is referred to as “lumpectomy” or a “tumorectomy.” This differs from a quadrantectomy, in which a resection of the tumor with the overlying skin and the involved quadrant of the breast is performed. The six randomized trials differed with respect to the type of wide local excision performed as well as tumor size in the patients who were randomized. In the Milan trial, a quadrantectomy was performed. In the Insitut Gustave Roussy trial, the “tumorectomy” performed was removal of the tumor and a 2-cm margin of normal tissue. In the United States, the NSABP B-06 trial did not specify the margins on the lumpectomy specimen, providing they were grossly free of tumor. In all trials, the authors noted comparable disease-free survival in both arms. The only difference noted was in local recurrence. In addition to the randomized trials, there are many nonrandomized reports published with similar results in survival between breast conservation and mastectomy. Radiation therapy is an important component of breast conservation. Adequate surgical margins are required to be negative on pathologic inspection. After healing, the radiation therapy is planned. Treatment to the entire breast should be at a dose of 1.8 to 2.0 Gy per day for a total of 45 to 50 Gy. A 10- to 15-Gy electron-therapy boost is often given to the lumpectomy bed. Breast conservation depends on the use of radiation. The question of whether lumpectomy alone would yield similar results has been addressed through randomized trials. Local recurrence rates of 18% to 40% with lumpectomy alone have been reported compared with 2% to 14% with lumpectomy and radiation. In these studies, a significant reduction in relapse was noted in the lumpectomy-and-radiation arm. The NSABP B-17 trial is the only randomized trial of lumpectomy alone versus lumpectomy and radiation in patients with DCIS. Ipsilateral relapse was 7% in the radiation arm versus 16.4% in the lumpectomy-alone arm. Actuarial 5-year survival showed an ipsilateral recurrence rate of 7.5% for noninvasive and 2.9% for invasive recurrence in the lumpectomy-plus-radiation arm. By comparison, the rates were 10.4% and 10.5%, respectively, in the lumpectomy-alone arm. The overall survival was excellent and comparable to that seen with mastectomy. In cases of invasive recurrence, patients in the lumpectomy-plus-radiation arm can be salvaged with mastectomy. Patient Selection Possible contraindications can interfere with offering a patient breast-conservation therapy. Cosmetic result should be considered when deciding whether to offer breast-conserving surgery; tumor size in relation to breast size is an important consideration. In addition, other factors that may affect the ability to receive irradiation must be considered (i.e., history of prior breast or chest irradiation, concurrent pregnancy, autoimmune connective-tissue disease). Particularly with a history of systemic lupus erythematosis, radiation may not be a possibility. Women with multiple cancers in the breast are not candidates for this approach and require a mastectomy. In addition, in patients with extensive calcifications on a mammogram suggesting a diffuse process may be better treated with mastectomy. An area of controversy remains over the status of negative margins at the time of lumpectomy. Generally a reexcision should be performed. Extensive intraductal component (EIC) is a condition that exists when greater than 25% of the tumor is associated with DCIS. In these cases, the invasive component may be outside the area of the intraductal carcinoma. This has been reported to have a higher relapse rate with breast conservation; however, it is thought to be secondary to margin status, since margin involvement may be an indication of residual intraductal carcinoma. In DCIS, Silverstein and associates developed the Van Nuys prognostic index (VNPI). The system combines the scores for histologic grade, tumor size, and margin status of a lesion in order to obtain an overall score. It considers margins of 1 cm to indicate a decreased rate of local relapse, even with no radiation. Management of the Axilla The axilla is a pyramidal space between the arm and thoracic wall. It contains the axillary vessels and their branches, the brachial plexus and its branches, and lymph nodes embedded in fatty tissue. The primary route of lymphatic drainage of the breast is through the axillary lymph nodes. The lymph nodes are also divided into levels based on location relative to the pectoralis minor. Level 1 lymph nodes lie lateral to the lateral border of the pectoralis minor muscle. Level 2 nodes lie behind the pectoralis minor muscle, and level 3 nodes are medial to the medial border of this muscle. In an axillary dissection, nerve branches of the brachial plexus are encountered. The lateral and medial pectoral nerves supply the pectoralis muscles. The thoracodorsal nerve runs downward, and innervates the latissimus dorsi. The long thoracic nerve is located on the medial wall of the axilla on the serratus anterior. It arises from the C5 to C7 roots, and injury to these nerves results in paralysis to part or all of the serratus anterior. The functional deficit is inability to raise the arm above the level of the shoulder. At the time of Halsted's radical mastectomy procedure, a complete axillary dissection was performed. The status of the axilla is the most important prognostic factor for breast cancer. The use of axillary dissection has, in the past, been demonstrated to significantly decrease local recurrence, which may ultimately translate to a survival advantage. Clinical examination of the axilla is inaccurate, since even in patients with a T1 lesion, there is a 10% risk of lymph node metastasis. Prior to the use of sentinel lymph node biopsy, there was no accurate method to adequately stage the axilla without an axillary dissection. Axillary lymph node sampling (i.e., removal of only a few lymph nodes) was inadequate. Metastatic involvement of lymph nodes usually occurs in a stepwise manner. Rosen and co-workers demonstrated the incidence of ?skip metastasis? to be less than 2%. A complete level 1 and 2 lymph node dissection provides excellent local control, and local recurrence after this procedure has been shown to be less than 1%. The NSABP B-04 trial randomized patients with clinically negative nodes to one of three groups: radical mastectomy, total mastectomy with nodal irradiation, or total mastectomy alone. The patients who had no axillary nodal therapy had an overall worse survival if recurrence occurred. The morbidity associated with axillary lymphadenectomy is not inconsequential. About 10% to 15% of patients will develop lymphedema. In addition, numbness, pain, or weakness contribute to a significant decrease in the quality of life in these patients. Sentinel Lymph Node Biopsy Sentinel lymph node (SLN) biopsy in breast cancer evolved out of efforts to minimize the morbidity associated with axillary lymph node dissection while still providing important staging information. Initial studies in melanoma, by Morton and colleagues in 1992, demonstrated the feasibility of the concept. Methods employed include use of blue dye or radioisotope ( Fig. 50.5).

FIG. 50.5. Intraoperative identification of the sentinel lymph node. Several studies have confirmed the accuracy of SLN biopsy. In the majority of the studies, successful identification of the SLN occurs between 92% and 98% of the time. The combination of blue dye and isotope has been reported to be better for identification of the SLN; when used in combination, the positive predictive value of the technique approaches 100%, with a negative predictive value close to 95%. The false-negative rate is about 5% to 10% in most studies. The validity of the SLN concept lies in the ability of the SLN to predict the status of the regional lymphatic basin. Turner and colleagues performed IHC staining of all lymph nodes, sentinel and nonsentinel, in a series of patients undergoing standard axillary dissection with negative nodes. Of 157 SLNs, 10 (6%) demonstrated IHC positivity compared with 1 of 1,087 (0.09%) of the non-SLNs. This provided validity of the SLN concept. Overall, greater scrutiny is paid to SLNs through serial sectioning and IHC stains. This is

possible because only a few SLNs are obtained at the time of the procedure, and would not be cost-effective in a standard axillary dissection that yields an average of 20 nodes. The use of the SLN biopsy procedure is widely employed in invasive breast cancer. In certain situations, however, a standard axillary dissection should still be considered. These include: in cases of palpable suspicious nodes in cases of large lesions (the majority of reports indicate that the procedure is reliable if there are no suspicious nodes in the axilla) in cases of prior radiation, of a large excisional cavity close to the axilla, or any condition in which disruption of the lymphatics is suspected. Systemic Treatment Adjuvant treatment in breast cancer was first used over 100 years ago. In 1894, Beatson reported on the results of oophorectomy and response rate in the metastatic breast cancer setting. Initially, the use of systemic therapy involved the use of single-agent chemotherapy; later, multi-agent chemotherapy was employed. As in most breast cancer trials, the initial reports were of patients with metastatic disease. Multiple randomized studies have demonstrated that the addition of chemotherapy improves overall survival in patients with breast cancer. The decision to use adjuvant chemotherapy or hormonal therapy depends on certain factors such as the size of the primary tumor, lymph node status, and the presence or absence of metastatic disease. In addition, the expression or lack of expression of ER or PR (ER/PR status) is also an important factor. Adjuvant chemotherapy is standard treatment for patients with positive nodes or large tumors. The combination of CMF (cyclophosphamide, methotrexate, and 5-fluorouracil [5-FU]) has been used for many years in the treatment of patients with breast cancer. An anthracycline-based regimen such as FAC (5-FU, adriamycin, and cyclophosphamide) has been used in patients with high risk factors for recurrence. The use of paclitaxel is also considered in this setting. For patients with intermediate risk factors, such as ER-negative tumors (>1 cm in size) and negative nodes, chemotherapy is considered. The Early Breast Cancer Trialists' Collaborative Group was formed in 1985 to analyze all available, properly conducted, randomized trials. A second overview was done in 1990, and a third in 1995. In women under the age of 50, administration of multi-agent chemotherapy decreased the annual risk of relapse by 35% and mortality by 27%. With 10 years of follow-up, this translates into absolute gains of 7% in patients with node-negative tumors and 11% in those with node-positive tumors. For women over the age of 50 years, the benefits of chemotherapy were smaller but still significant. Annual risk reduction was 20% for recurrence and 11% for mortality. At 10 years of follow-up, this risk reduction translated into absolute gains of 2% in patients with node-negative tumors and 3% in patients with node-positive tumors. It is important to note that in the overview, different regimens of CMF were used; but the greatest benefit was seen in those using CMF for 6 months or longer. The question of whether to use CMF or FAC in high-risk patients (tumors = 2 cm in size or ER/PR-negative with negative nodes) has been addressed. While the anthracycline regimen is more toxic than CMF, trials have shown superiority with this regimen. Presently, other factors are used to determine which regimen to recommend in this subgroup. The most promising candidate factor is HER2/ neu overexpression because increased response rates with an anthracycline-based therapy have been reported in cases where there is overexpression of HER2/ neu. The use of taxanes such as docetaxel and paclitaxel as adjuvant treatment in combination with anthracycline-based chemotherapy is being investigated. Postmenopausal women with negative nodes and ER-negative tumors greater than 1 cm in size are also considered for chemotherapy. In these patients, CMF is usually the treatment of choice. Metastatic Disease The goal of therapy in metastatic disease is palliation of symptoms, as cure is unlikely. The majority of patients with metastatic disease receive antihormonal therapy. First-line agents include tamoxifen, or aromatase inhibitors such as letrozole or anastrozole. These agents offer a 20% response with in ER/PR-positive tumors. Disease stabilization is the goal of therapy, and because these therapies are less toxic than chemotherapy, most patients will remain on them for prolonged periods of time. Upon failure of these agents, however, chemotherapy is the next step. High-Dose Chemotherapy Five large randomized trials have been conducted addressing the use of high-dose chemotherapy with bone-marrow or stem-cell rescue in metastatic breast cancer. Only one trial, conducted in South Africa, showed a lower rate of relapse. The other four trials showed no increase in overall survival. The investigators of the trial conducted in South Africa subsequently admitted to fraud. Thus, it is felt that high-dose chemotherapy offers no survival advantage over conventional treatment approaches. Neoadjuvant Chemotherapy Preoperative or neoadjuvant chemotherapy is attractive, as it may reduce the amount of disease present and thereby facilitate in obtaining clean surgical margins when the disease is still confined to the breast. This is often the case in inflammatory breast cancer, or in N2 disease, in which neoadjuvant chemotherapy may improve surgical resectability. A significant response of 50% to 90% has been seen with this approach. Down-staging of the tumor, as well as the axillary lymph nodes, has been reported. The NSABP B-18 trial randomized patients with operable breast cancer to receive adjuvant chemotherapy preoperatively for four cycles followed by reevaluation for surgical treatment. Results for this group were compared with those for patients who received surgery followed by adjuvant chemotherapy. Preoperative chemotherapy was associated with a higher incidence of breast conservation. Also, a significant reduction in the number of positive axillary nodes was noted in the preoperative chemotherapy group. There was, however, no difference in overall survival between those patients who received preoperative chemotherapy and those who received postoperative chemotherapy. Several subsequent trials have reported a higher local recurrence rate for patients treated with neoadjuvant chemotherapy and breast-conservation therapy. Radiation Therapy Radiation therapy is used in conjunction with lumpectomy for patients opting for breast conservation. The dose used is 1.8 to 2.0 Gy per day for a total of 45 to 50 Gy to the entire breast. A 10- to 15-Gy electron-therapy boost is often given to the lumpectomy bed. Postmastectomy chest-wall irradiation is used with increasing frequency. Patients with large tumors (T3 lesions) and more than four positive nodes are offered chest-wall radiation because they are at risk for local–regional failure. Chest-wall radiation is 50 Gy over 5 weeks, with a 10-Gy boost to the mastectomy scar. For patients with a chest-wall recurrence, the option of surgical debulking followed by chest-wall irradiation is employed. Radiation therapy can also be used in the palliative setting. It can be used for metastatic lesions to the bone or brain and can help to alleviate the patient's symptoms. Stage-Directed Therapy Patients with intraductal carcinoma and stage I and stage II breast cancer have the options of breast-conservation therapy and mastectomy. For patients with invasive breast cancer, the axillary nodes can be addressed with an SLN biopsy and, possibly, an axillary dissection. In intraductal carcinoma, an SLN biopsy or axillary dissection is generally not indicated. There are special circumstances of intraductal carcinoma such as high-grade DCIS or extensive DCIS in which a microinvasive component may be associated with the intraductal carcinoma. In these circumstances, particularly if a mastectomy is being performed, an SLN biopsy may be considered. For patients with more extensive disease, a mastectomy may be necessary. The use of postmastectomy chest-wall irradiation in these patients may also be considered. Breast Reconstruction

Breast reconstruction represents a major advance in cancer rehabilitation for patients undergoing a mastectomy. Previously, a 2-year surveillance period was recommended prior to reconstruction for detection of local disease recurrence. Immediate reconstruction has not interfered with disease detection, however, and it has the advantage of combining the two procedures into one. In addition, a greater amount of skin can be saved with planned immediate reconstruction, and the scar tissue that would be encountered with delayed procedure can be avoided. A delayed reconstruction can be performed if the patient is ambiguous about the reconstruction, or if operative risk is increased with prolonged anesthesia. It is also considered in those patients with locally advanced disease if a delay in adjuvant irradiation or chemotherapy is anticipated because of the reconstruction. Reconstruction options include expandable breast prosthesis (implant) and autologous tissue transfer. Tissue-transfer operations may yield the greatest symmetry between breasts, especially with larger breasts; however, they take longer, require greater surgical expertise, involve a longer recovery, and result in another scar at the donor site. The donor site may be the latissimus dorsi, transverse rectus abdominus, or gluteal muscle.

SPECIAL ISSUES Hereditary Breast Cancer One of the most characteristic features of hereditary breast cancer is its tendency to manifest at a young age. In the Breast Cancer Consortium's study of BRCA1-linked families that transmit BRCA1 mutations, more than 80% of breast cancers occurred in women under 50 years of age. Pathologic Features Most BRCA1-associated breast cancers have been reported to be of an infiltrating ductal type with an over-representation of poorly differentiated high-grade types. The tumors tend to be ER/PR-negative. Less than 20% of these cancers are ER/PR-positive, even when age matched with non– BRCA1-associated controls. In BRCA2 there is less of this over-representation of aggressive histology. Overall, BRCA2-associated breast cancers tend to be ER/PR-positive. Stage Most studies demonstrate that BRCA-associated breast cancers are seen at a stage comparable to non– BRCA-associated breast cancers. The incidence of axillary metastasis does not appear to be significantly different in patients with BRCA-associated breast cancers. In the literature, conflicting study data exist regarding the prognosis of patients with BRCA-associated breast cancer. Some studies have conferred a worse prognosis in patients with certain mutations in BRCA1. Some of these reports have been of highly selected groups of women, and thus further study is necessary in larger series of women with BRCA1 and BRCA2 mutations. Treatment Although some researchers have questioned the role of breast-conservation therapy in women with hereditary predisposition, there is no reason to suspect a unique survival advantage for mastectomy in these women. Studies that have examined the outcomes of breast-conservation therapy in women with BRCA mutations are small with variable follow-up. Local ipsilateral recurrence appears to be about 15% at 5 years. Although this is higher than would be expected for patients treated with breast-conservation therapy, it is within the range observed for treatment of young women with breast cancer. It is likely to be an influence of the age at diagnosis and not because of radiation resistance. After breast-conservation therapy, however, the breast tissue remains at risk for developing a second primary. Women with BRCA mutations may be at risk for a late ipsilateral recurrence because of the development of a second primary breast cancer. The degree of contralateral risk needs to also be addressed with patients. Some studies have reported an estimated average risk of 2.5% to 5% per year in BRCA1 mutation carriers, and this risk may be higher with younger age at diagnosis. In patients with BRCA2, the risk appears lower, estimated at 1.8% per year. Unlike the BRCA1-associated risk, in BRCA2 the age dependence is unknown. Chemoprevention Chemoprevention is the principle that cancer prevention can be achieved through pharmacologic intervention. It refers to the use of a medication in a healthy patient to reduce the risk of a particular cancer. It is an option for all women with significant risk for future breast cancer development. Currently, the only FDA-approved medication for prevention of breast cancer is tamoxifen. Tamoxifen is a selective estrogen-receptor modulator that acts as an antiestrogen in breast tissue. It has been used in the adjuvant setting in breast cancer since 1972 for both metastatic and early breast cancers. A reduction of 40% to 50% in contralateral breast cancer was seen in the adjuvant setting; thus, it was thought to be an ideal agent to use in a chemopreventive setting. The trial to determine the role of tamoxifen in chemoprevention was performed by the NSABP. In this randomized, double-blinded trial, women with a projected risk of breast cancer of greater than 1.66% over a 5-year period received either tamoxifen or a placebo for a period of 5 years. The Gail model was used to assess the risk. When an independent reviewing agency verified a 50% reduction in both invasive and noninvasive breast cancer cases in the population taking tamoxifen, the trial results were unblinded earlier than expected. Shortly thereafter, the use of tamoxifen was approved for chemoprevention. Two other chemoprevention trials using tamoxifen have been reported: the Italian tamoxifen-prevention study and the Royal Marsden Hospital tamoxifen trial in the United Kingdom. These two trials did not reveal a statistically significant reduction in breast cancer risk in women randomized to tamoxifen. However, the studies differed in respect to subject numbers, median age, eligibility criteria, risk, and use of HRT. Raloxifene is also a selective estrogen-receptor modulator (SERM) and is FDA-approved for use in osteoporosis. In studies in which raloxifene was used for treatment of osteoporosis, a secondary finding was a noted reduction of breast cancer risk of 50% to 70% in the population taking the medication. The women in those trials were at fairly low risk of breast cancer development. A randomized, double-blinded trial to compare raloxifene with tamoxifen in a population of postmenopausal women at increased risk for breast cancer development is currently open, the Study of Tamoxifen and Raloxifene (STAR) trial/P-2 study. In the NASBP P-1 study there was an increased risk of developing endometrial cancer with tamoxifen use (annual risk 2.3/1,000 women vs. 0.9/1,000 women in the placebo group). All cases of endometrial cancer in the patients taking tamoxifen were early stage. Another important side effect noted was a higher incidence of thromboembolic phenomena. Raloxifene does not increase the risk of endometrial cancer and thus may prove to be an alternative to tamoxifen in a chemopreventive role. An ideal SERM would have an antiestrogenic activity on breast and uterine tissues, but estrogenic effects on bone and the cardiovascular system among others. Current investigations into other SERMs are being conducted.

CONCLUSIONS Screening for breast cancer is clearly indicated for all women at the appropriate age. Determining a woman's unique risk factors will help to determine both the age at which that screening should begin and also the intensity of that screening. It will also help to identify those women who need to be counseled regarding options for prevention of breast cancer. The hope is that, by correctly identifying high-risk populations and then applying appropriate screening schedules and chemopreventive agents, many cases of breast cancers will be averted completely and that those that still occur will be found at the earliest stages. The role of the obstetrician and gynecologist in providing information on breast cancer diagnosis and screening is very important. In addition, the understanding of breast disease, benign and malignant, is crucial not only in the diagnosis of disease, but also in helping to guide women in their treatment and follow-up. SUGGESTED READINGS Introduction Jemal A, Thomas A, Murray T, et al. Cancer statistics, 2002. CA Cancer J Clin 2002;52:23–47.

Anatomy of the Breast Bland KI, Copeland III EM. The breast: comprehensive management of benign and malignant disease. Philadelphia: WB Saunders, 1998.

Epidemiology of Breast Cancer Bhatia S, Robison LL, Oberlin O, et al. Breast cancer and other second neoplasms after childhood Hodgkin's disease. N Engl J Med 1996;334:745. Kelsey J. A review of the epidemiology of human breast cancer. Epidemiol Rev 1979;1:74.

Risk Factors and Assessment Colditz GA, Hankinson SE, Hunter DJ, et al. The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. N Engl J Med 1995;332:1589.

Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the women's health initiative randomized controlled trial. JAMA 2002;288:321–333. Stanford JL, Weiss NS, Voigt LF, et al. Combined estrogen and progestin hormone replacement therapy in relation to risk of breast cancer in middle-aged women. JAMA 1995;274:137. Willet WC, Stampfer MJ, Calditz GA, et al. Moderate alcohol consumption and risk of breast cancer. N Engl J Med 1987;315:1174.

Relative Risk Gail MH, Brinton LA, Byar DP, et al. Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst 1989;81:1879.

BRCA1 and BRCA2 Berry DA, Parmigiani G, Sanchez J, et al. Probability of carrying a mutation of breast-ovarian cancer gene BRCA1 based on family history. J Natl Cancer Inst 1997;89:227–238. Porter D, Cohen B, Wallace M, et al. Breast cancer incidence, penetrance and survival in probable carriers of BRCA1 gene mutation in families linked to BRCA1 on chromosome 17q12-21. Br J Surg 1994;81:1512. Wooster R, Neuhausen SL, Mangion J, et al. Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q12-13. Science 1994;265:2088.

History and Physical Examination Donegan WL. Evaluation of a palpable breast mass. N Engl J Med 1992;327:937.

Mammography Alexander FE, Anderson TJ, Brown HK, et al. 14 years of follow-up from the Edinburgh randomised trial of breast-cancer screening. Lancet 1999;353:1903–1908. Andersson I, Aspegren K, Janzon L, et al. Mammographic screening and mortality from breast cancer: the Malmo mammographic screening trial. BMJ 1988;297:943–948. Bjurstam N, Bjorneld L, Duffy SW, et al. The Gothenburg breast screening trial: first results on mortality, incidence, and mode of detection for women ages 39–49 years at randomization. Cancer 1997;80:2091–2099. Chu KC, Smart CR, Tarone RE. Analysis of breast cancer mortality and stage distribution by age for the Health Insurance Plan clinical trial. J Natl Cancer Inst 1988;80:1125–1132. D'Orsi C. The American College of Radiology mammography lexicon: an initial attempt to standardize terminology. AJR 1996;166:779. Frisell J, Lidbrink E, Hellstrom L, et al. Followup after 11 years—update of mortality results in the Stockholm mammographic screening trial. Breast Cancer Res Treat 1997;45:263–270. Gemignani ML. Breast Cancer. In Barakat RR, Bevers M, Gershenson DM, et al, eds. The Memorial Sloan-Kettering & MD Anderson Handbook of Gynecologic Oncology. London: Martin Dunitz Publishers, 2002:297–319. Gotzsche PC, Olsen O. Is screening for breast cancer with mammography justifiable? Lancet 2000;355:129–134. Hendrick RE, Smith RA, Rutledge JH 3rd, et al. Benefit of screening mammography in women aged 40–49: a new meta-analysis of randomized controlled trials. J Natl Cancer Inst Monogr 1997;22:87–92. Miller AB, Baines CJ, To T, et al. Canadian National Breast Screening Study: 1. Breast cancer detection and death rates among women aged 40 to 49 years. CMAJ 1992;147:1459–1476. Miller AB, To T, Baines CJ, Wall C. The Canadian National Breast Screening Study: update on breast cancer mortality. J Natl Cancer Inst Monogr 1997;22:37–41. Statement of the American Society of Clinical Oncology: genetic testing for cancer susceptibility, adopted February 20, 1996. J Clin Oncol 1996;14:1730–1736. Tabar L, Fagerberg G, Chen HH, et al. Efficacy of breast cancer screening by age. New results from the Swedish Two-County Trial. Cancer 1995;75:2507–2517.

Breast Ultrasound and Magnetic Resonance Imaging Dao TH, Rahmouni A, Campana F, et al. Tumor recurrence versus fibrosis in the irradiated breast: differentiation with dynamic gadolinium-enhance MR imaging. Radiology 1993;187:751.

Diagnostic Evaluation Dershaw DD, Caravella BA, Liberman L, et al. Limitations and complications in the utilization of stereotaxic core breast biopsy. Breast J 1996;2:1. Dershaw DD. Needle localization for breast biopsy. In: Dershaw DD, ed. Interventional breast procedures. New York: Churchill Livingstone, 1996:25. Layfield LJ, Glasgow BJ, Cramer H. Fine-needle aspiration in the management of breast masses. Pathol Annu 1989;24(Pt 2):23–62. Liberman L, Abramson AF, Squires FB, et al. The breast imaging reporting and data system: positive predictive value of mammographic features and final assessment categories. AJR Am J Roentgenol 1998;171:35–40. Liberman L. Advantages and disadvantages of minimally invasive breast biopsy procedures. Semin Breast Dis 1998;1:84–94. Liberman L, Feng T, Dershaw DD, et al. US-guided core beast biopsy: use and cost-effectiveness. Radiology 1998;208:717. Orel SG, Kay N, Reynolds C, et al. BI-RADS categorization as a predictor of malignancy. Radiology 1999;211:845–850. Osborne C, Yochmowitz M, Knight W, et al. The value of estrogen and progesterone receptors in the treatment of breast cancer. Cancer 1980;46:2884.

Benign Breast Conditions Nonmalignant conditions of the breast. ACOG Technical Bulletin. In: The American College of Obstetricians and Gynecologists Compendium of Selected Publications. Washington, DC: 2001:616–621.

Breast Cancer Pathology, Biologic Markers and Prognostic Factors Fisher B, Bauer M, Wickerham DL. Relation of number of positive axillary nodes to the prognosis of patients with primary breast cancer: a NSABP update. Cancer 1983;52:1551. Fisher ER, Sass R, Fisher B, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (Protocol 6). I. Intraductal carcinoma (DCIS). Cancer 1986;57:197. Prost S, Le M, Douc-Rasy S, et al. Association of c-erbB2-gene amplification with poor prognosis in non-inflammatory breast carcinomas but not in carcinomas of the inflammatory type. Int J Cancer 1994;58:763. Silverstein MJ, Lagios MD, Craig PH, et al. A prognostic index for ductal carcinoma in situ of the breast. Cancer 1996;77:2267.

Staging of Breast Cancer Breast. In: American Joint Committee on Cancer: AJCC Cancer Staging Manual, fifth ed. Philadelphia: Lippincott-Raven Publishers, 1997:171–180.

Treatment of Breast Cancer

Breast-conservation Therapy Fisher B, Dignam J, Wolmark N, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from the National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol 1998;16:441. Silverstein MJ, Lagios MD, Groshen S, et al. The influence of margin width on local control of ductal carcinoma in-situ of the breast. N Engl J Med 1999;340:1455.

Management of the Axilla Petrek JA, Blackwood MM. Axillary dissection: current practice and technique. Curr Probl Surg 1995;32:257–323.

Sentinel Lymphnode Biopsy Morton DL, Wen DR, Wong JH, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992;127:392. Turner RR, Ollila DW, Krasne DL, et al. Histologic validation of the sentinel lymph node hypothesis for breast carcinoma. Ann Surg 1997;226:271.

Systemic Treatment Early Breast Cancer Trialists' Collaborative Group. Polychemotherapy for early breast cancer: an overview of the randomised trials. Lancet 1998;352:930. Fisher B, Brown A, Mamounas E, et al. Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997;15:2483. Weiss RB, Gill GG, Hudis CA. An on-site audit of the South African trial of high-dose chemotherapy for metastatic breast cancer and associated publications. J Clin Oncol 2001;19:2771–2777.

Metastatic Disease Fisher ER, Palekar A, Rockette H, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (Protocol No. 4). V. Significance of axillary nodal micro- and macrometastases. Cancer 1978;42:2032–2038. Rosen PP, Lesser ML, Kinne DW. Discontinuous or “skip” metastases in breast carcinoma. Ann Surg 1983;197:276.

Special Issues Hereditary Breast Cancer Robson ME, Boyd J, Borgen PI, et al. Hereditary breast cancer. Curr Probl Surg 2001;38:377–480.

Chemoprevention Cummings SR, Eckert S, Krueger KA, et al. The effect of raloxifene on risk of breast cancer in postmenopausal women: results from the MORE randomized trial. Multiple Outcomes of Raloxifene Evaluation. JAMA 1999;281:2189–2197. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371–1388. Powles T, Eeles R, Ashley S, et al. Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomised chemoprevention trial. Lancet 1998;352:98–101. Veronesi U, Maisonneuve P, Costa A, et al. Prevention of breast cancer with tamoxifen: preliminary findings from the Italian randomised trial among hysterectomised women. Italian Tamoxifen Prevention Study. Lancet 1998;352:93–97.

Chapter 51 Vulvar and Vaginal Cancer Danforth’s Obstetrics and Gynecology

Chapter 51 Natalie S. Gould, Joan L. Walker, and Robert S. Mannel

Vulvar and Vaginal Cancer

VULVAR CARCINOMA IN SITU Epidemiology Presentation Staging/Spread Patterns Treatment Other Histologic Subtypes Sarcoma VAGINAL CARCINOMA Vaginal Intraepithelial Neoplasia Squamous Carcinoma Clear Cell Adenocarcinoma Melanoma Sarcoma Botryoides Endodermal Sinus Tumor Lymphoma Metastatic Disease SUMMARY POINTS SUGGESTED READINGS Vulvar Intraepithelial Neoplasia Vulvar Carcinoma: Etiology, Presentation, and Staging Treatment VULVA MELANOMA Paget Disease Vaginal Intraepithelial Neoplasia Vaginal Cancer

Vulvar and vaginal cancer represent uncommon gynecologic cancers that occur most often in older women. Squamous lesions are the most frequent histology and risk factors are similar for both disease sites. This chapter describes the epidemiology, clinical presentation, patterns of spread, and treatment of squamous cell carcinomas of the vulva and vagina. Less common histologic subtypes including Paget disease, melanoma, adenocarcinomas, and sarcomas are also reviewed.

VULVAR CARCINOMA IN SITU Vulvar carcinoma theoretically results from malignant transformation of a vulvar carcinoma in situ as is seen with cervical squamous lesions. Unlike squamous lesions of the cervix, the natural history of vulvar intraepithelial neoplasia (VIN) is less well understood. The incidence of vulvar dysplasia has increased over the last 20 years, particularly among younger women. A report from Austria demonstrated a 307% increase in the overall incidence of high-grade VIN and a 394% increase among women under 50 years of age between 1985 and 1998. In a review of Surveillance, Epidemiology, and End Results (SEER) data, Sturgeon found that the incidence of VIN III nearly doubled between 1973 and 1976 and 1985 and 1987, from 1.1 to 2.1 per 100,000 woman years. Factors implicated in this increase include increased human papillomavirus (HPV) infection, increased surveillance, tobacco use, and immunosuppression, either with human immunodeficiency virus (HIV), organ transplant, or diabetes. Patients with VIN most commonly present with pruritis and vulvar lesions. These lesions may appear scaly, white, red, or hyperpigmented ( Fig. 51.1, Fig. 51.2, Fig. 51.3 and Fig. 51.4). Careful inspection with 5% acetic acid and liberal use of punch biopsy are the cornerstones of diagnosis. An underlying malignancy may be present in 7% to 22% of patients who undergo surgical excision for vulvar carcinoma in situ. Wide local excision with at least a 5-mm margin is the preferred management option as it allows pathologic confirmation and is associated with less morbidity than skinning vulvectomy. Skinning vulvectomy with split thickness skin graft may be an option in patients with widespread disease. Laser ablation is also an effective nonmutilating option in patients with multifocal or clitoral disease. Recurrences are frequent (10%–50%) despite negative surgical margins and therapy should be tailored to symptom control and ruling out underlying malignancy. Patients should be followed every few months with careful visual inspection of the vulva and taught self-exam skills as well.

FIG. 51.1. Vulvar carcinoma in situ presenting as white or hyperpigmented lesion. See color figure 51.1.

FIG. 51.2. Vulvar carcinoma in situ before application of 5% acetic acid. See color figure 51.2.

FIG. 51.3. Vulvar carcinoma in situ after application of 5% acetic acid. See color figure 51.3.

FIG. 51.4. Vulvar carcinoma in situ with full thickness involvement of dysplastic cells.

Epidemiology Vulvar carcinoma is the fourth most common genital tract malignancy in women, representing 3% to 5% of gynecologic malignancies and affecting an estimated 3,600 women in the United States in 2001. The majority of cancers are squamous in origin with occasional cases of basal cell carcinoma, melanoma, adenocarcinoma, and Paget disease. Vulvar cancer tends to be a disease of older women with a mean age at diagnosis of approximately 65 years. As the population ages, more older women will be at risk for vulvar carcinoma. Evidence exists that there are two distinct types of vulvar carcinoma with different etiologies. Tumors in older women are often unifocal and may be associated with chronic vulvar inflammation of long-standing duration such as lichen sclerosus or hyperplastic dystrophy. These keratinizing squamous cell carcinomas have associated HPV changes in only 6% of cases. While retrospective studies indicate that up to 50% of vulvar carcinomas are related to hyperplastic dystrophies and lichen sclerosus, prospective studies have demonstrated only 5% of women with vulvar dystrophies develop invasive carcinoma. Cancers found in younger women tend to be multifocal with adjacent VIN and have a basaloid or warty histology. Nearly 90% are associated with HPV infection, particularly HPV 16. Thirty-eight percent of women with HPV-associated lesions in a series by Trimble and colleagues were under age 55, compared with only 17% of those with classic keratinizing squamous cancers. Other associated risk factors include immunosuppression from chronic steroid use, diabetes or HIV, smoking, and a history of other lower genital tract dysplasia or neoplasia. Presentation Most women present with pruritis and an identifiable lesion ( Fig. 51.5, Fig. 51.6). Less common symptoms include pain and bleeding. Unfortunately, there is often a delay of many years between onset of symptoms and diagnosis. This is in part because patients self-medicate with a variety of over-the-counter preparations rather than seeking care and also because physicians may not biopsy liberally. The cornerstone for diagnosis of vulvar malignancy is a low threshold for a punch biopsy as it may be extremely difficult to distinguish between dysplasia, chronic vulvar dystrophy, and carcinoma. Any patient with symptoms lasting for longer than 2 weeks deserves a thorough exam and a biopsy. In order to adequately evaluate a patient with a vulvar lesion, 5% acetic acid is applied to the vulva for 5 minutes and then the area is examined either with the naked eye or a handheld magnifying glass. The entire vulva including the hair-bearing, perianal, and periclitoral regions should be examined for suspicious ulcerations and hyperpigmented, acetowhite, or gross warty lesions. Up to 5% of patients will have multifocal disease and may require multiple punch biopsies.

FIG. 51.5. Exophytic vulvar carcinoma involving posterior fourchette.

FIG. 51.6. Well-to-moderately differentiated invasive squamous cell carcinoma of the vulva.

Staging/Spread Patterns An International Federation of Gynecologists and Obstetricians (FIGO) surgical classification system replaced clinical staging for vulvar cancer in 1989 ( Table 51.1). By relying on histopathologic classification of the lymph nodes, the new system more accurately reflects prognosis since the previous system provided an inaccurate assessment of groin node involvement in one-fourth of cases.

TABLE 51.1. Vulvar carcinoma: FIGO staging 1994

Local spread occurs to contiguous structures such as the vagina, urethra, and rectum. Metastatic spread via lymphatics is common and has been well characterized by anatomic studies. Ipsilateral lesions tend to be characterized by spread to the ipsilateral groin nodes first, followed by contralateral groin, and then pelvic nodes. Lesions that cross the midline may have lymphatic drainage to both groins. Metastasis to the contralateral groin or pelvic nodes is highly unusual in the absence of ipsilateral groin node involvement. The superficial inguinal nodes above the cribriform fascia are believed to be involved prior to the deep femoral nodes below the cribriform fascia. Risk of lymphatic spread is related to tumor size, tumor thickness, older patient age, tumor grade, presence of lymphovascular space invasion, and clinically suspicious groin nodes. Groin node metastasis was found in 19% of patients with lesions =2 cm and in 42% of patients with lesions greater than 2 cm. The Gynecologic Oncology Group (GOG) surgicopathologic staging analysis demonstrated that lymph node dissection could be safely omitted in patients with less than 1 mm of stromal invasion. Hematogenous metastasis to distant organs such as bone, liver, and lungs is rare as a primary event but may be seen in patients with recurrent disease. Treatment Historically vulvar cancer has been treated with en-bloc radical vulvectomy and bilateral inguinofemoral lymphadenectomy or “longhorn incision” ( Fig. 51.7, Fig. 51.8 and Fig. 51.9). This technique resulted in a median hospitalization of 30 days, 70% to 90% wound breakdown, and chronic debilitating lymphedema in nearly 9% of women. Significant disruption in self-image and sexual function also occurred. Attempts to decrease morbidity first led to the use of separate incisions for groin node dissection while continuing to use radical vulvectomy. In 1981, Hacker and colleagues reported a series of 100 patients who underwent radical vulvectomy and lymphadenectomy through three separate incisions. The incidence of major wound breakdown was only 14% with a mean hospitalization shortened to 19 days. This procedure lead to the shortest length of hospitalization reported at that time and produced a dramatic decrease in complications. The majority of local recurrences were salvaged with repeat excision. Overall survival was not compromised by deviating from en-bloc dissection and the authors concluded that this technique is appropriate for patients with stage I and II vulvar carcinoma.

FIG. 51.7. A “longhorn” incision. Radical excision of the vulva from genitocrural fold to genitocrural fold, deep to the inferior fascia of the urogenital diaphragm along with en-bloc lymphadenectomy.

FIG. 51.8. Surgical specimen after en-bloc radical vulvectomy and inguinofemoral lymphadenectomy via longhorn incision.

FIG. 51.9. Healed vulva after en-bloc resection via longhorn incision.

DiSaia and associates first proposed conservative vulvar surgery in 1979 as a way of preserving sexual function in young patients. Requirements for conservative surgery were an invasive cancer 1 cm or less in diameter confined to the vulva with less than a 5-mm depth of invasion. Patients were treated with radical wide excision ensuring a 3-cm margin and lymphadenectomy was performed through separate groin incisions. There were no recurrences at a mean follow-up of 32 months and sexual function was deemed preserved. The authors concluded that there existed a subset of patients with vulvar cancer who could be treated with less radical vulvar procedures. Berman and co-workers expanded the experience with conservative surgery in 1989 with a report of 50 patients with T1 lesions treated with radical local excision, again providing a 3-cm margin and superficial lymphadenectomy. No patient had a resection margin positive for carcinoma. Median hospital stay was shortened to 7 days and only 12% required wound debridement compared to 50% or more with historical controls. There were five recurrences (10%), four of them local. All local recurrences were salvaged with a second local resection. In 1990, Burke and colleagues provided the first report of a conservative vulvar surgery in patients with both T1 and T2 lesions. Thirty-two patients (15 with T2 lesions) underwent radical wide excision removing a 1- to 2-cm margin of normal tissue in addition to selective inguinal dissection. The mean lesion diameter was 23 mm with a mean depth of invasion of 4.1 mm. No patients had invasive cancer at the surgical margins but 19% were positive for VIN. The authors reported only 15.5% wound separation with a mean hospital stay of 10 days. Three patients (10%) developed local recurrence and two were salvaged with repeat excision. The authors concluded that radical wide excision appears to be an acceptable surgical option for patients with resectable vulvar carcinomas. As these retrospective trials used differing criteria to identify patients suitable for less radical approaches, the GOG conducted a prospective trial beginning in 1983 which evaluated modified radical hemivulvectomy and ipsilateral superficial inguinal lymphadenectomy in 155 patients with clinical stage I vulva carcinoma limited to a depth of invasion of less than 5 mm. A 2-cm margin of normal skin around the lesion was excised. There were 19 recurrences and 7 deaths among 122 patients available for evaluation. Recurrences were evenly distributed between local and distant failures. Acute and long-term morbidity was decreased significantly compared to historical controls, but at the cost of increased risk of local recurrence. The authors concluded that although there was no consensus regarding the characteristics that would make a patient with early vulvar carcinoma a candidate for a limited surgical approach, this approach appeared to be an alternative to traditional radical surgery for a highly selected group of patients with stage I vulvar carcinoma ( Fig. 51.10, Fig. 51.11).

FIG. 51.10. Modified radical vulvectomy. Radical excision of the vulva deep to the inferior fascia of the urogenital diaphragm with a 2-cm margin around the lesion.

FIG. 51.11. Surgical specimen after modified radical vulvectomy. Lymphadenectomy is performed through separate incisions.

A need for establishing a clear margin for treatment of patients with vulvar cancer may be more important than the particular treatments. Heaps and colleagues reviewed surgicopathologic factors predictive of local recurrence in 135 patients treated with radical vulvectomy or radical local excision. Twenty-one developed local recurrence after radical resection. No patient with a surgical tumor-free margin = 8 mm suffered from local recurrence but 48% with less than an 8-mm margin recurred locally. As an 8-mm margin on fixed tissue corresponds to 10-mm margin with fresh tissue, the authors concluded that use of a 1-cm margin should successfully prevent local recurrence and reduce the currently used standard margin by over 50%. This would allow further modification of the procedure in order to decrease morbidity and avoid disfigurement and loss of organ function without sacrificing survival. Other modifications in the surgical management of vulvar carcinoma over the last 30 years include the elimination of routine pelvic lymphadenectomy and elimination of contralateral groin node dissection in patients with lateral T1 lesions and negative ipsilateral nodes. Overall survival in vulvar carcinoma is related to groin node status and lesion diameter with nodal status being the most important factor. Five-year survival is 98% for patients with negative nodes and a T1 lesion. Patients with one to two positive ipsilateral nodes have over 70% to 80% 5-year survival while those with three or more positive nodes or bilaterally positive nodes have survival in the range of 12% to 36% ( Table 51.2). Postoperative adjuvant groin radiotherapy is indicated in patients with more than one positive node or clinically evident groin nodes.

TABLE 51.2. Survival by groin node status and lesion size in vulvar carcinoma

Radiotherapy for vulvar carcinomas is indicated in patients with unresectable vulvar tumors or unresectable groin nodes. Boronow popularized treatment of locally advanced tumors with radiotherapy followed by resection of the primary tumor in order to preserve urethral, bladder, or rectal function and to avoid the morbidity of pelvic exenteration. In recent years, concurrent radiotherapy and chemotherapy has been shown to improve both local control and overall survival. The GOG has demonstrated high rates of resectability and local control of the lymph nodes in patients with clinically suspicious, fixed, or ulcerated nodes treated preoperatively with radiotherapy, concurrent cisplatin/5-fluorouracil (5-FU) chemotherapy, and followed by tailored surgery. New approaches to decrease morbidity of groin node dissection include the use of lymphatic mapping and sentinel lymph node biopsy. Intraoperative mapping with vital dye with or without lymphoscintigraphy may identify 56% to 100% of sentinel nodes. While this technique is widely used for breast cancer and melanoma, it is being prospectively evaluated by the GOG and other investigators. If the sentinel node can predict lymph node involvement with sufficiently high negative predictive value, then more extensive lymphadenectomy may be avoided. This may decrease one of the major morbidities of vulvar cancer surgery. Recurrence after definitive therapy for vulvar carcinoma may occur locally in the vulva or in the groin. The risk of local recurrence depends on prior surgical technique with higher rates of recurrence seen with less aggressive vulvar approaches. Vulvar recurrences may be characterized as new primaries due to field effect or recurrence at the prior site of excision. They are usually salvageable with radical excision though patients who have had prior radiotherapy to the groin will require some type of reconstructive flap to ensure adequate vascular supply and healing. Groin recurrences are much more difficult to treat. If the patient has not been irradiated, surgical excision of the nodes with postoperative radiotherapy is the treatment of choice. If radiotherapy has already been administered, no satisfactory therapy exists and survival is dismal. Other Histologic Subtypes Rare histologic types of vulvar carcinoma exist, representing about 10% of vulvar malignancies. These include melanoma, invasive Paget disease, basal cell carcinoma, verrucous carcinoma, Merkel cell carcinoma, adenocarcinoma, adenosquamous carcinoma, transitional cell carcinoma, sarcomas, and metastatic disease from other sites. Melanoma Melanoma is the second most common vulvar malignancy, accounting for about 6% of vulvar lesions and 1.3% of all melanomas among women. It is highly aggressive and characterized by both local recurrence and distant metastasis through hematogenous spread. The median age at presentation is 66 years, several decades older than for cutaneous melanomas. It is more common in white women with a relative risk of 2.6 compared to other races. Overall 5-year survival ranges from 35% to 50%. Patients frequently present with pruritis and visible lesions but many may be asymptomatic. The differential diagnosis of a pigmented lesion includes melanoma, lentigo, vulvar melanosis, squamous dysplasia, hemangioma, Paget disease, various nevi, and acanthosis nigricans. Biopsy should be considered of any pigmented lesion, particularly if the borders are indistinct r spreading or the lesion is raised. Excisional biopsy is preferred to better delineate depth of invasion. A punch biopsy of the most nodular area is the next best option. Three histologic subtypes of vulvar melanoma have been described. The most common is superficial spreading melanoma followed by nodular and acral lentiginous melanoma. Immunohistochemical testing may be necessary to distinguish between melanoma and Paget disease. Paget disease is typically positive for carcinoembryonic antigen (CEA) and melanoma is positive for S-100. Unlike squamous lesions, vulvar melanoma is staged using classification systems for cutaneous melanomas. The Clark system is based on level of invasion, while the Breslow system is based on vertical thickness of the lesion from the surface of intact epithelium to the deepest point of invasion. Other staging systems include Chung's modification of the Clark's system and the American Joint Committee on Cancer (AJCC) staging system ( Table 51.3). While the Breslow and Chung microstaging systems are more accurate than the Clark system, prospective evaluation by the GOG found that the AJCC staging system is most predictive of survival ( Table 51.4).

TABLE 51.3. Melanoma staging: Breslow, Clark, and Chung microstaging levels

TABLE 51.4. American Joint Committee on Cancer staging for vulvar melanoma As with squamous lesions, more conservative vulvar surgery is now being advocated for melanoma. Excision should include a 1- to 2-cm lateral margin and at least a one centimeter deep margin. The presence of capillary lymphatic space involvement and central tumor location has been correlated with positive groin node status. The role of lymphadenectomy in vulvar melanoma remains unresolved. Prospective evaluations have failed to show a survival benefit with lymphadenectomy. Adjuvant therapy for advanced disease includes the use of chemotherapy, immune modulators, and tumor vaccines. Given the small number of patients, no trials specific to vulvar melanoma have evaluated the role of adjuvant therapy. Paget Disease Paget disease accounts for 1% to 2% of vulvar malignancies. It occurs most frequently in postmenopausal white women and is characterized by red eczematous lesions with a superficial white coating and intense pruritis. As with other vulvar malignancies, there may be a long delay between initial symptoms and diagnosis. Histologic disease may spread well beyond the visible lesion. Biopsy reveals characteristic large eosinophilic cells in the basal layer of the epithelium ( Fig. 51.12). Paget disease of the vulva may exist as four clinical entities, the first of which is noninvasive or intraepithelial Paget disease. This represents 60% of cases and is cured with local excision. In invasive Paget disease, the Paget cells penetrate the basement membrane and invade the dermis. Intraepithelial Paget disease may also be associated with an underlying adenocarcinoma of sweat gland origin or a coexistent cancer. Underlying malignancy is seen in approximately 20% to 30% of patients, a rate lower than that seen with mammary Paget disease. Parker and co-workers reported poor survival in patients with invasive Paget disease and in patients with underlying malignancy. Patients with clitoral disease also had a poorer prognosis. Therapy for Paget disease involves wide excision. A skinning vulvectomy with split thickness skin graft is often needed to remove large areas of involved skin. Recurrences tend to be local and range from 7% to 58% with an average of about 30%. Some authors recommend sending margins for frozen section to ensure complete resection while others have not demonstrated a benefit to this approach with respect to recurrence. Radical surgery is reserved for patients with underlying malignancy.

FIG. 51.12. Paget disease of the vulva. Large eosinophilic cells with large nuclei and prominent cytoplasm in basal cells represent Paget cells. Basal Cell Carcinoma Basal cell carcinoma accounts for 2% to 4% of vulvar malignancies. It is typically seen in older women and tends to be well circumscribed, firm, and typically less than 2 cm in diameter. Metastasis is extremely rare and wide local excision is the preferred therapy. Verrucous Carcinoma Verrucous carcinoma is a variant of squamous cell carcinoma with distinctive pathologic and clinical characteristics. It is characterized by a large condylomatous or cauliflower-like lesion with minimal cellular atypia, a pushing border, and rare metastasis. Unless biopsies are deep enough, it may be confused with benign condyloma or papilloma. Wide local excision is the preferred therapy. Radiotherapy is contraindicated given the potential for anaplastic transformation. Merkel Cell Carcinoma Merkel cell tumors are neuroendocrine tumors of the skin and resemble small cell carcinomas. They are associated with lymphatic and distant metastases and have a very poor prognosis. Adenocarcinoma Adenocarcinoma of the vulva tends to arise in the Bartholin gland, although it may also arise from the sweat glands or Skene glands of the vulva ( Fig. 51.13). Adenoid cystic carcinoma also arises from the Bartholin gland and is characterized by slow growth and a tendency for local and perineural invasion ( Fig. 51.14). Distant metastasis is most commonly seen in the lung. Therapy involves radical excision, ipsilateral lymphadenectomy, and postoperative radiotherapy.

FIG. 51.13. Moderately differentiated adenocarcinoma arising from vulva.

FIG. 51.14. Adenoid cystic carcinoma arising from Bartholin gland. Adenosquamous Carcinoma Adenosquamous carcinoma is a rare cause of Bartholin gland carcinoma and is composed of both malignant squamous and glandular components. Transitional Cell Carcinoma Transitional cell carcinoma may arise within the Bartholin gland but is more commonly seen as metastatic spread from a urethral or bladder carcinoma. Sarcoma Sarcomas of the vulva are extremely rare, representing only 1% to 3% of vulvar malignancies. While leiomyosarcoma is the most common histologic subtype, malignant fibrous histiocytoma, and various other sarcomas have been reported. Radical surgery is the primary therapy, although the behavior of these neoplasms is not well understood. Metastatic Disease Metastatic disease to the vulva usually occurs from other lower genital tract malignancies, most commonly cervix. Rare causes include malignancies arising from the vagina, endometrium, ovary, breast, kidney, stomach, or lung.

VAGINAL CARCINOMA Primary vaginal cancer is the fifth most common gynecologic malignancy, with an expected 2,000 new cases in the United States in 2001. Only fallopian tube carcinomas are a less common gynecologic malignancy than vaginal carcinoma. The vast majority of vaginal neoplasms are metastatic from other sites, particularly direct extension from the cervix or endometrium. Lymphatic or hematogenous spread to the vagina is also possible. Primary vaginal neoplasms are typically squamous in origin and occur in postmenopausal women. Risk factors for vaginal dysplasia or carcinoma include a history of prior lower genital tract dysplasia or carcinoma, tobacco use, low socioeconomic status, history of genital warts or HPV infection, prior abnormal Pap smear, prior radiotherapy of the genital tract, immunosuppression, vaginal discharge or irritation, and early hysterectomy. History of diethylstilbestrol (DES) exposure is associated with development of clear cell adenocarcinoma of the vagina. Vaginal Intraepithelial Neoplasia The vagina is the least common site for lower genital tract dysplasia, occurring 100 times less frequently than cervical dysplasia. A review by Dodge and associates of 121 patients with biopsy-proven vaginal intraepithelial neoplasia (VAIN) showed that 65% of women with a uterus in place had associated cervical intraepithelial neoplasia and 10% had associated VIN. The upper third of the vagina is the most common site of involvement and over 60% of these women will have multifocal disease. The mean age for patients with VAIN III is approximately 40 years. Diagnosis is through careful colposcopic inspection after application of 5% acetic acid and directed biopsy ( Fig. 51.15, Fig. 51.16). Due to the rugated nature of the vagina, colposcopic evaluation may be a challenge. It is important to remember to colposcopically evaluate the vagina in patients with cytologic abnormalities, particularly those with normal appearing cervices. Acetowhite epithelium is the most common colposcopic feature. Application of half-strength Lugol solution allows for identification of nonstaining areas, which may represent dysplasia. Most patients with vaginal dysplasia are completely asymptomatic, fewer than 5% will present with bleeding or abnormal vaginal discharge. Postmenopausal patients with atrophy may benefit from a course of vaginal estrogen prior to undergoing colposcopy and directed biopsies.

FIG. 51.15. Vaginal intraepithelial neoplasia III after application of acetic acid. Note acetowhite lesions. See color figure 51.15.

FIG. 51.16. Vaginal intraepithelial neoplasia III after application of Lugol solution. Note nonstaining lesions.

Various treatment options have been employed for therapy of VAIN. These include laser ablation, partial or total vaginectomy, topical 5% 5-FU cream, cavitron ultrasonic aspiration, and loop electrosurgical excision procedure (LEEP). Choice of therapy for VAIN depends on the number and severity of lesions and their location, whether the patient is sexually active, and prior radiotherapy, as well as physician and patient preference. Hoffman and associates found invasive carcinoma in 28% of patients undergoing upper vaginectomy for VAIN III. Overall recurrence rates after therapy range from 10% to 42%. While partial vaginectomy is well suited to unifocal lesions and those involving the upper aspect of the vagina, it may lead to vaginal shortening and sexual dysfunction. Total vaginectomy with split thickness skin graft is reserved for patients who have failed more conservative approaches. Laser ablation to a depth of 2 to 3 mm is associated with minimal blood loss and is well suited for multifocal lesions. The challenge with laser therapy is ensuring that no lesions have been missed among the vaginal rugae. While 5-FU allows patients to treat themselves medically, it may be associated with significant skin desquamation, chronic ulceration, and be poorly tolerated. Dodge and colleagues found that recurrence is highest after laser ablation (38%) and 5-FU (59%) as well among those patients with multifocal disease (45%). Recurrence is lowest after partial vaginectomy (0%). Overall 2% to 12% of patients progress to vaginal cancer after therapy for vaginal dysplasia so careful cytologic and colposcopic surveillance is critical. Squamous Carcinoma Squamous carcinoma accounts for 80% to 90% of vaginal carcinomas with a mean age at presentation of 60 to 65 years. As with vaginal dysplasia, the most common location will be the posterior upper third of the vagina but the entire vaginal cylinder is at risk for malignant transformation. Abnormal vaginal bleeding occurs in over half of patients. Less common symptoms include abnormal vaginal discharge, pain, and dysuria ( Fig. 51.17).

FIG. 51.17. Vaginal carcinoma located in upper third of the vagina.

Vaginal carcinoma may spread by local extension, lymphatic spread to pelvic or inguinal nodes, and hematogenously. Lesions involving the upper two-thirds of the vagina tend to spread to pelvic nodes while lesions involving the lower third of the vagina tend to spread to inguinal nodes as would be seen with vulvar malignancies. To be considered primary vaginal carcinoma, a patient must have no evidence of disease in either the vulva or cervix. Staging of vaginal cancers is clinical and based on either the FIGO or AJCC systems ( Table 51.5).

TABLE 51.5. Vaginal carcinoma: FIGO and AJCC clinical staging

Therapy for vaginal carcinoma typically involves both external irradiation and brachytherapy. Surgical resection is typically reserved for small stage I tumors confined to the posterior aspect of the vaginal apex. Complications from therapy include fistula formation and stenosis. Overall prognosis is poorer than for cancer of the cervix. Recurrent disease tends to occur locally within 2 years of primary therapy. Extenterative procedures are required to adequately treat locally recurrent or persistent disease but are associated with considerable morbidity. Chemotherapy does not yet have a significant role in treatment of vaginal carcinoma. Clear Cell Adenocarcinoma Clear cell adenocarcinoma of the vagina is a rare malignancy most commonly seen in young women whose mothers received DES during pregnancy for prevention of miscarriage ( Fig. 51.18). This association was first noted in 1970, leading to removal of the drug from the U.S. market in 1971. Approximately 60% of patients with clear cell carcinoma are known to have been exposed to DES or other synthetic estrogens. The risk of developing clear cell adenocarcinoma of the cervix or vagina has been estimated at 1 per 1,000 to 1 per 1,500, with DES exposure in the first trimester conveying the greatest risk. The median age at diagnosis for DES-exposed patients is 19 years. DES-exposed daughters do not appear to have an increased risk of developing any other malignancies but do have an increased risk of cervical dysplasia. This may be due to increased surveillance or congenital abnormalities in the lower genital tract. No increased risk in squamous cancers of the cervix or vagina has been reported in DES-exposed women.

FIG. 51.18. Clear cell adenocarcinoma of the vagina associated with diethylstilbestrol exposure.

Most patients present with early stage disease, located in the exocervix or upper third of the vagina. Therapy is typically surgical and involves radical hysterectomy with removal of the upper vagina and pelvic lymphadenectomy. While this results in sterility, it may allow ovarian preservation and improved vaginal function over treatment with radiotherapy. Overall 5-year survival for stage I disease is more than 90% and approximately 80% for patients with stage II disease. Favorable factors in survival include early stage disease, older age, a tubulocystic histology, superficial invasion, small tumor diameter, and negative nodal status. Adenocarcinoma unrelated to DES exposure is a rare entity seen in postmenopausal women. Melanoma Melanoma is the second most common cancer of the vagina, accounting for less than 5% of vaginal malignancies and 0.3% of all melanomas among women. Nodular lesions are the most frequent histologic subtype. While vulvar melanoma is more common among white women, no association with race is seen for vaginal melanoma. It is a disease of older women and overall 5-year survival is poor at only 19%. Sarcoma Botryoides Sarcoma botryoides, or embryonal rhabdomyosarcoma, is a rare sarcoma found in young girls. It presents with vaginal bleeding and a polypoid “grapelike” mass. Historically this was treated with radical surgery, but now multimodality chemotherapy with VAC (vincristine, dactinomycin, and cyclophosphamide) and limited surgery can allow for a cure while preserving reproductive function. Radiotherapy may also be used in conjunction with multi-agent chemotherapy. Endodermal Sinus Tumor This is an extremely rare germ cell tumor arising in the vagina. It is seen in infants and is treated with combination chemotherapy as is used for germ cell tumors of the

ovary. Lymphoma Vaginal involvement with lymphoma may be secondary to systemic disease or may be localized to the vagina. This rare entity presents with bleeding and a vaginal mass. Primary vaginal lymphomas are typically diffuse, large, B-cell type. Metastatic Disease Although primary vaginal malignancy is quite rare, metastatic spread to the vagina is common and occurs through direct extension of vulvar, cervical, endometrial, and rectal malignancies. A thorough search for another primary site should be undertaken before concluding that a patient has a primary vaginal malignancy.

SUMMARY POINTS The incidence of vulvar intraepithelial neoplasia (VIN) is increasing with younger patients more likely to have human papillomavirus (HPV)-related lesions. The underlying risk for cancer in a field of VINIII is 7% to 22%. Vulvar carcinoma in older women tends to be unifocal and is associated with chronic vulvar irritation rather than HPV changes. Carcinomas in younger women tend to be multifocal and HPV changes are seen in 90% of cases (HPV 16 is most common). The risk of groin node metastasis with vulvar carcinoma increases with increasing tumor size, depth of invasion, and presence of lymphovascular space involvement. Modifications in surgery to reduce the radical nature of vulvar and groin surgery are possible in select patients. Vulvar resection margins should be at least 1 cm. The use of sentinel node biopsy to modify groin node dissections remains investigational. Squamous cell carcinoma of the vagina is a rare entity best treated with combined external beam radiation therapy and brachytherapy. SUGGESTED READINGS Vulvar Intraepithelial Neoplasia Joura EA, Lösch A, Haider-Angeler, MG, et al. Trends in vulvar neoplasia: increasing incidence of vulvar intraepithelial neoplasia and squamous cell carcinoma of the vulva in young women. J Reprod Med 2000;45:613–615. Küppers V, Stiller M, Somville T, et al. Risk factors for recurrent VIN, role of multifocality and grade of disease. J Reprod Med 1997;42:140–144. Sturgeon SR, Brinton LA, Devesa SS, et al. In situ and invasive vulvar cancer trends (1973 to 1987). Am J Obstet Gynecol 1992;166:1482–1485.

Vulvar Carcinoma: Etiology, Presentation, and Staging Abramov Y, Elchalal U, Abramov D, et al. Surgical treatment of lichen sclerosus: a review. Obstet Gynecol Surv 1996;51:193–199. Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer statistics, 2001. CA Cancer J Clin 2001;51:15–36. Homesley H, Bundy B, Sedlis A, et al. Prognostic factors for groin node metastasis in squamous cell carcinoma of the vulva (a gynecologic oncology group study). Gynecol Oncol 1993;49:279–283. Trimble CL, Hildesheim A, Brinton L, et al. Heterogeneous etiology of squamous carcinoma of the vulva. Obstet Gynecol 1996;87:59–64.

Treatment American College of Obstetricians and Gynecologists. Vulvar cancer. ACOG Technical Bulletin No. 186, Nov 1993. Berman M, Super J, Creasman W, et al. Conservative surgical management of superficially invasive stage I vulvar carcinoma. Gynecol Oncol 1989;35:352–357. Boronow RC. Combined therapy as an alternative to exenteration for locally advanced vulvo-vaginal cancer: rationale and results. Cancer 1982;49:1085–1091. Burke T, Stringer C, Gershenson D, et al. Radical wide excision and selective inguinal node dissection for squamous cell carcinoma of the vulva. Gynecol Oncol 1990;38:328–332. Cavanagh D, Fiorica J, Hoffman M, et al. Invasive carcinoma of the vulva, changing trends in surgical management. Am J Obstet Gynecol 1990;163:1007–1015. DiSaia P, Creasman W, Rich W. An alternate approach to early cancer of the vulva. Am J Obstet Gynecol 1979;133:825–830. Hacker N, Leuchter R, Berek J, et al. Radical vulvectomy and bilateral inguinal lymphadenectomy through separate groin incisions. Obstet Gynecol 1981;58:574–579. Heaps J, Fu Y, Montz F, et al. Surgical-pathologic variables predictive of local recurrence in squamous cell carcinoma of the vulva. Gynecol Oncol 1990;38:309–314. Homesley HD, Bundy BN, Sedlis A, et al. Assessment of current International Federation of Gynecology and Obstetrics staging of vulvar carcinoma relative to prognostic factors for survival (A Gynecologic Oncology Group study). Am J Obstet Gynecol 1991;164:997–1003. Levenback C, Coleman RL, Burke TW, et al. Intraoperative lymphatic mapping and sentinel node identification with blue dye in patients with vulvar cancer. Gynecol Oncol 2001;83:276–281. Montana GS, Thomas GM, Moore DH, et al. Preoperative chemo-radiation for carcinoma of the vulva with N2/N3 nodes: a Gynecologic Oncology Group study. Int J Radiat Oncol Biol Phys 2000;48:1007–1013. Stehman F, Bundy B, Dvoretsky, et al. Early stage I carcinoma of the vulva treated with ipsilateral superficial inguinal lymphadenectomy and modified radical hemivulvectomy: a prospective study of the gynecologic oncology group. Obstet Gynecol 1992;79:490–497.

VULVA MELANOMA Phillips GL, Bundy BN, Okagaki T, et al. Malignant melanoma of the vulva treated by radical hemivulvectomy: a prospective study of the Gynecologic Oncology Group. Cancer 1994;73:2626–2632. Weinstock MA. Malignant melanoma of the vulva and vagina in the United States: patterns of incidence and population-based estimated of survival. Am J Obstet Gynecol 1994;171:1225–1230.

Paget Disease Parker LP, Parker JR, Bodurka-Bevers D, et al. Paget's disease of the vulva: pathology, pattern of involvement, and prognosis. Gynecol Oncol 2000;77:183–189.

Vaginal Intraepithelial Neoplasia Dodge JA, Eltabbakh GH, Mount SL, et al. Clinical features and risk of recurrence among patients with vaginal intraepithelial neoplasia. Gynecol Oncol 2001;83:363–369.

Vaginal Cancer Hatch EE, Palmer JR, Titus-Ernstoff L, et al. Cancer risk in women exposed to diethylstilbestrol in utero. JAMA 1998;280:630–634. Hoffman MS, DeCesare SL, Roberts WS, et al. Upper vaginectomy for in situ and occult, superficially invasive carcinoma of the vagina. Am J Obstet Gynecol 1992;166:30–33.

Chapter 52 Cervical Cancer Danforth’s Obstetrics and Gynecology

Chapter 52 Robert E. Bristow

Cervical Cancer

RISK FACTORS Screening Race Sexual and Reproductive Factors Smoking Contraceptive Use Immunosuppression Human Papillomavirus Infection CLINICAL FEATURES Presenting Symptoms Physical Findings Spread of Disease DIAGNOSIS AND STAGING Diagnosis and Evaluation of Disease Extent Surgical Staging PROGNOSTIC VARIABLES Tumor Characteristics Host Factors TREATMENT MODALITIES Primary Surgery Primary Radiation Therapy Chemotherapy GENERAL MANAGEMENT BY STAGE Stage IA1 Stage IA2 Stages IB1, IB2, IIA Stages IIB, III, IVA, and IVB Post-treatment Surveillance TREATMENT OF RECURRENT CERVICAL CANCER General Considerations Surgical Treatment of Recurrent Cervical Cancer GLANDULAR LESIONS OF THE CERVIX Adenocarcinoma In Situ Cervical Adenocarcinoma SMALL CELL CARCINOMA CARCINOMA OF THE CERVICAL STUMP INCIDENTAL CERVICAL CANCER FOUND AT SIMPLE HYSTERECTOMY CERVICAL CANCER IN PREGNANCY SUMMARY POINTS SUGGESTED READINGS Risk Factors Clinical Features Diagnosis and Staging Prognostic Variables Treatment Modalities General Management by Stage Glandular Lesions of the Cervix Cervical Cancer in Pregnancy

In the United States, cancer of the uterine cervix is the sixth most common solid cancer in women after carcinoma of the breast, lung, colorectum, endometrium, and ovary. The American Cancer Society has estimated that in 2002 there will be 13,000 new cases of invasive carcinoma of the cervix in the United States, over 50,000 cases of carcinoma in situ, and 4,100 deaths from the disease. The average age of diagnosis for cervical cancer is 52 years, and the distribution of cases is bimodal with peaks at 35 to 39 years and 60 to 64 years. Worldwide, cervical cancer continues to be a leading cause of death due to cancer among women, with approximately 500,000 cases occurring annually. Lifetime risks for cervical cancer show significant geographic variation, ranging from 0.4% in Israel to 5.3% in Cali, Colombia, where cervical cancer is the most common malignancy in women.

RISK FACTORS Screening Cytologic evaluation of cells obtained from the cervix and vagina was first proposed by Papanicolaou and Traut in the 1940s as a method for detecting cervical cancer and its precursors. Since that time, cervical cytology has proved to be the most efficacious and cost-effective method for cancer screening. By increasing detection of preinvasive and early invasive disease, cervical cancer screening with the Pap smear has decreased both the incidence and mortality from cervical cancer in communities with active screening programs. A single negative Pap smear may decrease the risk for cervical cancer by 45%, and nine negative smears during a lifetime decreases the risk by as much as 99%. Eddy, using a mathematical model, indicated that in women 35 to 64 years of age, screening intervals of 10, 5, and 3 years would reduce the incidence of invasive cervical cancer by 64%, 84%, and 91%, respectively. Despite the recognized benefits of cytologic screening, substantial subgroups of women in the United States have not been screened or are not screened at regular intervals. One-half of women with newly diagnosed invasive cervical carcinoma have never had a Pap smear, and another 10% have not had a Pap smear in the 5 years preceding diagnosis. The absence of prior regular Pap smear screening is associated with a two- to six-fold increase in the risk of developing cervical cancer. Unscreened populations include older women, the uninsured, ethnic minorities, and women of lower socioeconomic status, particularly those in rural areas. Women over age 65 years should continue to be screened, as 25% of all cases of cervical cancer and 41% of deaths from the disease occur in women in this age group. Race Although the incidence of cervical cancer in the United States has declined significantly over the past 50 years, the rates among African Americans remain about twice as high as those among whites. The incidence is also approximately two times higher for Hispanic Americans and even higher for Native Americans, while most Asian-American groups experience rates similar to whites. These differences are at least partially accounted for by the strong inverse association between cervical cancer incidence and socioeconomic factors. When socioeconomic differences are controlled for, the excess risk of cervical cancer among African Americans is substantially reduced, from over 70% to less than 30%. Racial differences are also apparent in survival, with 59% of African Americans with cervical cancer surviving 5 years, compared to 67% of whites with the disease. Sexual and Reproductive Factors First intercourse before 16 years of age is associated with a two-fold increased risk of cervical cancer compared with that for women whose first intercourse occurred after age 20 years. Cervical cancer risk is also directly proportional to the number of lifetime sexual partners. Although difficult to separate epidemiologically, there is evidence to indicate that both early age at first coitus and the number of lifetime sexual partners have independent effects on cervical cancer risk. Increasing parity also appears to be a separate risk factor for cervical cancer, even after controlling for socioeconomic and reproductive characteristics. There is little evidence to support an association between age of menarche, age at menopause, or character of menses with carcinoma of the cervix.

Smoking Cigarette smoking has emerged as an important etiologic factor in squamous cell carcinoma (SCC) of the cervix. The increased risk for smokers is approximately two-fold, with the highest risk observed for long-term or high-intensity smokers. Proposed mechanisms include genotoxic or immunosuppressive effects of smoke-derived nicotine and cotinine, which can be detected in high levels in the cervical mucus of smokers. Contraceptive Use Numerous confounding factors—time interval since last cervical smear, sexual behavior, and role of the male partner, among others—complicate the interpretation of the data on oral contraceptive use and cervical cancer. After controlling for these and other variables, it appears that long-term oral contraceptive users (5 years or more) have about a two-fold increased risk of cervical cancer, compared to nonusers. Use of barrier methods of contraception, especially those that combine both mechanical and chemical protection, have been shown to lower the risk of cervical cancer, presumably because of reduced exposure to infectious agents. Immunosuppression Cell-mediated immunity appears to be a factor in the development of cervical cancer. Immunocompromised women (e.g., from renal transplantation or human immunodeficiency virus [HIV] infection) may not only be at higher risk for the disease but also demonstrate more rapid progression from preinvasive to invasive lesions and an accelerated course once invasive disease has been diagnosed. Maiman and colleagues report that HIV-positive women with cervical cancer may have a higher recurrence risk and cancer-related death rate compared to HIV-negative control subjects. Human Papillomavirus Infection During the past decade, epidemiologic evidence has accumulated, implicating infection with human papillomavirus (HPV) as a likely etiologic agent in cervical SCC. All known types of HPV have a similar structural and genomic organization. They are non-enveloped virions with a double-stranded circular DNA genome of 7,800 to 7,900 base pairs and an icosahedral capsid. HPV DNA is present in virtually all cases (93%) of cervical cancer and its precursor lesions. While HPV infection is thought to be a component of neoplastic transformation, it is unlikely to be entirely sufficient in and of itself. HPV colonizes mucosal or cutaneous epithelium and may induce hyperproliferation, resulting in the formation of warts at the site of infection. Based on differences in DNA sequencing, over 70 different types of HPV have been identified, 23 of which are known to infect the anogenital tract. Low oncogenic risk–type viruses include types 6, 11, 42, 43, and 44 and are associated with condyloma acuminatum and some cases of low-grade squamous intraepithelial lesions, but rarely with invasive cancer. High oncogenic risk–type viruses include types 16, 18, 31, 45, and 56 and are commonly detected in women with high-grade squamous intraepithelial lesions (HGSIL) and invasive cancer. HPV types 33, 35, 39, 51, and 52 can be considered as being of intermediate oncogenic risk, as they are associated with HGSIL but are uncommonly detected in invasive carcinomas. Following acute HPV infection, three clinical sequelae are possible: 1. Latent viral infection occurs when the HPV genome becomes stabilized as a nonintegrated episome and remains in the host cell without clinical or morphologic changes in the squamous epithelium. Clinically, patients may display no gross or morphologic evidence of infection but still harbor virus as demonstrated by DNA detection techniques. 2. Active infection, manifested by proliferation of squamous epithelium into benign tumors (warts), is present when HPV undergoes vegetative replication. 3. Highly oncogenic HPV viruses associated with high-grade lesions can become integrated into the host genome, interrupting control of proliferation by certain oncoproteins. Initiation of cervical dysplasia and carcinoma may involve interactions between HPV and specific genes that regulate cell growth. The E6 and E7 open reading frames of the HPV genome are particularly important in the immortalization and transformation of infected cells. In HPV types 16 and 18, the protein products synthesized from the E6 and E7 open reading frames can bind to the gene products of the p53 and retinoblastoma ( Rb) tumor suppressor genes, respectively. Viral integration results in the overexpression of the E6 and E7 viral protein products with increased binding and inactivation of their respective tumor suppressor proteins. Removal of these two inhibitory influences on cellular proliferation is thought to provide HPV-infected cells with a growth advantage, ultimately leading to neoplastic transformation.

CLINICAL FEATURES Presenting Symptoms The most common symptom of cervical cancer is abnormal vaginal bleeding or discharge. Abnormal bleeding may take the form of postcoital spotting, intermenstrual bleeding, menorrhagia, or postmenopausal spotting. If bleeding has been chronic, a patient may complain of fatigue or other symptoms related to anemia. Serosanguineous or yellowish vaginal discharge, frequently associated with a foul odor, may accompany an advanced or necrotic carcinoma. Pelvic pain may result from locally advanced disease or tumor necrosis. Tumor extension to the pelvic side wall may cause sciatic pain or back pain associated with urinary tract obstruction and hydronephrosis. Metastatic tumor to the iliac and paraaortic lymph nodes can extend into the lumbosacral nerve roots and present as lumbosacral back pain. Urinary or rectal symptoms (e.g., hematuria, hematochezia, fistulae) can be associated with bladder or rectal invasion by advanced-stage cervical carcinoma. Physical Findings Invasive carcinoma of the cervix displays a wide range of gross appearances. Early lesions may be focally indurated or ulcerated, or present as a slightly elevated and granular area that bleeds readily on contact ( Fig. 52.1). More advanced tumors have several types of gross appearance: exophytic, endophytic, or infiltrative. Exophytic tumors characteristically have a polypoid or papillary appearance and will often demonstrate contact bleeding. Endophytic tumors are usually ulcerated or nodular but may be clinically inapparent if located high within the endocervical canal. Such tumors frequently invade deep into the cervical stroma to produce an enlarged, hard, barrel-shaped cervix that may only be appreciated on rectovaginal pelvic examination. The infiltrative pattern of tumor growth may produce surrounding tissue necrosis and erosion of normal anatomic landmarks.

FIG. 52.1. Gross appearance of squamous cell carcinoma of the cervix. This is an ulcerative type of lesion.

Spread of Disease Parametrial Extension Cervical cancer generally follows an orderly pattern of disease progression. Initially, tumor cells usually spread through parametrial lymphatic vessels, expanding and replacing parametrial lymph nodes ( Fig. 52.2). These individual tumor masses enlarge and become confluent, eventually replacing the normal parametrial tissue. Less commonly, the central tumor mass reaches the pelvic sidewall by direct contiguous extension through the cardinal (Mackenrodt) ligament. Significant involvement of the medial portion of this ligament may result in ureteral obstruction and hydronephrosis.

FIG. 52.2. Anatomic pathways of spread in invasive cervical carcinoma. Lymph Node Involvement The pelvic lymph nodes will usually be the first site of embolic lymphatic spread of cervical cancer. The lymph node groups most commonly involved are the obturator, external iliac, and hypogastric ( Fig. 52.2). The inferior gluteal and presacral lymph nodes are less frequently involved. Secondary nodal involvement (i.e., common iliac, paraaortic) rarely occurs in the absence of pelvic nodal disease. The percentage of involved lymph nodes increases directly with primary tumor volume. Rarely, retrograde lymphatic embolization may occur to the inguinal lymph nodes. Patients with locally advanced pelvic disease may have detectable metastatic spread to the scalene nodes. Consequently, careful clinical assessment of the groin and supraclavicular fossa areas should be included as part of the physical examination. Vaginal Extension When the primary tumor has extended beyond the confines of the cervix, the upper vagina is frequently involved (50% of cases). Anterior extension through the vesicovaginal septum is most common, often obliterating the dissection plane between the bladder and underlying cervical tumor, making surgical therapy difficult or impossible. Posteriorly, a deep peritoneal cul-de-sac (pouch of Douglas) can represent an anatomic barrier to direct tumor spread from the cervix and vagina to the rectum. Bladder and Rectal Involvement In the absence of lateral parametrial disease, anterior and posterior spread of cervical cancer to the bladder and rectum is uncommon. Approximately 20% of patients with tumor extending to the pelvic sidewall will have biopsy-proven bladder invasion. Endometrial Involvement The endometrium is involved in 2% to 10% of cervical cancer cases treated with surgery, although the overall incidence (including nonsurgical cases) is unknown. Although endometrial extension does not alter a patient's stage of disease, it has been associated with decreased survival and a higher incidence of distant metastases. Ovarian Metastasis Ovarian involvement with cervical cancer is rare but, when present, most likely occurs through the lymphatic connections between the uterus and adnexal structures. In patients undergoing surgical treatment for early-stage disease, ovarian metastasis is present in less than 1% of SCCs and slightly more than 1% of adenocarcinomas. The true incidence of ovarian involvement with advanced-stage tumors is unknown, as pathologic evaluation of the adnexae is not commonly performed in such cases. Hematogenous Spread Hematogenous spread of cervical cancer is uncommon, particularly at the time of initial diagnosis. However, when blood-borne metastases do occur, the lung, liver, and bone are the sites most frequently involved. Metastasis to the bowel, adrenal gland, spleen, and brain are extremely rare.

DIAGNOSIS AND STAGING Diagnosis and Evaluation of Disease Extent Pathologic documentation of invasive disease should always be obtained prior to initiating therapy for cervical cancer. The International Federation of Gynecology and Obstetrics (FIGO) staging system used for the diagnosis and evaluation of cervical carcinoma is based on clinical evaluation (visual inspection, palpation, colposcopy, endocervical curettage, cervical biopsy, or conization), radiographic examination of the chest, kidneys, and skeleton, and ECC and biopsies. Cervical cancer is one of the few gynecologic malignancies (vaginal cancer being the other) that still uses a clinically based staging system by international convention, which facilitates comparison of results between institutions irrespective of the availability of diagnostic technological resources. Staging procedures allowed by FIGO convention are shown in Table 52.1. Lymphangiograms, arteriograms, computed tomographic (CT) scans, magnetic resonance imaging (MRI), and laparoscopy or laparotomy should not be used for clinical staging. However, information from such additional studies and procedures may be used to modify the treatment approach.

TABLE 52.1. Staging procedures for cervical cancer

In 1995, FIGO revised the clinical staging system of cervical carcinoma ( Table 52.2). Stage I disease includes those neoplasms that are clinically confined to the cervix. In the current staging classification, stage IA tumors are those that are diagnosed by microscopic evaluation of a conization specimen (microinvasive carcinoma). Stage IA1 is defined as a tumor with stromal invasion no greater than 3 mm in depth beneath the basement membrane and no wider than 7 mm. This definition reflects data indicating that patients with less than 3 mm of invasion are at very low risk of metastatic disease and may be treated more conservatively. All grossly visible lesions should technically be classified as stage IB. The 1995 staging system divides stage IB lesions into stage IB1 (no greater than 4 cm in maximal diameter) and stage IB2 (greater than 4 cm in maximal diameter), reflecting the prognostic importance of tumor volume for macroscopic lesions limited to the cervix.

TABLE 52.2. FIGO staging of cervical carcinoma

Determination of clinical stage depends on careful inspection and palpation of the cervix, vagina, and pelvis. Examination under anesthesia is recommended. It is important to palpate the entire vagina to determine whether disease is limited to the cervix (IB), extends to the upper two-thirds of the vagina (IIA), or also involves the

lower third of the vagina (IIIA) ( Fig. 52.3). Tumor extension into the parametrial tissue (IIB) or to the pelvic sidewall (IIIB) is best appreciated on rectovaginal examination. It is impossible at clinical examination to decide whether a smooth and indurated parametrium is truly cancerous or only inflammatory, and the case should be considered stage III only if the parametrium is nodular on the pelvic wall or if the growth itself extends to the pelvic wall. Biopsy-proven invasion of bladder or rectal mucosa by cystoscopy or proctoscopy is required for a diagnosis of stage IVA disease. In the United States, the distribution of patients by clinical stage is stage I, 38%; stage II, 32%; stage III, 26%; and stage IV, 4%.

FIG. 52.3. Clinical stages of carcinoma of the cervix. In stage I, only the cervix is involved. In stage II, the parametrium or upper two-thirds of the vagina is involved. In stage III, the malignancy extends to the pelvic sidewall or involves the lower third of the vagina. Stage IV reflects involvement of the bladder or rectal mucosa (stage IVA) or distant metastasis (stage IVB).

When there is doubt concerning which stage a tumor should be assigned, the earlier stage is mandatory. Once a clinical stage has been determined and treatment initiated, subsequent findings on either extended clinical staging (CT, etc.) or surgical exploration should not alter the assigned stage. Assignment of a more advanced stage during treatment will result in an apparent but deceptive improvement in the results of treatment for earlier stage disease. Surgical Staging The current FIGO staging classification for cervical cancer is based on pretreatment clinical findings. Only the subclassification of stage I (IA1, IA2) requires pathologic assessment. Discrepancies between clinical staging and surgicopathologic findings range from 17.3% to 38.5% in patients with clinical stage I disease, to 42.9% to 89.5% in patients with stage III disease. This has led some authors to emphasize surgical staging of cervical carcinoma to identify occult tumor spread and determine the presence of extrapelvic disease so that adjunctive or extended-field radiation therapy may be offered. Transperitoneal surgical staging procedures (e.g. laparotomy), when followed by abdominopelvic irradiation, are associated with appreciable complications, particularly enteric morbidity. The extraperitoneal surgical approach can be performed through a paraumbilical or paramedian incision and allows accurate assessment of pelvic and paraaortic nodal status. Using this approach, an incision in the peritoneum is avoided and adhesion formation is minimized. It is associated with few complications and does not delay institution of radiation therapy. Laparoscopic surgical staging for cervical cancer has also been advocated; however, the safety and efficacy of this approach has yet to be determined.

PROGNOSTIC VARIABLES Tumor Characteristics Clinical stage of disease at the time of presentation is the most important determinant of subsequent survival, regardless of treatment modality. Five-year survival declines as FIGO stage at diagnosis increases: stage IA, 97%; stage IB, 70% to 85%; stage II, 60% to 70%; stage III, 30% to 45%; stage IV, 12% to 18%. Prognostic variables directly related to surgicopathologic tumor characteristics and their effect on survival were compiled by Kosary for the National Cancer Institutes Surveillance, Epidemiology, and End Results (SEER) program for the period of 1973 through 1987. This study included 17,119 cases of invasive cervical cancer and found that FIGO stage, tumor histology, histologic grade, and lymph node status are all independent prognostic variables relating to survival. SCCs accounted for 74.9% of cases in the SEER database and can be categorized according to the degree of histologic tumor differentiation. Well-differentiated tumors account for about 5% of cervical SCCs and are composed of sheets and cords of cells with abundant acidophilic cytoplasm, clearly visible intercellular bridges, and often production of variable amounts of keratin. Moderately differentiated tumors are the most common variety, with 85% of SCCs falling in this category ( Fig. 52.4). These tumors are characterized microscopically by masses and cords of spindle-shaped squamous cells with elongated nuclei and scant cytoplasm and more frequent mitoses. Poorly differentiated tumors have a rapid growth rate, with numerous mitoses and cells with closely crowded nuclei and scant cytoplasm. These tumors are difficult to recognize as having originated in squamous cells and constitute approximately 10% of cervical SCCs. The degree of histologic tumor differentiation does correlate with overall survival. In the SEER database, 5-year survival for patients with well-differentiated tumors was 74.5%; for those with moderately differentiated tumors it was 63.7%, and it fell to 51.4% for those patients with poorly differentiated carcinomas. Approximately 15% to 20% of cervical cancer cases are adenocarcinomas. After controlling for known prognostic variables, the SEER study found no difference in overall survival between patients with cervical SCC and adenocarcinoma. However, adenosquamous histology was associated with decreased survival. Survival is also orrelated with depth of tumor invasion into the cervical stroma and overall tumor volume.

FIG. 52.4. Moderately differentiated squamous cell carcinoma of the cervix.

Among surgically treated patients, survival is directly related to the number and location of lymph node metastasis. The frequency of positive lymph nodes increases with the stage of disease ( Table 52.3). For all stages of disease, when both pelvic and paraaortic lymph nodes are negative, the 5-year survival rate is 75.2%. Survival decreases to 45.6% with positive pelvic nodes, and the risk of recurrence is related to the number of nodes involved. The recurrence rate is 35% with one positive pelvic lymph node, 59% with two or three positive nodes, and 69% with metastases to more than three pelvic lymph nodes. When paraaortic nodes are involved, the 5-year survival rate ranges from 15% to 45%.

TABLE 52.3. Incidence of pelvic and paraaortic lymph node metastasis by FIGO stage of cervicalcarcinoma

Host Factors Whether or not patient age at diagnosis of cervical cancer is a significant and independent predictor of clinical outcome remains controversial. Some investigators have observed decreased survival in women younger than 35 to 40 years, who have a greater frequency of poorly differentiated tumors. Others have found no significant

difference in survival between younger and older patients. Several hematologic parameters have been associated with cervical cancer survival outcome. The incidence of pretreatment anemia (hemoglobin of 12 g/dL or less) increases with advancing stage of disease, occurring in 25%, 33%, and 45% of patients with stages I, II, and III disease, respectively. Anemia is associated with a higher incidence of pelvic recurrences and decreased survival, primarily due to more frequent radiation therapy failures. Tumor hypoxia is the proposed mechanism of radio resistance in the presence of anemia. Another prognostic hematologic parameter is thrombocytosis (>400,000/mm 3 ), which has been associated with decreased survival after controlling for cell type, stage, and age. Coexistent medical conditions may also affect the success of treatment. Diabetes and hypertension are frequently associated with significant vascular disease and potentially contribute to both tumor hypoxia and decreased blood supply to normal pelvic tissues. Patients with these conditions are subject to a higher incidence of treatment complications and pelvic tumor recurrence, as well as decreased survival.

TREATMENT MODALITIES Surgery and radiation therapy are the two primary therapeutic modalities most commonly used to treat invasive cervical carcinoma. In general, primary surgical management is limited to patients with stage I and IIA disease, while radiation therapy can be applied to patients with all stages of disease. For patients with early-stage disease, multiple factors should be considered in selecting the most appropriate treatment program. Age is not a contraindication to surgical management, provided the patient does not have significant medical comorbidity. Young patients desiring ovarian preservation and sexually active patients are preferentially managed surgically. Other reasons for the selection of radical surgery over radiation include concomitant inflammatory bowel disease, previous radiation for other disease, and the presence of a coexistent adnexal neoplasm. Primary Surgery Surgery provides the opportunity to perform a thorough pelvic and abdominal exploration, which can identify patients with a disparity between the clinical and surgicopathologic stages. Such patients can then be offered an individualized treatment plan based on their precise disease status. The primary surgical management of cervical cancer generally consists of hysterectomy. For patients with stage IA1 lesions who desire fertility preservation, cervical conization with clear surgical margins is acceptable treatment. The safety and efficacy of surgery to preserve fertility (e.g., radical trachelectomy) for patients with larger stage I lesions however has yet to be fully evaluated. There are five distinct variations or types of hysterectomy used in the treatment of cervical cancer. Types of Hysterectomy Type I A type I hysterectomy refers to the standard extrafascial total abdominal hysterectomy. This procedure ensures complete removal of the cervix with minimal disruption to surrounding structures and is appropriate treatment for stage IA1 disease. Type II A type II hysterectomy is also referred to as a modified radical hysterectomy ( Fig. 52.5, Fig. 52.6 and Fig. 52.7). This procedure involves dissection of the ureters from the parametrial and paracervical tissues down to the ureterovesical junction. This permits removal of all parametrial tissue medial to the ureters, as well as the medial half of the uterosacral ligament and proximal 1 to 2 cm of vagina. This operation is usually performed in conjunction with pelvic lymphadenectomy.

FIG. 52.5. Anatomic dissection of radical hysterectomy. The cardinal ligaments are transected ( dashed line) at the level of the ureter (type II) or at the pelvic side wall (type III).

FIG. 52.6. Anatomic dissection of radical hysterectomy. The uterosacral ligaments are divided at the sacrum (type III) or midway between the sacrum and the uterus (type II).

FIG. 52.7. Anatomic dissection of radical hysterectomy. In a type II radical hysterectomy, the upper 1 to 2 cm of vagina is excised. In type III radical hysterectomy, the proximal one-third to one-half of the vagina is removed. Type III In a type III or radical abdominal hysterectomy, the ureters are completely dissected from within the paracervical tunnel, and the bladder and rectum are extensively mobilized ( Fig. 52.5, Fig. 52.6 and Fig. 52.7). Establishing the paravesical and pararectal spaces facilitates removal of all the parametrial tissue out to the pelvic sidewall, complete resection of the uterosacral ligaments, and excision of the upper one-third to one-half of the vagina. Bilateral pelvic lymphadenectomy is performed with this procedure, removing all lymph-bearing tissue between the mid-common iliac vessels distally to the circumflex iliac vessels and from within the obturator fossa ventral to the obturator nerve. Type IV/Type V A type IV or extended radical hysterectomy includes removal of the superior vesical artery, periureteral tissue, and up to three fourths of the vagina. In a type V or partial exenteration operation, the distal ureters and a portion of the bladder are resected. Type IV and type V procedures are rarely performed today because most patients with disease extensive enough to require these operations can be more adequately treated with primary radiation therapy. Complications of Radical Abdominal Hysterectomy Modern surgical techniques and anesthesia have reduced the operative mortality associated with radical hysterectomy to 0.6%. Potentially fatal pulmonary embolism occurs in 1% to 2% of patients. Urinary and bowel fistula formation and incisional complications related to surgical treatment tend to occur early in the postoperative period and are usually amenable to surgical repair. Ureterovaginal and vesicovaginal fistulae occur in 2%

and 0.9% of patients, respectively. The most commonly observed complication after radical hysterectomy is urinary dysfunction resulting from partial denervation of the detrusor muscle during excision of the paracervical and paravaginal tissue. Radical hysterectomy results in vaginal shortening; however, with sexual activity gradual lengthening will occur. Pelvic lymphocyst formation occurs in 2% to 6.7% of patients following radical hysterectomy and pelvic lymphadenectomy. The incidence is somewhat lower when the retroperitoneal spaces are left open. Most lymphocysts are asymptomatic and do not require intervention; however, lymphocysts may occasionally produce pelvic pain, ureteral obstruction, or partial venous obstruction with thrombosis. Primary Radiation Therapy Radiation therapy can be used for all stages of disease and for most patients regardless of age, body habitus, or coexistent medical conditions. The recommended nomenclature for measurement of absorbed radiation dose is the gray (Gy); 1 Gy is equal to one joule (J) of energy absorbed per kilogram of substance. Radiation dose is also commonly expressed as centi-gray (cGy), with 100 cGy equal to 1Gy. By convention, the irradiation dose used in the treatment of cervical cancer is described relative to two anatomic landmarks within the pelvis. Point A is defined as a point 2 cm above the lateral vaginal fornix and 2 cm lateral to the uterine canal corresponding to the paracervical triangle. Point B reflects the dose delivered to the pelvic sidewall and is located 3 cm lateral to point A. The technical treatment modalities used in modern radiation therapy for cervical cancer consist of a combination of external irradiation and local irradiation. External irradiation is delivered from a source remote from the body (e.g., linear accelerator, cobalt-60) and is used to treat the regional lymph nodes, decrease tumor volume, and reduce the anatomic distortion produced by larger tumor masses. External beam irradiation treatment for cervical cancer is usually delivered using a four-field technique (anterior, posterior, and lateral fields). The precise treatment volume is determined according to individual patient anatomy, but usually measures 15 cm × 15 cm to 18 cm × 18 cm. The pelvic radiation field extends 1 to 2 cm beyond the lateral borders of the bony pelvis and inferiorly beyond the border of the obturator foramen. The cephalad margin may be extended to 18 cm in length to treat the common iliac lymph nodes or even higher if paraaortic lymph node coverage is necessary. A typical external beam radiation treatment portal for cervical cancer is shown in Fig. 52.8.

FIG. 52.8. Whole pelvis radiation treatment field for cervical cancer. In this case, the lower margin of the treatment field extends below the pubic symphysis to provide coverage of the proximal vagina. Lead tapes (white stripes) are used for excluding the corners of a square field, reducing the total irradiated volume by approximately 10%.

Brachytherapy, refers to a radiation source in direct proximity to the target tissue and may be delivered using a variety of intracavitary applicator devices, but the intrauterine tandem and vaginal colpostats are used most frequently for primary treatment ( Fig. 52.9). Once adequate placement is assured, the device is after-loaded with radioactive isotope (e.g., radium-226, cesium-137, iridium-192). Alternatively, vaginal cylinders or interstitial needle implants may be used to deliver local radiation therapy, depending on patient anatomy and tumor distribution.

FIG. 52.9. Diagram of a typical intrauterine tandem and vaginal colpostat brachytherapy placement for cervical cancer showing the anatomic landmarks point A and point B.

The total dose delivered to point A is determined by the volume of disease to be treated and ranges from 6,500 to 7,000 cGy for small stage IB lesions to 8,500 to 9,000 cGy for bulky stage IIB and stage III lesions. For patients with documented paraaortic node metastasis, or those at high risk, 4,500 cGy of extended field irradiation is delivered to the paraaortic region. Cure rates with irradiation therapy are stage-dependent, with 5-year survival rates averaging 70% to 85% for stage I, 60% for stage II, 45% for stage III, and 18% to 20% for stage IV disease. Concurrent Chemotherapy and Radiation Therapy In 1999, five multi-institutional, randomized controlled trials reported a survival advantage associated with the concurrent administration of chemotherapy and radiation therapy in the management of cervical cancer. Although these trials differed in their inclusion criteria, chemotherapy schedules, and prescribed radiation treatment, all demonstrated a similar improvement in progression-free survival (10%–27%) and overall survival (10%–17%). Many centers now administer concurrent treatment with chemotherapy and radiation therapy as standard practice for patients with locally advanced cervical cancer. Cisplatin (Platinol), 5-fluorouracil (Adrucil), and hydroxyurea (Hydrea) have been the chemotherapeutic agents most extensively studied as part of a combined modality treatment program. A commonly used contemporary regimen is cisplatin 40 mg per m 2 administered intravenously on a weekly basis during the radiation treatment interval. Complications of Radiation Therapy Radiation therapy is associated with both acute and chronic complications. Perforation of the uterus may occur at the time of intracavitary insertion and, if unrecognized, may result in significant blood loss, radiation damage, and peritonitis. Appropriate management consists of removal of the implant and broad-spectrum antibiotic coverage if signs of infection are present. Vaginal fibrosis and stenosis is the most common chronic complication of radiation therapy for cervical cancer and is seen in up to 70% of cases. Ovarian function is lost in virtually all patients undergoing radiation therapy to the pelvis. Proctosigmoiditis occurs in up to 8% of patients undergoing radiation therapy for cervical cancer. Symptoms include abdominal pain, diarrhea, and nausea. An antispasmodic agent, a low-gluten and low-lactose diet, and steroid enemas may be useful; however, severe cases may require hyperalimentation and a diverting colostomy. Hemorrhagic cystitis is seen in approximately 3% of patients undergoing radiation therapy for cervical cancer. In contrast to surgical therapy, fistulous complications associated with radiation therapy tend to occur late and are more difficult to repair secondary to poorly vascularized tissues from radiation fibrosis and vasculitis. Rectovaginal and vesicovaginal fistulae each occur in approximately 1% of cervical cancer patients treated with irradiation. In such cases, biopsy specimens should be obtained from the edge of the fistula to rule out recurrent cancer. Diversion of the fecal (colostomy) or urinary (percutaneous nephrostomy) stream is usually required to allow adequate healing (3–6 months) prior to surgical repair. Two percent of patients experience small bowel obstruction as a consequence of radiation therapy; it is more common in those patients with vascular disease or a history of previous abdominal surgery. The most common site of small bowel obstruction is the terminal ileum, which is relatively fixed within the radiation field by the cecum. Complete small bowel obstruction or cases recalcitrant to conservative management require surgical intervention. Chemotherapy Chemotherapy as the sole mode of treating cervical cancer is indicated for patients with extrapelvic metastases (stage IVB) or those with recurrent disease who are not candidates for radiation therapy or exenterative surgery. Cisplatin has been the most extensively studied agent and has demonstrated the most consistent clinical

response rates. Complete clinical responses have been observed in 24% of patients, with an additional 16% demonstrating a partial response. Unfortunately, in most series, responses to cisplatin are short-lived (3–6 months). Other agents demonstrating at least partial activity against cervical cancer include carboplatin (Paraplatin), ifosfamide (Ifex), doxorubicin hydrochloride (Adriamycin), vinblastine sulfate (Velban), vincristine sulfate (Oncovin), 5-fluorouracil, methotrexate, and altretamine (Hexalen). There is little objective evidence to suggest that combination chemotherapy is superior to single-agent cisplatin treatment in improving the overall survival of patients with advanced or recurrent cervical cancer.

GENERAL MANAGEMENT BY STAGE Stage IA1 The 5-year survival rate of these patients approaches 100% with primary surgical therapy. Extrafascial hysterectomy is adequate treatment for this group of patients. Conization may be used selectively if preservation of fertility is desired, provided the surgical margins are free of disease. In the absence of lymph–vascular invasion, the incidence of pelvic lymph node metastasis is 0.3%, and lymphadenectomy is not indicated. In the presence of lymph–vascular involvement, the risk of pelvic node metastasis increases to 2.6%. Pelvic lymphadenectomy and extrafascial hysterectomy should be performed in these cases. In patients who are medically inoperable, stage IA1 carcinoma can be effectively treated with intracavitary radiation. Stage IA2 Microinvasive carcinoma with stromal invasion of 3.1 to 5.0 mm is associated with positive pelvic lymph nodes in 6.2% of patients. The preferred treatment for these lesions is modified radical (type II) hysterectomy with pelvic lymphadenectomy. Radiation therapy is equally effective from a survival standpoint but may carry a greater risk of post-treatment morbidity compared to modified radical hysterectomy. Stages IB1, IB2, IIA Both radical surgery and radiation therapy are equally effective in treating stages IB and IIA carcinoma of the cervix. Numerous uncontrolled studies support the merits of each modality, with no significant differences in pelvic tumor control or overall survival. Zander and colleagues reported on 1,092 patients with stages IB and II cervical cancer treated with radical (type III) hysterectomy and pelvic lymphadenectomy. Five-year survival rates were 84.5% for stage IB and 71.1% for stage II disease. Similar survival rates are obtained with primary radiation therapy. In one series, Perez and associates reported 5-year survival rates of 85% for 312 patients with stage IB disease and 70% for 98 patients with stage IIA disease treated with primary radiation therapy. Treatment should be individualized for patients with bulky stage I (IB2) tumors. Tumor expansion of the upper endocervix and lower uterine segment can distort cervical anatomy and lead to suboptimal placement of intracavitary radiation sources. Consequently, the central failure rate has been reported as high as 17.5% in patients with cervical lesions greater than 6 cm treated with radiation alone. In such situations, a “completion” extrafascial hysterectomy is usually performed following radiation therapy. While many clinicians limit the use of radical hysterectomy to patients with small stage IB (