The Pediatric and Adolescent Hip: Essentials and Evidence [1st ed.] 978-3-030-12002-3;978-3-030-12003-0

Table of contents :
Front Matter ....Pages i-xi
Front Matter ....Pages 1-1
The History of Pediatric Hip Surgery: The Past 100 Years (Dennis R. Wenger, James D. Bomar)....Pages 3-27
Anatomy and Physiology of the Pediatric Hip (Emily K. Schaeffer, Kishore Mulpuri)....Pages 29-51
Biomechanics of the Hip During Gait (Morgan Sangeux)....Pages 53-71
Front Matter ....Pages 73-73
Developmental Dysplasia of the Hip in Young Children (Stuart L. Weinstein, Joshua B. Holt)....Pages 75-129
Acetabular Dysplasia in the Reduced or Subluxated Hip (Jonathan G. Schoenecker, Ira Zaltz, Justin Roth, Perry L. Schoenecker)....Pages 131-165
Front Matter ....Pages 167-167
Legg-Calve-Perthes Disease (Benjamin Joseph)....Pages 169-191
Coxa Vara (Arnold Suzuki, Anthony Cooper, James Fernandes)....Pages 193-206
Slipped Capital Femoral Epiphysis (Balakumar Balasubramanian, Sattar Alshryda, Sanjeev Madan)....Pages 207-252
Femoroacetabular Impingement (Erika Daley, Ira Zaltz)....Pages 253-271
Front Matter ....Pages 273-273
Musculoskeletal Infection of the Hip (Michael Benvenuti, Megan Johnson, Jonathan G. Schoenecker)....Pages 275-309
Tuberculosis Involving the Hip (Vrisha Madhuri)....Pages 311-323
Front Matter ....Pages 325-325
Transient Synovitis (James S. Huntley)....Pages 327-346
Juvenile Idiopathic Arthritis and the Hip (James S. Huntley, Peter S. Young, Sanjeev Patil)....Pages 347-374
Idiopathic Chondrolysis of the Hip (Vrisha Madhuri, Noel Malcolm Walter, Jyoti Panwar)....Pages 375-390
Front Matter ....Pages 391-391
Pediatric Proximal Femoral Fractures (Mohamed Kenawey, Emmanouil Liodakis, Marcel Winkelmann, Christian Krettek)....Pages 393-408
Pediatric Pelvic Injuries (Mohamed Kenawey)....Pages 409-443
Traumatic Hip Dislocation in Children (Hossam Hosny, Wael Salama, Ahmed Abdelaal, Mohamed Kenawey)....Pages 445-463
Front Matter ....Pages 465-465
The Hip in Cerebral Palsy (Jason J. Howard, Abhay Khot, H. Kerr Graham)....Pages 467-530
The Hip in Myelomeningocele (Emmanouil Morakis, Jason J. Howard, James Wright)....Pages 531-551
The Hip in Spinal Muscular Atrophy (Jill E. Larson, Brian Snyder)....Pages 553-570
The Hip in Muscular Dystrophy (Deborah M. Eastwood)....Pages 571-581
The Hip in Charcot-Marie-Tooth Disease (Neil Saran)....Pages 583-598
The Hip in Poliomyelitis (Hugh G. Watts, Benjamin Joseph, Sanjeev Sabharwal)....Pages 599-616
Front Matter ....Pages 617-617
The Hip in Rett Syndrome (Deborah M. Eastwood)....Pages 619-629
Hip Problems in Children with Trisomy 21 (Matthew Lea, Sattar Alshryda, John Wedge)....Pages 631-649
Larsen Syndrome and the Hip (James S. Huntley)....Pages 651-672
The Hip in Mucopolysaccharidoses (Kevin Walker)....Pages 673-689
The Hip in Arthrogryposis (Katie Rooks, Haemish Crawford)....Pages 691-713
The Hip in Osteogenesis Imperfecta (Maegen Wallace, Paul Esposito)....Pages 715-734
Front Matter ....Pages 735-735
Osteoid Osteoma Involving the Hip (Karl Logan, Felix Brassard, Jason J. Howard, Pierre Schmit)....Pages 737-749
Osteochondroma Involving the Hip (Daniel E. Porter, Fei Li)....Pages 751-768
The Hip in Fibrous Dysplasia (Brian L. Dial, Benjamin A. Alman)....Pages 769-783
Bone Cysts Involving the Hip (Laura Deriu, Sattar Alshryda, James Wright)....Pages 785-817
Front Matter ....Pages 819-819
Bladder and Cloacal Exstrophy (Jason J. Howard, James S. Huntley, Jonathan G. Schoenecker, Sattar Alshryda, Joao Pippi Salle)....Pages 821-840
Athletic Injuries Involving the Hip (Justin Roth, Jeffrey J. Nepple)....Pages 841-853
Snapping Hip Syndrome (Ling Hong Lee, Ed Gent, Sattar Alshryda)....Pages 855-874
The Hip in Congenital Femoral Deficiency (Fergal Monsell)....Pages 875-891
Total Hip Arthroplasty for Pediatric Disorders (Stephen M. Engstrom, Gregory G. Polkowski)....Pages 893-910

Citation preview

The Pediatric and Adolescent Hip Essentials and Evidence Sattar Alshryda Jason J. Howard James S. Huntley Jonathan G. Schoenecker Editors

123

The Pediatric and Adolescent Hip

Sattar Alshryda  •  Jason J. Howard James S. Huntley Jonathan G. Schoenecker Editors

The Pediatric and Adolescent Hip Essentials and Evidence

Editors Sattar Alshryda Clinical Director of Paediatric Trauma and Orthopaedic Surgery Royal Manchester Children Hospital Manchester University NHS Foundation Trust Manchester, UK James S. Huntley Senior Attending Physician in Pediatric Orthopedic Surgery Sidra Medicine Doha, Qatar

Jason J. Howard Weill Cornell Medicine Chief of Orthopaedic Surgery Sidra Medicine Doha, Qatar Jonathan G. Schoenecker Vanderbilt University Medical Center Jeffrey Mast Chair of Orthopaedic Hip and Trauma Surgery Nashville, TN USA

ISBN 978-3-030-12002-3    ISBN 978-3-030-12003-0 (eBook) https://doi.org/10.1007/978-3-030-12003-0 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To Gavin De Kiewiet, an inspiring surgeon, mentor, and friend And to all my other trainers who taught me to treat the whole child and not just the hip. Sattar Alshryda To my friend and teacher, Dr. Gerry Kiefer, who first introduced me to the beauty of pediatric hip surgery. And to my wife, Rhonda, and children, Seamus and Cara, for your love and support during projects like this that steal away our precious time together. Jason J. Howard To my beautiful wife and children, in case you were wondering what I was doing. James S. Huntley To my father Perry for his inspiration to make the world a better place as a pediatric orthopedic surgeon, my mother Sally for inspiring me to communicate through art, and Susan, Tyler, and Abby for encouraging me to chase my dreams. Jonathan G. Schoenecker

Foreword

Disease and deformity of the pediatric and adolescent hip are among the most frequent, challenging, and perplexing conditions faced by orthopedic surgeons treating this age group. Our understanding of the natural history of significant hip deformity has been slow in development because the time over which this evolves is protracted and may be longer than the career of an individual pediatric orthopedic surgeon, thus reducing the validity of experience as the basis for sound clinical decision-making. Similarly, the time from surgical intervention until an outcome can be accurately determined may extend almost two decades, making experience a matter of “too little, too late” in many instances. Evidence accumulated over more than one generation of surgeons in a particular institution is seldom recorded, with a few notable exceptions. Unless a surgeon is very subspecialized, experience with any hip problem, given the inherent variability, is most often very limited. Thus, there is a need for a reference source with a strong evidence-based focus, written by experts with both extensive clinical experience and a familiarization with what constitutes legitimate evidence. A reasonable question for the editors is: Why a textbook, stationary in time, when new information is immediately available online from multiple websites and open-access journals, or from a collaborative wiki that updates information as it is received? This dilemma may be resolved by distinguishing between isolated “information” and comprehensive “knowledge.” Potentially disruptive technology requires careful assessment, contextual reality, and profound understanding to ensure proper application. I believe this book does provide comprehensive and valid knowledge while at the same time utilizing sensible structure within an electronic platform. The consistent format for all chapters, combined with the authors’ genuine expertise, accomplishes a valuable evidence-based approach to complex and often rare clinical conditions seen with pediatric and adolescent hip disease and deformity. Each chapter distills topics having voluminous prior literature into manageable, readable, and informative content emphasizing key classic papers, essential evidence, and take-home messages. The editors have achieved an appropriate balance between more common conditions, such as developmental dysplasia of the hip (DDH), Legg-Calve-Perthes disease, and slipped capital femoral epiphysis, and more rare entities such as: the hip manifestations of Down Syndrome, bladder exstrophy, proximal femoral focal defi-

vii

Foreword

viii

ciency, and Larsen syndrome. The sections on poliomyelitis and tuberculosis are important inclusions for those working in developing countries. I congratulate Drs. Alshryda, Howard, Huntley, and Schoenecker for producing what should be an essential reference source and practical guide for those managing pediatric and adolescent hip conditions. 

John H. Wedge O.C., MD, FRCSC, FACS Toronto, ON, Canada

Contents

Part I Foundational Aspects 1 The History of Pediatric Hip Surgery: The Past 100 Years ��������   3 Dennis R. Wenger and James D. Bomar 2 Anatomy and Physiology of the Pediatric Hip������������������������������  29 Emily K. Schaeffer and Kishore Mulpuri 3 Biomechanics of the Hip During Gait��������������������������������������������  53 Morgan Sangeux Part II Developmental Dysplasia of the Hip 4 Developmental Dysplasia of the Hip in Young Children��������������  75 Stuart L. Weinstein and Joshua B. Holt 5 Acetabular Dysplasia in the Reduced or Subluxated Hip������������ 131 Jonathan G. Schoenecker, Ira Zaltz, Justin Roth, and Perry L. Schoenecker Part III Osteochondroses and Impingement 6 Legg-Calve-Perthes Disease������������������������������������������������������������ 169 Benjamin Joseph 7 Coxa Vara����������������������������������������������������������������������������������������� 193 Arnold Suzuki, Anthony Cooper, and James Fernandes 8 Slipped Capital Femoral Epiphysis������������������������������������������������ 207 Balakumar Balasubramanian, Sattar Alshryda, and Sanjeev Madan 9 Femoroacetabular Impingement���������������������������������������������������� 253 Erika Daley and Ira Zaltz Part IV Infectious Conditions 10 Musculoskeletal Infection of the Hip���������������������������������������������� 275 Michael Benvenuti, Megan Johnson, and Jonathan G. Schoenecker ix

x

11 Tuberculosis Involving the Hip ������������������������������������������������������ 311 Vrisha Madhuri Part V Inflammatory Conditions 12 Transient Synovitis�������������������������������������������������������������������������� 327 James S. Huntley 13 Juvenile Idiopathic Arthritis and the Hip�������������������������������������� 347 James S. Huntley, Peter S. Young, and Sanjeev Patil 14 Idiopathic Chondrolysis of the Hip������������������������������������������������ 375 Vrisha Madhuri, Noel Malcolm Walter, and Jyoti Panwar Part VI Traumatic Conditions 15 Pediatric Proximal Femoral Fractures������������������������������������������ 393 Mohamed Kenawey, Emmanouil Liodakis, Marcel Winkelmann, and Christian Krettek 16 Pediatric Pelvic Injuries������������������������������������������������������������������ 409 Mohamed Kenawey 17 Traumatic Hip Dislocation in Children ���������������������������������������� 445 Hossam Hosny, Wael Salama, Ahmed Abdelaal, and Mohamed Kenawey Part VII Neuromuscular Conditions 18 The Hip in Cerebral Palsy�������������������������������������������������������������� 467 Jason J. Howard, Abhay Khot, and H. Kerr Graham 19 The Hip in Myelomeningocele�������������������������������������������������������� 531 Emmanouil Morakis, Jason J. Howard, and James Wright 20 The Hip in Spinal Muscular Atrophy�������������������������������������������� 553 Jill E. Larson and Brian Snyder 21 The Hip in Muscular Dystrophy���������������������������������������������������� 571 Deborah M. Eastwood 22 The Hip in Charcot-Marie-Tooth Disease�������������������������������������� 583 Neil Saran 23 The Hip in Poliomyelitis������������������������������������������������������������������ 599 Hugh G. Watts, Benjamin Joseph, and Sanjeev Sabharwal

Contents

Contents

xi

Part VIII Syndromes and Skeletal Dysplasias 24 The Hip in Rett Syndrome�������������������������������������������������������������� 619 Deborah M. Eastwood 25 Hip Problems in Children with Trisomy 21���������������������������������� 631 Matthew Lea, Sattar Alshryda, and John Wedge 26 Larsen Syndrome and the Hip�������������������������������������������������������� 651 James S. Huntley 27 The Hip in Mucopolysaccharidoses������������������������������������������������ 673 Kevin Walker 28 The Hip in Arthrogryposis�������������������������������������������������������������� 691 Katie Rooks and Haemish Crawford 29 The Hip in Osteogenesis Imperfecta���������������������������������������������� 715 Maegen Wallace and Paul Esposito Part IX Tumours and Tumour-like Conditions 30 Osteoid Osteoma Involving the Hip����������������������������������������������� 737 Karl Logan, Felix Brassard, Jason J. Howard, and Pierre Schmit 31 Osteochondroma Involving the Hip ���������������������������������������������� 751 Daniel E. Porter and Fei Li 32 The Hip in Fibrous Dysplasia �������������������������������������������������������� 769 Brian L. Dial and Benjamin A. Alman 33 Bone Cysts Involving the Hip���������������������������������������������������������� 785 Laura Deriu, Sattar Alshryda, and James Wright Part X Miscellaneous Conditions 34 Bladder and Cloacal Exstrophy������������������������������������������������������ 821 Jason J. Howard, James S. Huntley, Jonathan G. Schoenecker, Sattar Alshryda, and Joao Pippi Salle 35 Athletic Injuries Involving the Hip������������������������������������������������ 841 Justin Roth and Jeffrey J. Nepple 36 Snapping Hip Syndrome ���������������������������������������������������������������� 855 Ling Hong Lee, Ed Gent, and Sattar Alshryda 37 The Hip in Congenital Femoral Deficiency ���������������������������������� 875 Fergal Monsell 38 Total Hip Arthroplasty for Pediatric Disorders���������������������������� 893 Stephen M. Engstrom and Gregory G. Polkowski

Part I Foundational Aspects

1

The History of Pediatric Hip Surgery: The Past 100 Years Dennis R. Wenger and James D. Bomar

Scientific Developments That Have Allowed Childhood Hip Surgery to Advance Key nineteenth and twentieth century inventions revolutionized surgical care of musculoskeletal disease and led to the modern era which allows safe, generally predictable childhood hip surgery. Mercer Rang listed the following important scientific achievements that allowed modern surgery to evolve. • Discovery of general anesthesia—1846 • Sterilization—discovery of bacterial origin of sepsis (Pasteur) and application to surgery (Lister—1860–1880s) • Discovery of X-ray (Röntgen)—1895 • Discovery of stainless steel (for implants)— 1920s • Discovery of antibiotics—1930s

D. R. Wenger (*) Rady Children’s Hospital, San Diego, San Diego, CA, USA Orthopedic Surgery, University of California, San Diego, San Diego, CA, USA e-mail: [email protected] J. D. Bomar Department of Orthopedics, Rady Children’s Hospital, San Diego, San Diego, CA, USA e-mail: [email protected]

© Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_1

Anesthesia Surgery has been performed since the Stone Age including trephining of skulls, primitive draining of abscesses, as well as amputations. The stress and pain of having surgery remained almost intolerable until 1846, when William T.G. Morton (Fig. 1.1), a dentist at the Massachusetts General Hospital in Boston, demonstrated the first use of ether as a general anesthetic. Morton’s method revolutionized surgical care and was quickly adopted by surgeons in the Boston area and then throughout North America. The understanding of the bacterial causation of surgical wound sepsis remained to be discovered.

Bacterial Basis for Surgical Infection In the late 1850s, Louis Pasteur, a non-MD French scientist whose main interest was wine fermentation, developed an understanding of bacteria and studied them under the microscope. His research clarified the bacterial basis for infections (Fig. 1.2). While searching for an answer to surgical infection, …Joseph Lister, a Glasgow surgeon (Fig. 1.3) was told by a chemistry professor to read an obscure French paper by Pasteur, which had been written several years earlier…

3

4

D. R. Wenger and J. D. Bomar

Fig. 1.1  William T. G. Morten of Boston demonstrating the use of ether as a general anesthetic

Fig. 1.2  Louis Pasteur, a French chemist, proved that bacteria were the cause of infections (late 1850s—Paris)

Fig. 1.3  Joseph Lister, a Glasgow surgeon, read Pasteur’s paper and then visited Pasteur in Paris. He took the ideas back to Scotland and established a method to minimize operating room infections (carbolic acid soaked bandages and spray)

1  The History of Pediatric Hip Surgery: The Past 100 Years

Lister visited Pasteur in the 1860s and took his ideas back to Glasgow where he applied them to the prevention of surgical infection. Lister traveled to North America and startled a Philadelphia surgical audience in 1876, when he stated that “I can open a joint with absolute ­certainty that no infection will occur.” Lister’s concepts were critical in allowing safe childhood musculoskeletal surgery.

Unfortunate Timing: Discovery of Anesthesia vs. Discovery of Antiseptic Technologies As Ponseti noted in his 1988 Presidential Guest Speaker Address at the Colorado Springs POSNA meeting, …the world would have been better off if understanding the bacterial cause of infections (about 1870) had occurred prior to the development of general anesthesia (about 1840)… so that methods for prevention of surgical wound sepsis would have been mastered before surgery was performed on a large scale. Fig. 1.4 Wilhelm Conrad Röntgen, a Würzburg physicist who lived above his laboratory, discovered the X-ray in 1895. The first image was of his wife’s hand, including her wedding ring!

5

For example, the American Civil War (1861– 1865) occurred in a period after general anesthesia was available but before antisepsis principles were understood. The loss of life and limb in this war was greater than in any war before or since because surgeons could now operate on almost anyone (without pain)—but with no sterile gloves or sterile methods. Routine complete closure of contaminated gunshot/cannon wounds to limbs also greatly contributed to morbidity/death because many patients became septic and died.

Röntgen and Radiographs (X-Rays) Wilhelm Conrad Röntgen (1845–1923—Fig. 1.4), a non-MD physicist in Würzburg, Germany (who lived with his wife above their scientific laboratory) developed the X-ray with the first known film performed on his wife’s hand in 1895. Röntgen subsequently received the first Nobel Prize in physics for his discovery (well deserved, we might add). Prior to Röntgen, many orthopaedic diseases were poorly understood, particularly childhood hip disorders. Differentiating between synovitis of the hip, sepsis, tuberculosis, Legg-Calvé-­Perthes disease (only defined later), and slipped capital femo-

D. R. Wenger and J. D. Bomar

6

ral epiphysis (SCFE) was impossible. The X-ray changed everything. Röntgen first announced his discovery in December, 1895 and within 3 months, X-ray units were available throughout the leading teaching hospitals in Europe and London and soon thereafter, throughout the world. “Skeletal imaging radically improved the understanding of hip disease in childhood and provided a framework for the development of corrective operations.”

Stainless Steel Implants Transition to the twentieth century brought with it many new ideas in culture, science, and surgery. Metallic implants had occasionally been used to fix fractures in the late nineteenth century, however the twentieth century brought more mechanically effective plates and screws. Bérenger-Féraud (France), who wrote on the topic in 1870, and Sir William Arbuthnot Lane (1856–1938—London), were among the first to use metal plates to treat fractures. William Sherman (1880–1979), popularized bone plating (Fig. 1.5) in the U.S. Early bone plates suffered from poor quality and corrosion thus internal fixation remained problematic until the 1920s when stainless steel was invented and used to make non-corrosive implants. This revolutionized implant predictability, setting the foundation for the A-O methods (1960s—Fig. 1.6) and development of the proximal femoral blade plate, which greatly improved modern childhood hip surgery.

Antibiotics The discovery of streptomycin, sulfa (Germany) and penicillin (Florey, Fleming—England) made

childhood hip surgery safer. Also, streptomycin allowed medical treatment of tuberculosis, which had until this time been a major cause of destructive childhood hip disease. In 1928, Alexander Fleming (1888–1955— Fig. 1.7), a Scottish bacteriologist working at St. Mary’s Hospital, London identified penicillin, a bread mould which readily lysed staphylococci. Penicillin was hard to produce and the scientific community paid little attention. Ten years later, a group of Oxford scientists led by Howard Florey (1898–1968—Fig. 1.8), an Australian working at Oxford, found Fleming’s report, developed more efficient methods for penicillin production, and in 1940 successfully treated infected mice. Human application soon followed and both British and American laboratories began mass production so that penicillin was available in World War II. Fleming, Florey, and Ernst Boris Chain (a German biochemist on the Oxford team) received the Nobel Prize for their work in 1945.

 ospitals and Children’s Orthopedic H Hip Surgery The transfer of orthopedic concepts and methods from Europe to North America in the nineteenth century laid a foundation for the development of children’s orthopedic sub-specialization in the mid twentieth century. This paralleled the development of major urban children’s hospitals in North America, as well as the Shrine Hospital system. The large urban North American children’s hospitals included orthopedic units that allowed a focus on children’s care. Out of this milieu, a group of skilled orthopedic surgeons, dedicated to the care of children’s problems, arose and provided the nidus for the establishment of the new sub-specialty of children’s orthopaedics.

1  The History of Pediatric Hip Surgery: The Past 100 Years Fig. 1.5  Metal plates to stabilize fractures were first used in the late eighteenth and early nineteenth centuries. This drawing demonstrates William Sherman’s surgical appliance application for his plate and self-­ tapping screw (1912— USA) (From: The Story of Orthopedics. Philadelphia: W.B. Saunders Company; 2000)

7

8

D. R. Wenger and J. D. Bomar

Fig. 1.6  The invention of stainless steel greatly improved bone plate efficacy The A-O blade plate (illustrated here) made childhood hip surgery more predictable (circa 1970)

Fig. 1.8  Howard Florey (Oxford) led the scientific team that performed animal and human research proving the life-saving efficacy of penicillin. Their team was awarded the Nobel Prize for their efforts (1945). Licensed under the Creative Commons Attribution 4.0 International license—Author Brigadier Sidney Smith

Fig. 1.7  Alexander Fleming (Scotland, London—1920s) identified a bread mould (penicillin), which lysed staphylococci. The scientific community paid little attention

The North American cities that developed powerful children’s orthopaedic traditions include Toronto (Fig.  1.9) and Montreal (in Canada), Boston, New York City, Philadelphia,

Wilmington, Atlanta, Chicago, Cincinnati, Memphis, New Orleans and others. Further west, Minneapolis, Iowa City, St. Louis, Dallas, Houston, Denver, Salt Lake City, Seattle, Portland, San Francisco, Los Angeles, and San Diego became important centers for children’s orthopaedic surgery. Similar developments were also occurring in Europe and elsewhere. These advances set the stage for the monumental advances in childhood hip surgery that have been made over the last 100  years. Our focus will be on the history of surgical advances in three common childhood hip conditions, namely developmental dysplasia of the hip (DDH), Perthes disease, and slipped capital femoral epiphysis. Our focus is on European and North America developments which we know the most about. Many parallel developments were occurring elsewhere in the world but we cannot cover everything in one chapter.

1  The History of Pediatric Hip Surgery: The Past 100 Years

a

9

b

Fig. 1.9 (a, b) The Hospital for Sick Children—Toronto (late nineteenth century) was typical of the urban children’s hospital movement in North America that helped

 ongenital Dysplasia of the Hip C ‘First CDH–Now DDH’ We will frequently use the classic CDH description since only recently has the term been changed to DDH (developmental dysplasia of the hip), for technical/legal and other reasons. We will use CDH when quoting from historic documents and DDH for contemporary references. …Interestingly Adolf Lorenz, of Vienna, in a classic 1920 treatise, used the title “So-Called Congenital Dislocation of the Hip” to express his conviction that CDH is not truly congenital… but instead a “preliminary phase” that can lead to subsequent upward displacement of the femoral head. He didn’t state what might alter the “preliminary phase” but clearly they could arise in the pre- or perinatal periods, or later.

Anatomy The anatomy of hip dislocation was poorly understood until Giovanni Batista Palletta (1747– 1832), who preceded Monteggia as the chief

to establish children’s orthopaedics as a specialty. The nurse/child image shows an orthopaedic patient (note leg casts)

s­ urgeon at the Ospedale Maggiore in Milan, gave the first anatomical description of a congenital dislocation of the hip discovered during the post-­ mortem exam of a 15-day-old boy. …Dupuytren in Paris (1777–1835) also gave an extensive report of the pathologic changes—after dissecting multiple specimens in older children—but stated that the condition could not be treated… Late diagnosed CDH in a child was extremely common in the past and is still occasionally seen in developed countries in the modern era (Fig. 1.10).

First Treatment: Pravaz, France Charles Gabriel Pravaz (1791–1853), served in Napoleon’s army at Waterloo and then studied medicine in Paris. Pravaz became associated with Jules Guerin (see below) in his Paris institute, and then founded his own institute in his hometown of Lyon. Pravaz described a traction treatment method for CDH that allowed reduction in 19 patients (Fig. 1.11). Pravaz was the first to describe a successful reposition of CDH with a technique that combined continuous traction with increasing

10

D. R. Wenger and J. D. Bomar

Fig. 1.10 Classic photograph depicting a 4-year-old child with bilateral CDH (Boston—1885). Note the typical increase in lumbar lordosis. (From Brown, Double Congenital Displacement of the Hip: Description of a Case with Treatment Resulting in a Cure. Boston Med Surg J 62(1885):541–46)

Fig. 1.11  Traction device for reduction of a dislocated hip designed by Charles Pravaz (Paris, Lyon) in the mid-­ nineteenth century. (From: Orthopedics: A History and Iconography. San Francisco: Norman Publishing; 1993)

1  The History of Pediatric Hip Surgery: The Past 100 Years

11

abduction and pressure over the greater trochanter. These methods are almost identical to the gradual treatment methods used for CDH reduction at the Institute Calot in Berk, France, throughout much of the twentieth century (Georges Morel).

Surgical Methods After learning of Pravaz’s CDH reduction methods, Guerin in Paris immediately modified the method (as is characteristic of the competitive spirit of surgeons) by utilizing preliminary traction and a subcutaneous tenotomy (surgery) along with manipulative reduction. He also used his tenotomy knife to cut the superior capsule over the lateral wall of the ilium, creating adhesions in this area that he thought would encourage hip stability (rather like Hippocrates’ “red-hot poker” inserted to prevent recurrent shoulder dislocation). Agostino Paci (1845–1902), a surgeon from Pisa, was the first to advise a formal, acute method for manipulative reduction of CDH. Prior to that, all treatment had been by gradual traction methods such as that described by Pravaz and Guerin. Paci’s method was later claimed by Lorenz in Austria, leading to some competition between the two.

Open Reduction for CDH Alfonso Poggi (1848–1930) of Bologna described the first successful open reduction for CDH in a 12-year-old girl. Albert Hoffa (1859–1907) of Würzburg, Germany (later Berlin), then developed and standardized an anterior open reduction for CDH that was soon used throughout Europe. The developments allowing safer surgery led to increasing conflict regarding whether or not a child with a dislocated hip should have a surgical open reduction versus a closed reduction. Adolf Lorenz [1] (1854–1946), the famous Austrian surgeon (Fig.  1.12), was a leader in the conflict related to these issues. He initially favored open reduction, but then developed an allergy to carbolic acid (the new mandatory intraoperative “system” to prevent infection). As a result, Lorenz

Fig. 1.12  Adolf Lorenz, the famous Austrian surgeon, first used open methods to reduce DDH but then reverted to closed methods due to his allergy to carbolic acid antisepsis. He traveled the world extolling his hip reduction technique

could no longer operate! He then changed his indications and decided that all children with a dislocated hip could be treated by closed reduction even up to age 7 or 8 years of age and became an international celebrity, traveling throughout the world demonstrating his technique. Other European surgeons found that the uniformly positive results reported by the famous Lorenz were not reproducible when they attempted his aggressive manipulative reduction methods and casting. Many children had capsular injury, nerve injury, fractured femurs, etc. Also Lorenz seemed blissfully unaware of the high incidence of avascular necrosis (AVN) that followed his methods (Fig. 1.13). Lorenz had a great influence on North American surgeons including training Arthur Steindler, who later became Chairman of Orthopedic Surgery at the University of Iowa in 1913. Steindler served as a mentor to I.V. Ponseti whose intellect and influence have helped define

12

a

D. R. Wenger and J. D. Bomar

Boston through Bradford, Sever, Green, Tachdjian, and other subsequent surgeons of the “Boston School”. Henry Jacob Bigelow of Boston (1818–1890) made an important contribution to CDH treatment by studying the hip capsule (the anterior Y ligament now bears his name) and then publishing a book entitled “The Mechanism of Dislocation and Fracture of the Hip: with the Reduction of the Dislocations by the Flexion Method”. Bigelow’s reduction concepts were used by Lorenz as well as subsequent surgeons who performed manipulative closed reduction for DDH.

 edial vs. Anterior Approach for Hip M Reduction

b

Fig. 1.13 (a) The somewhat extreme abduction hip spica position utilized in a Lorenz-style hip reduction. (b) Resulting AVN of the left hip noted at late follow-up of a CDH patient treated by the Lorenz method

CDH treatment in North America (subsequent trainees include Stuart Weinstein, Dennis Wenger, and many others) (Fig. 1.14).

Boston Contributions Buckminster Brown (1819–1891) of Boston spent a period of time in Paris working with Guerin and took his CDH treatment methods back to Boston. These concepts have evolved in

“In the early twentieth century, Ludloff in Germany described a medial approach for open reduction of the hip that was relatively blood-free and very effective for young children (up to age 2 years).” A capsulorrhaphy was not performed and thus excellent casting techniques were required to maintain reduction. Other surgeons, including Hey-Groves of Bristol, U.K., preferred a typical anterior open reduction which allowed some capsular repair. Bony surgery was not typically performed with open reduction in younger children; however, several experts developed wedge osteotomies performed above the acetabulum to improve the stability of the hip that had been surgically reduced.

Osteotomies to Stabilize Reduction As the twentieth century evolved, change occurred in DDH management, with younger children treated by closed reductions and older children treated with open reduction; often not effective, with many children having residual subluxation or re-dislocation. The need for simultaneous bony osteotomies became apparent. Two somewhat different concepts for reduction evolved in the U.K. versus North America.

1  The History of Pediatric Hip Surgery: The Past 100 Years

13

Fig. 1.14  The evolution of childhood hip knowledge (Univ. of Iowa—circa 1940). Professor Arthur Steindler (right circle) instructs orthopedic resident I.V. Ponseti (left circle)

Edgar Somerville (Fig.  1.15) at Oxford demonstrated that one could readily reduce a dislocated hip, even in an older child, if the labrum was excised along with an appropriate derotation, varus, and perhaps shortening femoral osteotomy [2]. Robert Salter (Fig.  1.16) in Toronto developed a quite different approach where he emphasized the importance of maintaining the labrum and the acetabular growth centers when he performed open reduction plus capsular repair plus innominate osteotomy [3] (now known as the Salter osteotomy) (Fig. 1.17). Salter’s very rigorous capsular excision/repair provided a very low incidence of re-dislocation (although occasionally the child was left with an internal rotation contracture). Ponseti in Iowa (Fig. 1.18) did important histologic work documenting the importance of the cartilage growth centers in the acetabulum [4] and both he and Salter realized that these centers, as well as the labrum, should not be excised as they contributed to subsequent acetabular growth. Ponseti and Weinstein (Iowa) also performed important studies to confirm the value of the medial open reduction for DDH.

Fig. 1.15  Edgar Somerville (Oxford) promoted open reduction for DDH that included labrum excision and often a femoral osteotomy (photo courtesy of Professor Michael Benson, Oxford UK)

14

D. R. Wenger and J. D. Bomar Salter

Fig. 1.16  Robert Salter (Toronto) promoted open reduction for DDH that preserved the labrum and included a secure capsulorrhaphy plus an innominate (acetabular) osteotomy

Fig. 1.17  Depiction of Salter’s supra-acetabular osteotomy that re-directs the acetabulum to provide improved anterolateral femoral head coverage

 urgical Improvements: Deepening S the Acetabulum Along with Open Reduction Alfonso Poggi (1848–1930) in 1880 wrote a historic paper describing the first deepening of the acetabulum for developmental dysplasia of the hip. The idea of extending or improving the acetabular roof over the femoral head was further advanced by Franz Koenig in 1871. Spitzy of Vienna, Albee of New York (Fig. 1.19), and Dega of Poland were early proponents of placing a bone graft over the acetabulum to encourage its growth. Thus Salter’s advice for a pelvic osteotomy was not new but was unique because it was a complete osteotomy that re-directed the ­acetabulum (without the risk of decreasing acetabular size—as can occur with bending osteotomies when performed with primary open reduction plus acetabular osteotomy).

Fig. 1.18  Professor I.V. Ponseti of the University of Iowa published multiple papers that have clarified the histology and later clinical course of acetabular development after DDH reduction

Salter advised the re-directional acetabuloplasty that now bears his name. The combination of reduction, a corrective capsulorrhaphy, and then a Salter innominate osteotomy (without femoral surgery) became a North American stan-

1  The History of Pediatric Hip Surgery: The Past 100 Years

15 Pemberton

Albee

Tibial graft

Fig. 1.19  Depiction of Albee’s method for improving acetabular coverage of the femoral head (New York— early twentieth century)

Fig. 1.20  Depiction of Pemberton’s acetabular “bending” osteotomy, designed to improve femoral head coverage in DDH (Salt Lake City)

dard in the 1960s and 1970s. The Pemberton osteotomy [5] (1960s) was also commonly used in North America (Fig. 1.20).

Klisic’s massive experience, presented to a North American orthopaedic audience at both the Tachdjian International Course and the landmark Royal Oak, Michigan Children’s Hip Course (mid 1970s), showed American surgeons the clear efficacy of femoral shortening. Following his presentations, most American centers began to use femoral shortening in older CDH patients. Klisic’s methods have stood the test of time and are now used widely throughout the world [6]. Initially, most North American surgeons used the Salter open reduction plus acetabuloplasty for children up to 4 or 5  years of age and avoided femoral osteotomy, except for older children. Over time, the age indications were lowered as surgeons recognized the great value of adding a derotational femoral shortening to the “Salter method” [7]. By the early twenty-first century, many (if not most) North American surgeons, faced with a completely dislocated hip in a 2 or 3-year-old child, will combine open reduction

 emoral Shortening for Older F Children Closed vs. open reduction for delayed diagnosis of CDH remained the orthopaedic standard throughout the late nineteenth and early twentieth century, but treatment in older children proved to be difficult. Hey-Groves of England may have been the first to mention femoral osteotomy to aid in reduction in 1928. Ombredanne (France) also described the concept of femoral shortening to help reduce the hip in the older child in 1932, with Klisic of Yugoslavia popularizing the method in the mid to late twentieth century (Fig.  1.21). Klisic also included an acetabular osteotomy.

16

D. R. Wenger and J. D. Bomar

The Chiari procedure creates a shelf support via a complete “translational” pelvic osteotomy that places the femoral head below a “ledge” of bone, allowing the development of a fibrocartilaginous “roof”. The Chiari osteotomy is still used in rare cases to treat residual dysplasia.

Traction Prior to Reduction Surgery As noted earlier in this chapter, before the surgical era, traction methods were routine in reducing DDH.  A dislocated hip, even in a 4  year old, could be routinely reduced with prolonged traction applied through skin traction with gradual abduction and then internal rotation. The method of Petit (Paris) and Morel (Institut François Calot—Berck) were typical of this methodology and were used even into the second half of the twentieth century [8]. The method was quite safe and effective, but is no longer widely used because of current concepts regarding the expense of a prolonged hospital stay. “The liberal use of femoral shortening as part of reduction surgery has also contributed to the non-traction era”. Fig. 1.21 Predrag Klisic, an important Yugoslav (Serbian) orthopedic surgeon introduced North Americans to the importance of femoral shortening for treatment of DDH in older children (1970s) (photo courtesy of Darko Antonivich and Klisic family)

plus acetabuloplasty plus derotational femoral shortening.

Extreme Cases During the period from the 1940s to the 1960s, there were also operations done for more severe hips that were perhaps considered untreatable by traditional methods. These included the Colonna procedure (Philadelphia), and the Chiari osteotomy (Vienna). Colonna utilized excised hip capsule to cover the femoral head, which was then placed into the acetabulum which had been reamed to a greater depth. Extreme stiffness was common and the procedure is now rarely used.

 steotomies for Residual Hip O Dysplasia Many patients who have treatment for early childhood hip dysplasia end up with inadequate femoral head coverage. The European and North American literature is filled with many types of osteotomies that have been developed by surgeons in differing countries and continents, all designed to improve acetabular coverage and thus forestall the early onset of hip arthritis that can occur in a hip that is poorly covered. Although a varus femoral osteotomy can be selected, it can leave the child with a limp and the acetabular dysplasia may never fully correct. By the mid twentieth century in North America, most residual childhood hip dysplasia has been treated by relatively simple acetabular osteotomies (Salter, Pemberton) in younger children. As

1  The History of Pediatric Hip Surgery: The Past 100 Years

17

children get older, it becomes difficult to rotate the acetabulum with a single cut thus, procedures such as triple pelvic osteotomy (Steel [9] and others) (Fig.  1.22) and periacetabular osteotomy (Ganz) were developed and are commonly used.

Triple

Perthes Disease ‘Also Legg-Calvé-­Perthes Disease’ The directions for creating this chapter were to cover the history of children’s hip surgery over the last 100 years, and interestingly Perthes disease encompasses almost exactly that period. Radiographs were first discovered in 1895 by Röntgen, and in 1910, simultaneous publications were produced by Legg, Calve, Perthes and Waldenström (Fig. 1.23) describing this unusual condition of the hip in childhood. Many different developmental bone disorders were described almost immediately following the invention of the radiograph (Sever, Kohler, Osgood, Schlatter, many others) providing a “golden era” for an orthopedic surgeon to become famous (via Fig. 1.22 Depiction of triple pelvic osteotomy as an eponym). These were broadly consolidated into described by Steel, Tonnis, Carlioz, and others. Making three bony cuts allows free rotation of the acetabulum a group known as the “osteochondroses”.

Legg

Calvé

Perthes

Waldenstrom

Fig. 1.23 The “recent” (in 1895) discovery of X-rays Perthes disease. Waldenström’s name didn’t make “the made hip disease more understandable. In 1910, these cut” but he is eponymically recognized for his chronologifour surgeons simultaneously reported the condition we cal staging of Perthes disease now refer to as Perthes, Legg-Perthes, or Legg-Calvé-­

D. R. Wenger and J. D. Bomar

18

Perthes was recognized as a non-infectious sort of “coxitis” in children whose etiology was (and continues to be) enigmatic. The pathology of Perthes disease was defined in the mid-1920s when Phemister (University of Chicago) performed histologic studies clarifying that the peculiar radiographic findings were produced by AVN.  Some prefer the term “aseptic necrosis” because in the earlier orthopedic era, most femoral head destruction was the result of tuberculosis or other type of bacterial infection.

Non-operative Treatment Almost all Perthes treatment was non-operative during the first half of the twentieth century, utilizing devices that discouraged weight bearing on the affected limb. These included the stirrup crutch, as well as many types of braces. In the 1930s, A.O. Parker in Cardiff, Wales advised that femoral head deformation could be best prevented by femoral head “containment” which he achieved by applying so-called “broomstick plasters” (in the British tradition, two cylinder casts held separated by an attached broomstick/wooden dowel [10]). The method proved quite effective and was continued in the modern era by Petrie (Montreal) and more recently by Schoenecker, et al. in St. Louis. Orthoses which matched the abduction cast model were then developed by European and North American experts. The goal of both the casts and the orthoses was to hold the thigh and leg abducted and internally rotated so that the femoral head was deeply contained within the acetabulum. The problem is that the femoral head revascularizes and reossifies slowly over a 1½ to 2-year period. Although abduction casting and bracing proved to be quite effective, …as the twentieth century evolved, and the psychology of childhood development became better understood, surgeons looked for better methods to avoid prolonged cast or brace wear.

Surgical Containment In the 1960s, early reports began to appear advising surgical methods to provide containment of the

femoral head in Perthes disease. Axer (Israel) [11], and Lloyd-Roberts/Catterall (London) [12] (Fig.  1.24), and others reported that proximal femoral varus osteotomy would allow the femoral head to remain contained with the acetabulum during normal walking with the acetabulum serving as a “mold” to allow the femoral head to heal in a more normal shape (Fig. 1.25). Stulberg et al. noted that head shape (after treatment and healing) was predictive of early onset arthritis [13]. Soon thereafter, Salter (Toronto) decided that innominate osteotomy, which he had developed for the treatment of DDH, could also be applied to contain the femoral head, during healing, in Perthes disease [14]. As in any change from nonoperative to operative treatment, substantial disagreement occurs until a new paradigm is accepted. Such was the case for Perthes disease with many European and most North American, traditionalists continuing with abduction bracing for Perthes treatment. However, once it became clear that the femoral head can only be well-contained with a brace that extends from the hip, over the knees, and to the foot (to control hip rotation), more surgeons moved towards surgical containment as a more civilized treatment. Although Lloyd-Roberts and Catterall strongly suggested that surgical containment provided a better radiographic and clinical outcome as compared to nonoperative treatment, their paper did not meet the highest standards because of the absence of matched controls. Recognizing this deficiency, in the 1980s, Dr. Tony Herring (Fig.  1.26) and colleagues at the Texas Scottish Rite Hospital—Dallas began a prospective, multi-center, randomized study that compared patients with Perthes disease who had the following treatment: 1. No specific treatment—stretching of adductors, possible physical therapy. 2. Brace containment methods. 3. Proximal femoral varus osteotomy. 4. Salter innominate osteotomy. This well-funded and carefully performed prospective “surgeon choice” randomized study led to an important publication that clarified that Perthes patients have a better outcome if they are treated by surgical containment rather than minimal treatment or bracing [15, 16]. The study noted

1  The History of Pediatric Hip Surgery: The Past 100 Years

19

Fig. 1.24  George Lloyd-Roberts (left) and Anthony Catterall (right), both from London, provided an early understanding of the efficacy of varus femoral osteotomy in treating Perthes disease Fig. 1.25  Depiction of varus femoral osteotomy. Decreasing the neck-shaft angle allows the femoral head to be “contained” while the child is standing/ walking

Varus Femoral Osteotomy

165°

135°

D. R. Wenger and J. D. Bomar

20

problem because they are trained for and enjoy performing operations… Thus, following the studies by Herring et al., there has been a strong movement throughout the world toward surgical containment surgery for Perthes disease. Two of the largest studies confirming the efficacy of surgery with femoral osteotomy are those of Wiig et al. (Oslo, Norway) [17], as well as the very large series published by Benjamin Joseph (Fig.  1.27) in Manipal, India. Joseph reported the results of surgical treatment in more than 600 patients, and also developed a sophisticated algorithm (designed for surgeons) that advises early surgical containment to minimize head lateralization and flattening that occur when a child is left untreated [18, 19]. These algorithms are sometimes difficult to follow, but most surgeons rather like Joseph’s viewpoint since it appears to encourage an operative procedure when one is not certain whether to intervene (better to prevent head deformation than wait too long). Fig. 1.26  John (Tony) Herring of the Texas Scottish Rite Hospital—Dallas led the prospective Perthes study group that more scientifically established the efficacy of surgical containment

that mild cases in younger children do not benefit from treatment and that severe cases in older children may not benefit from simple containment methods (femoral osteotomy or Salter procedure). Their study did not include surgical methods designed for severe cases where one might select “hyper containment” (Salter plus femoral osteotomy, triple innominate osteotomy, etc.). Recognizing the need for more aggressive methods of containment for severe cases, methods were developed to treat such cases including combining proximal femoral and Salter osteotomy (Olney/Asher, Kansas City, also Staheli, Seattle) or the triple innominate osteotomy (Steel, Tonnis, Carlioz, others), which allows more complete coverage of almost any femoral head. …Most orthopedic surgeons are happy to accept a surgical solution for a complex

Newer Methods The surgical community continues to search for improvements in surgical treatment. Nuno Craveiro Lopes (Portugal) has recommended serial penetration of the avascular epiphysis with small K-wire drill holes to improve vascularity between the metaphysis and the epiphysis (even in mild cases). This can also be combined with hip distraction via an external fixator. His method has been popularized elsewhere but with little scientific support of efficacy and no prospective randomized studies that we are aware of. It remains unclear as to whether the drill-hole channels remain open long enough to speed head revascularization. Medical treatment has also been pursued in several centers (bisphosphonates etc.) but as of yet not a standard addition or supplementation to surgical containment. Despite all of the above, currently, straightforward surgical treatment remains highly effective

1  The History of Pediatric Hip Surgery: The Past 100 Years

21

Fig. 1.27  Benjamin Joseph of Kasturba Medical College (Manipal, India) has the world’s largest reported series of femoral osteotomy for Perthes disease. The right image shows him with his extensive (and beloved) X-ray

research files. Dogged determination in maintaining all records and X-rays of 600 patients symbolizes the dedication of nineteenth and twentieth century researchers who have provided our knowledge base

and most children can lead a relatively normal life following surgical containment for Perthes disease.

tion of the head from the neck—in the sense that one says a criminal has been beheaded—can also be seen when one has separation of the epiphysis from the neck of the femur. One should consider that the cartilage between the epiphysis and the neck are a form of cement that joins both parts”. Later, Ernst Muller in Tubingen, Germany (1888), gave a more complete description of SCFE when he examined several autopsy specimens of patients with a pathologic process in the proximal femur. He noted that the condition was seen in healthy young individuals from 14 to 18  years of age with no known cause or prior injury, that symptoms developed slowly, that hip joint movement was usually free and without crepitus, and that the femoral head must be in a healthy joint. He then described the pathologic findings he had noted in an anatomic specimen and thought that the gradual bending at the head/ neck junction was perhaps due to a rachitic condition.

 lipped Capital Femoral Epiphysis S (SCFE) “Before the advent of X-ray, SCFE was likely just another cause of limp and likely considered a rather benign disease.” Even after X-ray became available and the pathogenic process (slip) recognized, early diagnosis was rare and the main interest of surgical treatment during the early part of the twentieth century was to perform femoral osteotomies to correct late deformity. The first specific literature reference to what is now known as SCFE was by Jean-Louis Petit who in 1723 wrote “decollement, that is separa-

D. R. Wenger and J. D. Bomar

22

Surgical Stabilization As the century evolved and more patients were studied radiographically, the disease became better defined with an understanding that a slip could be progressive and that this could be prevented by surgical stabilization. Early methods used large diameter nails similar to those which had been invented to nail femoral neck fractures. Unfortunately, these large nails could cause further separation of the physis due to the blunt trauma of inserting them. The availability of threaded Steinman pins greatly improved this circumstance. Two or three large threaded pins placed across the slip would stabilize the slip in a predictable manner. A better understanding of the clinical presentation of SCFE (chubby pre-adolescent—foot turned out—knee pain) allowed earlier diagnosis with in-situ pinning before a substantial slip had occurred. “Surgical pinning in this era was cumbersome because power K-wire drivers had not yet been invented.” The surgeon had to use a manual drill plus large diameter threaded pins whose “tip position” had to be documented by serial plain radiographs (which required processing in an often distant X-ray development room between each pin “adjustment”). Early pinnings were plagued with an unfortunately high incidence of chondrolysis despite the surgeon feeling that the final X-ray showed good pin-tip position. …This chondrolysis was not fully understood until the brilliant radiographic and geometric studies of Walters and Simon (Boston Children’s Hospital and MIT),… who documented that many patients who appeared to have the pin tips well within the femoral head, were actually penetrating the joint [20]. This “blind spot” was most commonly noted when the pin tips were in the lateral quadrants of the head (rather than central). Thus, the cause of chondrolysis proved not to be a mysterious immune condition, but instead

was the result of pin tip joint penetration (Fig. 1.28). This understanding, plus the development of cannulated screw systems, the recognition that a single large central screw could stabilize most slips, and the development of high quality image intensifiers, radically improved SCFE care in the last third of the twentieth century.

In-Situ Pinning vs. Reduction In general, the prognosis for in-situ pinning slipped capital femoral epiphysis has been quite satisfactory. The landmark paper by Boyer and Ponseti (University of Iowa) [21] demonstrated that even more severe slips pinned in-situ could have good long-term function and would not require a total hip replacement until upper middle age. Similar papers demonstrating the efficacy of in-situ pinning also appeared in the Scandinavian literature.

Societal Change and Surgical Choices The initial papers from the University of Iowa and other conservative centers were developed in communities where most patients were from farmer/laborer stock who likely did not pursue athletics in adult life. Also, the patients were often (but not always) of normal weight. A rather rapid revolution then occurred in the late twentieth century in North America with a tremendous increase in the incidence of childhood obesity along with markedly less physical activity as the population left the farms for the city—plus access to TV watching/video games. In addition, city life and more leisure time allowed greater athletic activities in teenagers, with the associated increased stress on the sensitive proximal femoral physis.

 table vs. Unstable Slip and Surgical S Choices It had been known for at least the last 75 years that acute slips had a very poor prognosis. Loder

1  The History of Pediatric Hip Surgery: The Past 100 Years

23

Fig. 1.28  Severe SCFE treated by in-situ pinning. The X-ray shows the pin tips far from the joint (top). But the hip was stiff and painful. The 3DCT study (lower) shows that the pin tips were in the acetabulum

and colleagues (Fig. 1.29) studied a large series of SCFE patients in the early 1990s and were able to divide slips into “stable” slips and those that were “unstable” due to an acute change in epiphyseal position [22]. Loder’s logic and language were accepted almost immediately and led to clearer decision-making when a SCFE patient arrived—about 95% were stable and 5% unstable.

The conservative approach was to treat all stable slips with in-situ pinning with performance of a late osteotomy if the patient had an unacceptable limb deformity (external rotation, limited hip flexion). The late osteotomies were of the valgus-flexion type, as designed by Southwick and Imhauser, which corrected limb deformity (rotation) and better centered the femoral head in the acetabulum.

24

Fig. 1.29 Randall Loder (Indianapolis) studied and defined the concept of a stable vs. unstable SCFE in the early 1990s

Unstable Slips Unstable slips present with severe pain and are unable to walk, with an ultrasound study demonstrating fluid in the joint. These patients had a poor prognosis and if they were surgically reduced, very commonly developed avascular necrosis (AVN). The treatment of the unstable slip remained problematic. If the hip was pinned in-situ, in a very displaced position, a large anterior metaphyseal “bump” impinged on the acetabulum with significant risk of early arthritis. If they were manipulated and reduced, and then pinned, the risk of avascular necrosis was high. Thus, the hip seemed doomed.

 he Move Towards Anatomic T Reduction The 1990s and the first 20 years of the twenty-­first Century brought significant (some would say radical) re-thinking about significantly displaced slips.

D. R. Wenger and J. D. Bomar

Fig. 1.30  Reinhold Ganz (Bern, Switzerland) provided the intellectual drive, research studies, and surgical techniques that greatly improved our ability to anatomically reduce a severe SCFE

…Ganz (Fig.  1.30) and colleagues in Bern, Switzerland, recognized that not every slip patient had a good clinical result after in-situ pinning and began to advise anatomic reduction… of SCFE, first in unstable and severe slips and later even mild slips [23]. This was not an entirely new idea since Herndon in Cleveland (1950s) had recommended removal of the anterior femoral head bump at the time of an in-situ pinning [24]. Also, Dunn in the U.K. had recommended anatomic reduction at the epiphyseal level for significant slips. Fish in the U.S.A. similarly promoted anatomic reduction for severely displaced slips. Ganz et al. performed careful anatomic studies to better understand the blood supply to the femoral head epiphysis and also defined the location and nature of callus that forms subperiosteally in patients with a more severe slip [25]. They then developed a trans-trochanteric surgical hip dislocation approach which allows removal of the callous and careful anatomic reduction of both stable and unstable slips. Ganz described his

1  The History of Pediatric Hip Surgery: The Past 100 Years

operation as a modified Dunn procedure, thus crediting the prior work by Dunn in the UK [26]. The Ganz method was demanding, but when done carefully, with Swiss detail and proper training, yielded favorable results. The complexity of the surgery, however, kept the method from providing uniformly positive results worldwide. Many major hip centers in North America and Europe now use the surgical hip dislocation/modified Dunn method for acute unstable slips because the risk of AVN is so high in this patient group. There is much less tendency to use this method on a stable severe slip since the incidence of AVN is negligible after in-situ pinning plus later corrective osteotomy.

 ess Complete Reduction/Monitoring L Blood Supply Parsch (Germany) has described a likely safer method of treating the acute, unstable, SCFE that includes an open anterior capsulotomy, hematoma evacuation, and partial reduction plus k-wire stabilization of the slip. The AVN rate with this method is reported as less than 10% [27]. In an attempt to improve results in unstable slips, a great deal of current research focuses on intraoperative maintenance of femoral head blood flow. Schrader and colleagues in Atlanta have developed a method of measuring the oxygen tension within the femoral head by placing a catheter through a threaded screw [28]. Also with the surgical dislocation approach, a catheter can be inserted into the femoral head and attached to a pressure monitor, allowing minute-by-minute monitoring of blood flow.

 he “Athletic Hip”: Deformity T Due to Extreme Training Another contemporary interest of Ganz and the Swiss group relates to the “translational defor-

25

mity” at the proximal head/neck junction that develops in individuals who have participated in rigorous sports training during childhood/adolescence [29]. It appears that with vigorous physical training throughout this period, in pre-disposed individuals, a subtle deformity evolves at the head-neck junction (physeal level), producing an anterolateral bump (Fig.  1.31). The so-called “lack of offset” hip impinges on the acetabular rim, leading to labral injury, cartilage delamination, and early arthritis. Corrective surgery to remove the “bump” can be performed via open or arthroscopic approaches.

Hip Preservation Surgery The tremendous knowledge advances that have been made to manage hip deformities whether due to DDH, Perthes, SCFE, the “athletic hip”, or other causes has led to an entirely new specialty described as “hip preservation surgery”, which is surgery to postpone or prevent total hip replacement. Freeman questions the hypothesis that surgery can preserve a hip and prevent the need for arthroplasty [30], which currently remains unproven. The bi-annual Bernese Hip Symposium and the North American based Academic Network of Conservational Hip Outcomes Research (ANCHOR) study group keep interested experts updated on this new surgical subspecialty. Despite a growing understanding of the origin and nature of the athletic hip impingement syndrome, there is little common understanding or agreement regarding what sports children should be limited from, numbers of hours that they should practice, and whether or not children with differing biology and/or genetic make-up may have a different risk level for developing hip deformity when participating in “cutting sports”, such as soccer, basketball, hockey, etc.

D. R. Wenger and J. D. Bomar

26 The “Athletic Hip”

Fig. 1.31  Radiographs and 3D-CT study shows the antero-lateral prominence at the head-neck junction noted in a very athletic teenage male who presented with hip pain

Summary The history of hip surgery in childhood is complex and likely was not very interesting until the radiograph was invented in 1895. Further “inventions” have made modern surgery safe. High level imaging, with image intensifiers and then CT scans, as well as MRI studies, and then three dimensional imaging have revolutionized our understanding of how children’s hips become deformed and whether corrective surgery can be beneficial. Finally, the development of academic centers that focus on childhood hip disorders, as well as a growing number of centers that focus on hip problems in adolescents and young adults, assure a continuous and changing “history” of childhood hip disease. The rapid appearance of new concepts, implants, and operations, fueled by the

digital revolution, will continue to outpace our ability to carefully study the results of new operations. Internet savvy, subspecialty oriented, “advertising clinics” will continue to jeopardize sensible progress. Future historians will revel at our amazing advances—and perhaps despair from our occasional/frequent mis-steps.

References 1. Lorenz A. My life and work. New York, NY: Charles Scribner’s Sons; 1936. 2. Somerville EW. Open reduction in congenital dislocation of the hip. J Bone Jt Surg Br. 1953;35-B(3):363–71. 3. Salter RB. Innominate osteotomy in the treatment of congenital dislocation and subluxation of the hip. J Bone Jt Surg Br. 1961;43B(3):518–39. 4. Ponseti IV. Growth and development of the acetabulum in the normal child. Anatomical, histological,

1  The History of Pediatric Hip Surgery: The Past 100 Years and roentgenographic studies. J Bone Joint Surg Am. 1978;60(5):575–85. 5. Pemberton PA.  Pericapsular osteotomy of the ilium for treatment of congenital subluxation and dislocation of the hip. J Bone Joint Surg Am. 1965;47:65–86. 6. Klisic P, Jankovic L.  Combined procedure of open reduction and shortening of the femur in treatment of congenital dislocation of the hips in older children. Clin Orthop Relat Res. 1976;119:60–9. 7. Galpin RD, Roach JW, Wenger DR, Herring JA, Birch JG. One-stage treatment of congenital dislocation of the hip in older children, including femoral shortening. J Bone Joint Surg Am. 1989;71(5):734–41. 8. Rampal V, Sabourin M, Erdeneshoo E, Koureas G, Seringe R, Wicart P.  Closed reduction with traction for developmental dysplasia of the hip in children aged between one and five years. J Bone Jt Surg Br. 2008;90(7):858–63. 9. Steel HH. Triple osteotomy of the innominate bone. J Bone Jt Surg Am. 1973;55(2):343–50. 10. Harrison MH, Turner MH, Smith DN.  Perthes’ disease. Treatment with the Birmingham splint. J Bone Jt Surg Br. 1982;64(1):3–11. 11. Axer A.  Subtrochanteric osteotomy in the treatment of Perthes’ disease: a preliminary report. J Bone Jt Surg Br. 1965;47:489–99. 12. Lloyd-Roberts GC, Catterall A, Salamon PB. A controlled study of the indications for and the results of femoral osteotomy in Perthes’ disease. J Bone Jt Surg Br. 1976;58(1):31–6. 13. Stulberg SD, Cordell LD, Harris WH, Ramsey PL, MacEwen GD. Unrecognized childhood hip disease: a major cause of idiopathic osteoarthritis of the hip. St. Louis, MO: Mosby; 1975. 14. Salter RB.  Legg-Perthes’ disease. Treatment by innominate osteotomy. AAOS Instr Course Lect. 1973;22:309–16. 15. Herring JA, Kim HT, Browne R.  Legg-Calve Perthes disease. Part I: Classification of radiographs with use of the modified lateral pillar and Stulberg classifications. J Bone Joint Surg Am. 2004;86-A(10):2103–20. 16. Herring JA, Kim HT, Browne R. Legg-Calve-Perthes disease. Part II: Prospective multicenter study of the effect of treatment on outcome. J Bone Joint Surg Am. 2004;86-A(10):2121–34. 17. Wiig O, Terjesen T, Svenningsen S. Prognostic factors and outcome of treatment in Perthes’ disease: a prospective study of 368 patients with five-year ­follow-­up. J Bone Joint Surg Br. 2008;90(10):1364–71. https:// doi.org/10.1302/0301-620X.90B10.20649.

27

18. Joseph B, Nair NS, Narasimha Rao KL, Mulpuri K, Varghese G.  Optimal timing for containment surgery for Perthes disease. J Pediatr Orthop. 2003;23(5):601–6. 19. Joseph B, Price CT.  Principles of containment treatment aimed at preventing femoral head deformation in Perthes. Orthop Clin North Am. 2011;42(3):317–27. 20. Walters R, Simon S.  The hip: Proceedings of the Eighth Open Scientific Meeting of the Hip Society. St. Louis, MO: Mosby; 1975. 21. Boyer DW, Mickelson MR, Ponseti IV. Slipped capital femoral epiphysis. Long-term follow-up study of one hundred and twenty-one patients. J Bone Joint Surg Am. 1981;63(1):85–95. 22. Loder RT, Richards BS, Shapiro PS, Reznick LR, Aronson DD.  Acute slipped capital femoral epiphysis: the importance of physeal stability. J Bone Joint Surg Am. 1993;75(8):1134–40. 23. Ziebarth K, Zilkens C, Spencer S, Leunig M, Ganz R, Kim Y-J. Capital realignment for moderate and severe SCFE using a modified Dunn procedure. Clin Orthop Relat Res. 2009;467(3):704–16. 24. Herndon CH, Heyman CH, Bell DM.  Treatment of slipped capital femoral epiphysis by epiphyseodesis and osteoplasty of the femoral neck. J Bone Jt Surg Am. 1963;45-A:999–1012. 25. Gautier G, Ganz K, Krugel N, Gill T, Ganz R. Anatomy of the medial femoral circumflex artery and its surgical implications. J Bone Jt Surg Br. 2000;82:679–83. 26. Dunn DM.  The treatment of adolescent slipping of the upper femoral epiphysis. J Bone Jt Surg Br. 1964;46:621–9. 27. Parsch K, Weller S, Parsch D.  Open reduction and smooth Kirschner wire fixation for unstable slipped capital femoral epiphysis. J Pediatr Orthop. 2009;29(1):1–8. 28. Schrader T, Jones CR, Kaufman AM, Herzog MM.  Intraoperative monitoring of epiphyseal perfusion in slipped capital femoral epiphysis. J Bone Jt Surg Am. 2016;98(12):1030–40. 29. Leunig M, Beck M, Dora C, Ganz R. [Femoroacetabular impingement: trigger for the development of coxarthrosis]. Orthopade 2006;35(1):77–84. 30. Freeman CR, Azzam MG, Leunig M.  Hip preservation surgery: surgical care for femoroacetabular impingement and the possibility of preventing hip osteoarthritis. J Hip Preserv Surg. 2014;1(2):46–55.

2

Anatomy and Physiology of the Pediatric Hip Emily K. Schaeffer and Kishore Mulpuri

Subcapital Sulcus

Introduction The hip joint is an articulating “ball-and-socket” joint that is formed by the head of the proximal femur and the acetabulum. The proximal femur consists of the approximately spherical femoral head, the femoral neck, and the greater and lesser trochanters (Fig. 2.1). The femoral head joins the neck at the subcapital sulcus—a deep groove containing the intra-articular subsynovial vascular ring (Fig. 2.1). The acetabulum is a semi-­spherical concavity formed by three major components of innominate bone: the ilium, ischium and pubis. During development and maturation, these three separate bones unite at the centre of the acetabulum, known as the tri-radiate cartilage (Fig. 2.2). The femoral head and the acetabulum come into closest contact during maximum weight-bearing and extremes of range of motion, for example flexion and internal rotation [1]. The anatomy of the pediatric hip represents a changing landscape throughout all stages of growth and development, from birth until skeletal maturity. Abnormalities during different developE. K. Schaeffer ∙ K. Mulpuri (*) Department of Orthopaedics, University of British Columbia, Vancouver, BC, Canada Department of Orthopaedic Surgery, British Columbia Children’s Hospital, Vancouver, BC, Canada e-mail: [email protected] © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_2

Neck

Fovea

Greater Trochanter Intertrochanteric Line Lesser Trochanter

©JSchoenecker2018

Fig. 2.1  The proximal femur. The bony proximal femur consists of the femoral head and neck, separated by the subcapital sulcus. The greater and lesser trochanters are non-articular traction apophysis, providing attachment points for muscles, ligaments and tendons of the hip joint. The trochanters are connected by the intertrochanteric line

mental stages can cause a number of debilitating hip conditions that can have lasting ramifications into adulthood. Hip joint development begins in the embryonic phase of life, and is a progressive, dynamic process that continues throughout fetal development, infancy and childhood. The relatively rapid hip development that occurs during the prenatal and neonatal phases can have an impact on the different surgical and non-surgical management options used in the treatment of 29

E. K. Schaeffer and K. Mulpuri

30 Iliac Crest

Ilium AIIS Tri-radiate Cartilage Pubis

Ishcium Ishcial Tuberosity ©JSchoenecker2018

Fig. 2.2  The bony acetabulum. The acetabulum is a semispherical cavity formed by three components of innominate bone. The ilium (2/5), ischium (2/5) and pubis (1/5) unite during development at the tri-radiate cartilage in the centre of the acetabulum

pediatric hip disorders. In particular, the unique anatomical and physiological features of the developing pediatric hip merit the surgeon’s consideration before deciding on the most appropriate treatment for common disorders such as developmental dysplasia of the hip (DDH), slipped capital femoral epiphysis (SCFE), Perthes disease, and hip displacement in cerebral palsy.

 renatal Development of the Hip P Joint  he Embryonic Phase of Hip Joint T Development Primary tissue differentiation occurs in the embryonic phase of pre-natal development, weeks two through eight post-conception. It is in this phase that the entirety of the musculoskeletal system develops, including the immature origins of the hip joint [2]. All the elements comprising the pelvis and the hip joint arise from a single mass of mesoderm, transitioning sequentially to blastemic tissue, precartilage, and cartilage in the early embryonic stages [3]. Appearance of the lower limb buds occurs by 28  days post-­conception, and the first recognizable structure of the hip joint—the cartilage model of the femoral diaphysis—appears dur-

ing the sixth week [2]. At this point, the hip joint is nonarticulating, the precartilage that will eventually form the femoral head being contiguous with the cartilaginous acetabulum. Primitive chondroblasts in this area begin to undergo differentiation and, as their nuclei separate, matrix material is secreted into the cytoplasm and a club-shaped femur begins to form [4]. Undifferentiated mesenchymal cells—blastemal cells—make up the trochanteric projection from the femur that will subsequently form an apophysis with its associated muscle attachments. The precartilage covers the long bone articulations and the cartilage model will form the basis for the development of osseous structures. During the seventh week, a precartilaginous centre develops in the middle of the femoral shaft and the acetabulum begins to distinguish itself from the femoral head, starting as a shallow depression of approximately 65 of the arc of a circle that eventually deepens to 180 over the course of development [2]. There is evidence to suggest that movement is necessary for appropriate joint cavity formation during this period, as studies using neuromuscular blockers in chick embryos resulted in failed hip joint cavitation, failed development of intra-articular ligaments, and replacement of muscle with fat [5]. Concurrent to joint cavity development, the femoral head and articular cartilage are forming, with apoptotic cell death occurring in the interzone to create the joint space [4]. Due to the development of a discernable joint space, this stage is notable as the first time during development that it would be hypothetically possible to “dislocate” the hip. Should that occur, said dislocation would most likely be inferior as a result of poor definition of the transverse acetabular ligament [2]. Differentiation begins in the ilium just above the acetabulum. The deepening of the acetabular cavity throughout the embryonic phase is influenced by pressure from the femoral head, and during this time, differentiation of the ligamentum teres femoris and other protective capsular structures begin [2]. From here, the cartilage of the ilium with attached labrum grows over the femoral head. By the eighth week, at the end of the embryonic phase, both primary and secondary ossification centres in the ilium appear, in addition to early formation of the synovial tissue, liga-

2  Anatomy and Physiology of the Pediatric Hip

Hip Capsule

Acetabular Epiphysis

31

Labrum Femoral Epiphysis

Transverse Acetabular Ligament

Posterior Capsule (Everted) Lateral Epiphyseal Vasculature

Anterior Capsule (Everted)

Ligamentum Teres Femoris ©JSchoenecker2018

Fig. 2.3  Structures and features of the hip joint. The hip capsule encompasses the articulating components of the acetabulum and femoral head. The cartilaginous labrum

mentum teres and the acetabular labrum (Fig. 2.3). While the neck of the femur is now elongated, unlike the ilium, the femur has not yet developed a secondary centre of ossification [1, 6].

encircles the femoral head, and the ligamentum teres femoris attaches at the fovea of the femoral head and blends with transverse acetabular ligament

ricity decreases throughout the fetal phase, resulting in a 50% spherical head at the time of birth [1, 9]. The femoral head gradually regains some sphericity in the postnatal period. As a consequence of this dynamic sphericity, acetabular coverage of the femoral head is at its lowest at birth, before again The Fetal Phase of Hip Joint increasing through infancy and childhood [1, 9]. Development In the acetabulum, the primary ossification centre appears in the ilium during the early fetal period at Ossification, vascularization and maturation of 9 weeks. Between the 11th and 16th weeks of fetal both the proximal femur and acetabulum occur in development, the muscles form around the hip joint the fetal phase, which represents week 8 post-­ and capsule, the capsule joins with the femoral periconception through to birth. Femoral ossification chondrium and the acetabular labrum, the ligamenproceeds from the centre of the diaphysis in both tum teres and transverse acetabular ligament form proximal and distal directions while the lower within the capsule, and the articulating surfaces of limb bud internally rotates [7]. By the 11th week, the femoral head and acetabulum are covered by the spherical femoral head has reached 2 mm in hyaline cartilage (Fig. 2.3). By the end of the 16th size and the femur has 5–10° of anteversion [8]. week, the ischial ossification centre of the acetabuThis anteversion will continue to increase lum has appeared and the femoral shaft has fully throughout fetal development, reaching a maxi- ossified [6]. Following delineation of the three main mum of 45° at 36 weeks. acetabular epiphyseal centres delineated by the triraThroughout the fetal phase, the femoral head is diate cartilage, ossification of the ischium begins in also changing shape. During the embryonic phase, the fourth gestational month, and then a few weeks an anatomical study of 44 hip joints in fetuses and later in the pubis [10]. During iliac ossification, the children reported that the femoral head represents lateral cortex of the ilium is persistently thicker than 80% of a complete sphere however, femoral sphe- the medial cortex, possibly due to asymmetric

E. K. Schaeffer and K. Mulpuri

32

mechanical stresses imposed by the gluteal muscles. Iliac ossification is also marked by haversian bone remodelling, becoming visible in the 28th week [3]. By 32 weeks post-conception, femoral shaft ossification reaches the greater trochanter and ilial and ischial ossification are complete [6]. Additionally, the vascular supply both to the femoral head and acetabulum matures throughout this time. The nutrient proximal femur artery, extracapsular circumflex arteries and the acetabular artery enter the acetabular fossa in the early fetal period (2 months). Between the second and third month, femoral vascularization also begins to develop distinct metaphyseal and epiphyseal supplies, while retinacular vessels perforate the femoral head and neck [11].

 ostnatal Development of the Hip P Joint

New Bone New Bone

FNI

TGP New Bone Organized Trabecular Bone

LGP New Bone Direction of New Bone Growth

©JSchoenecker2018

Fig. 2.4  The growth plates of the proximal femur. During postnatal development, three growth plates promote ossification of the femoral head and neck: the longitudinal growth plate of the femoral neck (LGP), the growth plates of the greater trochanter (TGP) and the femoral neck isthmus (FNI). Both the LGP and TGP provide longitudinal growth (black arrows), with the FNI connecting the two growth plates at the lateral neck

Acetabulum The acetabulum remains immature at birth, primarily consisting of a cartilaginous ring around the femoral head, with the tri-radiate cartilage at its deepest, central point. The ischial, ilial and pubic arms of the tri-radiate cartilage eventually fuse to form the non-articulating portion of the acetabulum during the pubescent period [10]. Simultaneously, the cartilage ring grows along with the femoral head to create the load-bearing, articular surface. Three primary ossification centres—ilial, ischial and pubic—help to define the Y-shaped tri-radiate cartilage as ossification occurs (Fig. 2.2). The largest ossification centre is the os acetabuli from the pubis, forming the anterior acetabular wall by maturity. The ilial centre forms the superior acetabular dome, while the ischial centre forms the posterior acetabular wall. These ossification centres typically completely fuse to the body of the acetabulum between 17 and 18 years of age [12, 13].

Femur At birth, femoral ossification has reached the greater trochanter and femoral neck, while the proximal femur remains cartilaginous. Three

separate growth plates contribute to the growth and morphology of the proximal femur throughout postnatal development: the longitudinal growth plate (LGP) of the femoral neck, the growth plate of the greater trochanter (TGP) and femoral neck isthmus (FNI) that connects the two on the lateral neck. The LGP and TGP provide longitudinal growth [1, 14, 15]. LGP activity is influenced by acetabular pressure (Fig. 2.4) [1, 14, 15]. The ultimate shapes of both the femoral head and acetabulum at skeletal maturity are intimately connected and depend upon their dynamic interaction throughout development. An anteroposterior view of the pelvis demonstrating the relationship between the acetabulum and proximal femur is depicted in Fig.  2.5. Disruptions in their contact relationship or growth progression can result in angular deformities, shallow acetabuli, or otherwise imperfect hip joint formation. Despite their interconnected nature, few studies have focused on the development of the femur and acetabulum in parallel, instead focusing on one or the other in near isolation. To address this, Birkenmaier and colleagues undertook a geometrical analysis of their parallel development based on plain radiographs [16]. Studying 675 hips ranging in age from

2  Anatomy and Physiology of the Pediatric Hip

33

Sacrum

Iliac Apophysis

Ilium ASIS AIIS Capsule

Labrum

Greater Trochanter Inter-Trochanteric Line Ischial Apophysis ©JSchoenecker2018

Fig. 2.5  Anteroposterior view of the pelvis. The diagram denotes key elements of the acetabulum and proximal femur and their intimate contact relationship. The acetabulum and femoral head articulate at the hip joint capsule, held in place by the labrum—a ring of cartilage the follows the outside rim of the joint socket. The iliac apophysis pro-

vides attachment sites for the gluteal muscles and tensor fascia latae, the ischial apophysis provides attachment sites for the hamstrings, the ASIS (Anterior Superior Iliac Spine) provides attachment sites for the sartorius muscle and the AIIS (Anterior Inferior Iliac Spine) provides the attachment site for the direct head of the rectus femoris

9  months to 16  years, the authors found that no major changes in angular conformation of the proximal femur took place beyond the age of 10; however, acetabular coverage and centric alignment were dynamic through to skeletal maturity [16]. Load and muscle lever arms also increased through the end of the growth period, but their relative ratio remained constant beyond 10  years of age. These findings provide support for surgical timing considerations for corrective osteotomies of the proximal femur. Little overall impact on hip joint geometry would be expected if performed at or after 10 years of age [16].

acetabular fossa is defined by the crista articularis, or articular ridge, which reaches its highest point on the posterior and inferior surfaces, and the outer articular margin of the acetabulum is continuous with the labrum [1]. The fibrocartilaginous labrum, together with the transverse acetabular ligament, forms a complete ring around the acetabulum which is largely fully-­formed and developed at birth [17]. The posteroinferior third of the articular surface—the sustenaculum—supports the femoral head when sitting or in supine [3]. The internal anatomy of the acetabulum is characterized by heterogeneous bone due to the uneven distribution of weight-­bearing forces [1, 18]. The proximal femur is comprised of the femoral head, neck, and greater and lesser trochanters. The subcapital sulcus, where the head and neck join, contains the intra-articular subsynovial vascular ring (Fig.  2.1) [1]. On the surface of the medial femoral head, the fovea capitis provides the ligamentum teres attachment site, while the intertrochanteric line provides the attachment site for the iliofemoral ligament (Figs. 2.1 and 2.5) [1, 6]. The trochanteric crest provides attachment sites

 ormal Hip Joint Anatomy N and Geometry The surface anatomy of the acetabulum is characterized by its three major components: ilium (superior 2/5), ischium (inferolateral 2/5) and pubis (medial 1/5) (Fig. 2.2). At skeletal maturity, these three bones fuse into one at the central triradiate cartilage (Fig.  2.2). The non-articular

34

for the short external rotators, piriformis, obturator internus, gemelli, and quadratus femoris (Fig. 2.5) [1]. Finally, the greater and lesser trochanters provide attachment sites for gluteus medius and minimus, and psoas respectively (Figs. 2.1 and 2.5) [1].

Acetabular Depth Acetabular depth may be measured on radiographs by drawing a line from the anterior to posterior acetabular rim and constructing a perpendicular line extending to the deepest part of the acetabulum. A few studies have determined acetabular depth, by direct anatomic measurement and arthrographic measurement, in fetuses ranging in age from 11  weeks to term [4, 19]. These studies found that acetabular depth increased an average of 3  mm during weeks 11–24 of fetal development. There is limited availability of information on acetabular depth in children; however, studies in adults suggest a normal acetabular depth of 24–25 mm [17, 20]. Acetabular depth has been reported to correlate with degenerative changes during dysplasia, and a shallow acetabulum has been shown to be associated with hip instability and inferior clinical outcomes following hip arthroscopy for femoroacetabular impingement [21, 22]. Conversely, it has been postulated that a deeper acetabulum may be related to an elevated incidence of SCFE due to increased shear forces on the proximal femoral physis [23]. Whyte et  al. found that decreased acetabular depth correlates with poorer clinical outcomes following hip arthroscopy for treatment of femoroacetabular impingement (FAI), and therefore may potentially be used as a prognostic indicator for these patients [21].

Acetabular Diameter Very few studies have been performed specifically examining acetabular diameter, particularly in association with pediatric hip disorders. The measure of acetabular diameter, however, is important in surgical planning for procedures about the hip joint, particularly those involving the use of prosthetics. One study measured the

E. K. Schaeffer and K. Mulpuri

relative acetabular diameter of 200 fetuses [1, 24]. Their findings indicated that the majority of acetabula were oval (57.5% vertically and 13.5% transversely oval compared to 29% round). Additionally, round acetabula were more common in younger fetuses [24]. In adults, the mean anteroposterior acetabular diameter was found to be 5.1 cm in males and 4.7 cm in females [17]. This difference in acetabular size between males and females is consistent throughout growth to adulthood. During development, the rate of acetabular diameter widening is greater than the rate of growth of the remainder of the pelvis.

Femoral Head Diameter Femoral head shape is dynamic throughout embryonic and fetal development, and is influenced by its plasticity, joint contact forces, intrauterine fetal positioning, and fetal movement. Femoral head growth rates are accelerated in early development, with the growth rate slowing after birth [1, 24]. Dega noted the femoral head was more round in younger fetuses than older ones, postulating that the femoral heads may become flatter as the fetus grows and hip movement becomes more restricted [12]. Chung measured the femoral head diameters of 51 fetuses (2.8–4.8 months) and 78 children (birth to 15 years) [1]. Their findings indicated that femoral head diameter increased most rapidly in utero, with a growth rate of 23 mm/year. From birth to 6 months, this growth rate decreased to 10 mm/year, and then further decreased to 4 mm/year at 1–2 years of age. Acetabular growth also slows during this period. These observations may help explain decreased success rate associated with closed reduction and hip spica casting for children with hip dysplasia, once over 18 months of age [25, 26]. Chung further postulated another period of increased femoral head growth at approximately 8 years of age, correlating with typical height and weight increases in normal children at this point of development [1]. An anatomic study of 400 adult specimens reported an average head diameter of 49.7 mm in males and 43.8 mm in females [27]. Much like acetabular diameter, femoral head diameter is important to guide surgical planning for procedures involving prosthetics, such as total

2  Anatomy and Physiology of the Pediatric Hip

35

hip arthroplasty. However, femoral head diameter has been reported to be smaller in infants and toddlers in severe cases of DDH [28]. Whether this is a potential cause or consequence of dysplasia is difficult to definitively determine, and may potentially be some combination of the two. An important consideration related to, but extending beyond, femoral head size is that of femoral head shape. The overarching concept for the treatment of DDH is that of achieving congruent reduction of the hip. However, as Rosenberg and colleagues argue in a recent study, such congruent reduction is not possible if the shape of the femoral head and the acetabulum do not match [29]. Past studies have explored femoral head shape both in normal hips [30, 31] and in DDH [32, 33], and found the femoral head to be largely spherical in normal hips and aspherical in DDH. However, the two studies examining DDH femoral head shape used two-dimensional plain radiographs and arthrograms to address a three-­ dimensional question, and lacked normal hips for comparison. Using MRI scans to compare 14

110°

Coxa Vara (135°) ©JSchoenecker2018

Fig. 2.6  Neck-shaft angle (NSA) of the femur. The NSA represents the angle between the axis of the femoral neck and axis of the femoral shaft (black arrows). Postnatally, a normal

NSA measures between 120 and 135° (middle panel), with a decreased NSA resulting in coxa vara (left panel) and an increased NSA resulting in coxa valga (right panel)

E. K. Schaeffer and K. Mulpuri

36

Often in patients with cerebral palsy, or other paralytic conditions such as myelomeningocele or polio, the NSA is either increased, or remains constant in the postnatal period from birth [35]. This observation is likely due to the fact that load on the femoral head is less than normal (due to abductor weakness and altered weight-bearing; see Chap. 18) in these patients, and is evidence that normal muscle forces play an important role in normal postnatal hip development [1]. The NSA is important due to its involvement in both a number of pathologic hip conditions, and in the surgical decision-making process [36]. Specifically, Walton and colleagues demonstrated the importance of NSA in guiding device considerations for extracapsular proximal femur fixation [37]. NSA measurement can also impact femoral osteotomy planning for reconstructive and salvage hip procedures in patients with a number of pediatric hip conditions, including cerebral palsy (Chap. 18), coxa vara (Chap. 7), mucopolysaccharidosis (Chap. 27), and osteogenesis imperfecta (Chap. 29), among others [34].

a 40° Increased Femoral Anteversion

15°

Normal Femoral Anteversion

–30°

Femoral Retroversion

Femoral Anteversion Angle Femoral anteversion, or the femoral torsion angle, is formed by the intersection of the coronal plane of the posterior aspect of the femoral condyles and a line through the femoral neck and head centre (head-neck axis) (Fig.  3.3, Chap. 3) [1]. Femoral anteversion is defined as the femoral head and neck pointing anterior to the posterior coronal femoral condyle plane (Fig. 2.7a). In contrast, femoral retroversion is defined as the headneck axis pointing posterior to this plane. Femoral anteversion can first be measured in the 11th week of prenatal development, when it is typically approximately 5  [1, 38]. This anteversion increases throughout development, plateauing at 45° in the 36th gestational week [38]. Postnatally, muscle forces acting across the hip joint serve to gradually decrease femoral anteversion, with approximate values of 30°  at 1  year of age and

Angle of ante-version

b

©JSchoenecker2018

50

Fabry, Macewen, Shands (JBJS 55A Dec., 1973, p 1728) Crane (JBJS 41A April., 1959, p 423)

40 30 20 10

2

4

6

8 10 12 Age in years

14

16

Fig. 2.7  Femoral Anteversion. (a) The femoral torsion angle is formed by the intersection of the coronal plane of the posterior aspect of the femoral condyles (horizontal line) and a line through the femoral neck and head centre (head-neck axis, angled line). Femoral anteversion results when the femoral head and neck point anterior to the posterior coronal femoral condyle plane (top and middle panels). Femoral retroversion results when the head-neck axis point posterior to the femoral condyle plane. (b) Line graph depicting the change in femoral anteversion angle throughout development from birth to skeletal maturity. Figure 5(b) from Chung, 1981, Used with permission [1]

2  Anatomy and Physiology of the Pediatric Hip

15° at skeletal maturity (Fig. 2.7b). Hip instability can result when excessive anteversion occurs in both the femur and acetabulum; whereas, acetabular retroversion can compensate for femoral anteversion. Excessive anteversion improves spontaneously in many children; however, the capacity for spontaneous improvement markedly decreases beyond the age of 8 years [1, 39]. There is currently no strong evidence to suggest excessive femoral anteversion on its own is a potential contributing factor to degenerative changes in the adult hip. However, femoral anteversion is often associated with pathologic ­ conditions such as DDH [40] and metatarsus adductus [1].

 ostnatal Ossification of the Femoral P Head The single ossification centre of the femoral head is first radiographically detectable between the ages of 4 and 6 months [1]. Normal (but rare) variations are seen whereby multiple ossification centres occur, each with its own arterial supply [14]. The size of the femoral head ossification centre increases linearly with age through to puberty, accompanied by a corresponding decrease in the total amount of cartilage within the anlage. The proximal femur grows in length throughout this period by cartilage cell proliferation at the capital growth plate, the trochanteric growth plate, the pre-osseous femoral head cartilage, and the greater trochanter pre-osseous cartilage and perichondrium (Fig. 2.4) [1, 41]. Intraosseous blood vessels are most concentrated around areas of endochondral bone formation below the capital growth plate and femoral head articular cartilage [41]. Throughout this femoral growth phase, the shape of the proximal femur is consistently maintained by the coordinated growth and resorption of bone medially at the femoral neck and laterally just below the greater trochanter [1]. Timing of ossification follows a predictable schedule with postnatal ossification occurring first in the femoral head, then the greater trochanter and finally, the lesser trochanter.

37

Delays in hip joint ossification can be associated with specific pathologies. For example, in DDH, forceful reduction of a dislocated hip may result in delayed ossification. Conversely, prolonged dislocation without reduction can likewise result in delayed ossification, thought to be associated with reduced pressure on the femoral head from lack of contact with the acetabulum. Delays in bone age in comparison to chronological age are also seen associated with pathologies including Perthes disease, hypothyroidism and hypopituitarism, while advanced bone age is seen in Albright syndrome—a genetic disorder impacting bones, skin pigmentation and hormones.

 eriarticular Muscles About P the Hip Joint  unctional Muscle Groups About F the Hip Joint Muscles about the hip can be grouped based on their functions in relation to hip movement and stability. The major functional muscle groups of the hip joint include those involved in flexion, extension, medial rotation, lateral rotation, adduction and abduction. The primary hip flexors are the psoas major and iliacus that together form the iliopsoas, with assistance from the pectineus, rectus femoris and sartorius [42]. Primary extensors are the gluteus maximus and the hamstrings. The tensor fascia latae and fibres of the gluteus medius and gluteus minimus are the key contributors to medial hip rotation, while the obturator muscles, quadratus femoris and gemelli, with assistance from the gluteus maximus, sartorius and piriformis, are the key contributors to lateral rotation. The adductor longus, brevis and magnus, with assistance from the gracilis and pectineus, control hip adduction, while the gluteus medius and minimus, with assistance from the tensor fascia latae and sartorius, control hip abduction [42, 43]. Figure  2.8 depicts key functional muscles and their locations about the hip joint.

E. K. Schaeffer and K. Mulpuri

38 Gluteus Medius

Gluteus Medius

Tensor Fascia Lata

ASIS

Iliopsoas

Piriformis Obturator Internus

Rectus Femoris

Gemelli Quadratus Gluteus Maximus

Vastus Lateralis

Ischial Apophysis Hamstrings

©JSchoenecker2018

Iliopsoas Gluteus Minimus

Pubic Apophysis

Anterior Tubercle Iliopsoas Insertion

Ischial Apophysis Rectus Femoris Hamstrings

Lesser Trochanter

Adductors (Expand)

Vastus Intermedius Vastus Lateralis

©JSchoenecker2018

Fig. 2.8  Functional muscles about the hip joint. Primary hip flexors include the iliopsoas (upper and lower right panel) and rectus femoris (upper right and lower left panels). Primary hip extensors include the gluteus maximus (upper left panel) and hamstrings (upper right and lower

left panels). Tensor Fascia Lata and gluteus medius (upper left panel) contribute to medial rotation. The obturator internus, quadratus, gemelli and piriformis contribute to lateral rotation (upper right panel)

2  Anatomy and Physiology of the Pediatric Hip

Imbalance of these muscle groups can result in changes to the hip joint anatomy over a period of time, and the pediatric hip can be particularly susceptible to any of these muscle loading imbalances due to continual developmental changes during this period [44]. Consequently, increased or decreased muscle loading can influence bone shape and structure differently in newborns, adolescents and adults, and it has been shown that postnatal muscle loading at the hip is critical for the formation of the proximal femur and acetabulum [45–47]. A series of experiments in mice demonstrated that unilateral postnatal unloading of key hip stabilizing muscle groups decreased acetabular coverage and bone accumulation of the femoral head, while also altering the size and shape of the femoral head in unloaded hips versus the contralateral hip [48]. Adaptations to this altered loading during postnatal growth may have a profound influence on proliferation and differentiation of growth plate cells, subsequently resulting in regions of significant loss of acetabular coverage and altered contact mechanics of the proximal femur and acetabulum. Consistent with these findings, it has been reported that 79–90% of cases of osteoarthritis of the hip have hip joint abnormalities that lead to increased impingement or altered loading patterns [49].

 urgical Considerations for Muscles S About the Hip Joint Detailed discussion on the muscles, muscle groups and biomechanics of the hip joint can be found in Chap. 3. However, a brief discussion on their importance for surgical procedures will be outlined here. Consideration of the anatomical orientations of many of the muscles about the hip joint in relation to their function is of paramount importance for good surgical outcomes. The hip joint is covered by a large muscle envelope consisting of 21 muscles that cross the joint (Fig. 2.8) [50]. Specific muscles in this envelope carry major surgical significance depending on the intended procedure approach. Together with the iliotibial band, the tensor fascia latae and gluteus maximus form the main outer layer of this

39

muscular envelope (Fig. 2.8, upper panel); therefore, at least one of these three components must be split to allow access to deeper muscles in the gluteal region [50]. The gluteus medius is also a key muscle to consider in surgical planning (Fig. 2.8, upper panel). It functions as a major hip abductor, and together with the gluteus minimus (Fig.  2.8, lower panel), stabilizes the hip joint during swing phase of the gait cycle. Weakness or inefficiencies of these abductors can cause a Trendelenburg gait, while tightness can cause damage to the femoral head. Lateral surgical approaches to the hip aim to avoid the need to detach the gluteus medius, or facilitate its reattachment following the procedure [51–53]. Additionally, despite the relatively weak abduction role of the gluteus minimus (Fig. 2.8, lower panel), it also contributes to flexion and internal rotation; therefore, its accurate reattachment during surgery is also of critical importance [50]. Both the piriformis and the iliopsoas are other surgically important muscles (Fig. 2.8, upper panel). The piriformis is a short external rotator of the hip that provides insight into the area’s neurovasculature [50]. Specifically, the superior gluteal vessels and superior gluteal nerve enter the gluteal region above the pelvis by passing below the piriformis, necessitating caution during surgical procedures. The iliopsoas is comprised of both the psoas major and iliacus muscles that are usually separate in the abdomen, but merge in the thigh to form the primary hip flexor [54]. The iliopsoas tendon inserts into the lesser trochanter posteromedially (Fig. 2.8, lower panel) and must be released during anterior and medial surgical approaches to the hip joint to facilitate exposure [50]. Finally, the iliocapsularis muscle is often overlooked in its role in surgical planning, but appears to be an important stabilizer specifically in the dysplastic hip [55]. The iliocapsularis— also referred to as the iliacus minor or iliotrochantericus—overlies the anterior hip capsule, originating mainly from an elongated attachment to the anteromedial hip capsule and part of the border of the anterior inferior iliac spine (AIIS), inserting just distal to the lesser trochanter [55]. It serves as an important landmark for exposure

40

of the anteromedial hip capsule and psoas tendon interval during Bernese periacetabular osteotomy (PAO) [56–58]. The Bernese PAO is performed through an anterior iliofemoral incision, and the iliocapsularis muscle must be elevated to allow for the ischial osteotomy interval between the hip capsule and psoas tendon [58]. The iliocapsularis had long been unappreciated or even unmentioned in the surgical literature until this finding during PAO, as it is thought to have a superfluous role in normal hips or hips with excessive acetabular coverage [55]. Babst and colleagues demonstrated that the iliocapsularis is hypertrophied in dysplastic hips, likely due to increased demand on the muscle as a consequence of hip instability [55]. Hip instability causes chronic overloads and shear forces that are most pronounced when in full extension and external rotation. At this point, the iliocapsularis is maximally stretched. Hypertrophy of this muscle may therefore passively assist in constraining the femoral head in a deficient acetabulum. Conversely, when there is acetabular over-coverage, the bony anatomy naturally provides stabilization of the femoral head, thus mitigating the need for the iliocapsularis and leading to muscle atrophy [55].

Surgical Approaches to the Hip Joint The anterior approach to the hip is also referred to as the anterior iliofemoral or Smith-Petersen approach [55]. The anterior approach facilitates sufficient exposure of the acetabulum while leaving the abductor mechanism intact. The anterior approach is indicated in a variety of situations, including open reduction for a hip dislocation in DDH, synovial biopsy, hemiarthroplasty, pelvic osteotomies, total hip arthroplasty and joint drainage and irrigation. Superficial and deep fasciae are first divided, and then the attachments of the gluteus medius and tensor fasciae latae are freed from the iliac crest (Fig. 2.8 upper panel). A blunt dissection is then performed between the tensor fasciae latae and the sartorius below the anterior superior iliac spine (ASIS). Deep dissection is then performed through the interval between the rectus femoris and the gluteus medius, involving detachment of the rectus femoris from its origins

E. K. Schaeffer and K. Mulpuri

to expose the hip joint capsule (Fig.  2.8 upper panel). The anterior approach enables both the superficial and deep dissections to occur through the internervous planes, minimizing potential neurovascular complications, while providing exposure of the anterior column and medial wall of the acetabulum. The lack of disruption of the abductor mechanism mitigates post-operative limping, and it carries minimal risk of hip dislocation. However, the anterior approach does not provide sufficient access to the posterior column of the acetabulum and femoral medullary canal, limiting its utility for certain surgical procedures [55]. The anterolateral approach uses the intramuscular plane between the tensor fascia latae and gluteus medius and provides better exposure of the acetabulum than the anterior approach [55]. However, this approach necessitates disruption of the abductor mechanism to allow for hip adduction and facilitate acetabular exposure. This approach can be indicated for similar procedures to the anterior approach including total hip replacement, hemiarthroplasty and synovial biopsy of the hip, but also for open reduction and internal fixation of femoral neck fractures and biopsy of the femoral neck. In this approach, the fascia lata is incised at the posterior margin of the greater trochanter, and the interval between the gluteus medius and tensor fasciae latae is identified by blunt dissection. The gluteus medius and gluteus minimus are retracted proximally and laterally in order to expose the superior aspect of the joint capsule covering the femoral neck. In addition to the advantages of the anterior approach, the anterolateral approach also provides good exposure of the femoral neck and carries a low risk of avascular necrosis of the femoral head due to preservation of the superior retinacular vessels supplying the femoral head. However, exposure of the acetabulum remains limited, and it carries a risk of damage to the superior gluteal nerve [55]. Lateral approaches, either direct or trans-­ trochanteric, provide better exposure of the acetabulum than anterior/anterolateral approaches, but require complete or near-complete disruption of the abductor mechanisms [55, 57]. The direct lateral approach, first described by McFarland and Osborne [52], and later modified by Hardinge [53], involves division of the fascia lata and ilio-

2  Anatomy and Physiology of the Pediatric Hip

tibial band over the greater trochanter, followed by posterior retraction of the gluteus maximus and anterior retraction of the tensor fasciae latae. The gluteus medius is then separated from surrounding muscles by blunt dissection, and the tendon of the gluteus minimus is divided to expose the joint capsule. The trans-trochanteric technique involves an osteotomy of the attachment of the gluteus medius and vastus lateralis (Fig.  2.8, upper and lower panels) at the greater trochanter to enable lifting the muscles and the mobile trochanteric fragment in one piece [59]. This approach facilitates excellent exposure of the acetabulum and proximal femur and permits trochanteric transfer to restore abductor power after total hip replacement. This technique has been further popularized by Dr. Ganz in surgical hip dislocation approaches [60]. However, due to disruption of the abductor mechanism, post-­operative abductor weakness can result, and trochanteric nonunion may occur at the osteotomy site. There is also an increased incidence of trochanteric bursitis and heterotopic ossification [61]. Posterior approaches to the hip are most commonly used for total hip replacement but can also be used for posterior hip dislocations [50, 62]. Like the anterior approach, the posterior approach does not disrupt the abductor mechanism and facilitates a rapid post-operative rehabilitation process. However, there is a higher risk of post-­operative dislocation and injury to the sciatic nerve [50]. The medial approach was first developed by Ludloff to treat developmental hip dislocations in infants and young children [63]. Here, the dissection plane is between the adductor longus and pectineus; while Ferguson’s modification sees the superficial muscle interval between gracilis and adductor longus, and the deep interval between the adductor brevis and adductor magnus (Fig.  2.8, lower panel) [50, 64]. The medial approach requires little dissection, thus keeping operation time and blood loss to a minimum—an important consideration for infants and young children. However, it has limited use only in infants and carries a risk of damage to the medial femoral circumflex artery (MFCA), thus increasing the risk of AVN [50]. While there are distinct merits and drawbacks associated with each of these surgical approaches to the hip joint, the common theme is that the prac-

41

ticing surgeon should possess intimate knowledge and exercise great attention to the anatomical orientations and functional contributions of the hip joint muscles in order to maximize benefit and mitigate risk with any of these surgical approaches.

Vascular Supply to the Hip Joint Vascular Supply of the Femoral Head Vascularization of the femur is first seen at the end of the embryonic developmental phase. At the eighth week, capillaries penetrate the cartilage model of the femur at the mid-diaphysis [14]. This is the site of the primary ossification centre of the femur. These capillaries eventually form the nutrient artery in the mature bone. As development continues, a ring of vessels encircle the femoral neck, forming primitive retinacular vessels which invade the cartilage model at the articular cartilage-neck junction during the 14th week of gestation. Once in place, this ‘primitive’ vascular model remains largely intact through to skeletal maturity. In the postnatal period, the vascular supply to the proximal femur is composed of three main arterial Femoral Artery

LFCA MFCA

Deep Femoral Artery ©JSchoenecker2018

Fig. 2.9  Posterior view of the vascular supply to the femoral head. The medial femoral circumflex artery (MFCA) and lateral femoral circumflex artery (LFCA) branch from the profunda femoris artery or common femoral artery, forming the extracapsular arterial ring. The MFCA branches to form the medial ascending cervical arteries and the LFCA branches to form the anterior ascending cervical arteries that penetrate the hip joint capsule

E. K. Schaeffer and K. Mulpuri

42

systems: (1) the retinacular or extracapsular ring of vessels, (2) the foveal or ascending cervical vessels and (3) the intraosseous or intracapsular vessels (Fig. 2.9) [1, 14]. A fourth arterial group, those of the ligamentum teres, are also present but their precise role and contribution to the blood supply of the femoral head is more controversial and less well-defined. The ligamentum teres is routinely excised during surgical dislocation, and for DDHrelated procedures, usually with negligible impact [65]. Additionally, in a study involving injections to the femoral head, contributions of the ligamentum teres to vascularization was entirely absent in some cases [66].

a

Fig. 2.10  Photograph (a) and corresponding line diagram (b) of a posterosuperior view of the right hip depicting the penetration of the terminal branches of the medial femoral circumflex artery (MFCA) into the hip joint capsule. The line diagram shows (1) femoral head, (2) gluteus medius, (3) deep branch of the MFCA, (4) terminal subsynovial branches of the MFCA, (5) gluteus medius tendon inser-

The medial and lateral femoral circumflex arteries (MFCA and LFCA, respectively) comprise the extracapsular ring, and arise directly from the profunda femoris artery or directly from the femoral artery (Figs.  2.9 and 2.10) [1, 67]. These arteries are the primary vascular supplies to both the proximal femur and greater trochanter [14]. The second major arterial supply system is formed by ascending cervical branches from the extracapsular ring (Fig.  2.9). These branches penetrate the hip joint capsule and travel around the femoral neck under the synovium, anastomosing laterally at the piriformis fossa. Once intracapsular, the retinacular vessels travel within

b

tion, (6) piriformis tendon insertion, (7) lesser trochanter and nutrient vessels, (8) trochanteric branch, (9) first perforating artery branch and (10) trochanteric branches. Figure reproduced with permission of the Licensor through PLSclear from Gautier et  al., 2000 [67] (PLS Reference Number 4914)

2  Anatomy and Physiology of the Pediatric Hip

fibrous extensions of the capsule wall. These extensions have been termed the retinacula of Weitbrecht (RW), and they penetrate the femoral head just distal to the articular rim [14, 65, 68]. The femoral neck elongates during growth, and in this stage the anterior retinacular artery (the terminal branch of the LFCA) regresses to supply the anterior metaphyseal area, diminishing its overall contribution to the proximal femoral blood supply [69]. During this phase, the superior and inferior retinacular arteries (SRA and IRA, respectively; the terminal branches of the MFCA) grow along with the femoral neck and increase their overall contribution to the proximal femoral blood supply (Fig. 2.11). The SRA has long been considered to be the primary blood supply to the femoral head, with the IRA being a relatively minor contributor [70]. However, several more recent studies using quantitative MR imaging have demonstrated a potential larger role for the IRA in femoral head perfusion [71]. The third major supply system is the intracapsular synovial ring of vessels that ascend superficial to the perichondrial ring, encircle the femoral head and supply the epiphysis [72]. During growth and development, the proximal femoral capital physis acts as a barrier between

43

the blood supplies of the epiphyseal and metaphyseal ossification centres until the physis closes at skeletal maturity [72]. Throughout this phase, each penetrating vessel supplying the epiphysis is responsible for a distinct zone of vascularity. Consequently, until maturity, patterned necrosis may occur in the femoral head if a particular vessel (i.e. vascular zone) is compromised [72].

 renatal Development of the Femoral P Blood Supply In the 8-week-old embryo, the nutrient proximal femur artery, the MFCA, and the LFCA are all present. Vascular sprouts from the medial and lateral ascending cervical arteries then begin to branch, eventually forming a fine vascular network beneath the synovium that extends to the femoral head circumference in the 12–16 week-­ old fetus [1]. At this point, blood vessel rudiments originating from the MFCA and LFCA can be seen on the medial and lateral greater trochanter. In the following gestational weeks, the number and branches of these rudimentary arteries increases. By 20–24  weeks, four to five ascending cervical arteries are present on the anterior and posterior femoral neck, while the descending metaphyseal artery has branched and passes distally in close proximity to the femoral shaft nutrient artery. At 24–28 weeks, the nutrient artery of the femur anastomoses with the nutrient artery of the acetabulum and the blood vessels on the medial and lateral sides of the greater trochanter anastomose (Fig. 2.12) [1].

Postnatal Development of the Femoral Blood Supply

Fig. 2.11  Anterior aspect of the right hip following capsulotomy. Photograph depicting the retinacular artery derived from the medial femoral circumflex artery (MFCA). (1) Femoral artery; (2) MFCA; (3) Retinacular branch; (4) Anterior aspect of the femoral neck. cran = cranial and lat = lateral. Figure reproduced with permission of the Licensor through Copyright Clearance Center from Kalhor et al. 2009 [68] (License Number 4275531114422)

As described, once established prenatally, the femoral vascular supply remains largely intact throughout postnatal growth and development. The medial, posterior and lateral portions of the extracapsular ring surrounding the base of the femoral neck comprise a continuation of the MFCA, while the anterior portion of the ring is a continuation of the LFCA [1, 14]. The MFCA

E. K. Schaeffer and K. Mulpuri

44

Arterial branches to superior gluteal artery

Posterior ascending cervical arteries

Capsule Femoral neck Medial ascending cervical arteries Lateral ascending cervical artery

Arterial branches to obturator externus muscle

Medial circumflex femoral artery

Inconstant Branch

Iliopsoas

Anterior ascending cervical arteries

Femoral artery Profundus femoral artery

Lateral circumflex Femoral artery

Fig. 2.12  Cross-sectional diagram of the proximal left femur depicting the extracapsular arterial ring and arteries crossing the hip capsule. Figure from Chung, 1981 [1], Used with Permission

passes posteriorly between the iliopsoas and pectineus muscles and then between the medial capsule and obturator externus muscle before branching to form the medial ascending cervical branches. These branches subsequently breach the capsule and travel up the femoral neck in the synovium, with several branches also supplying the obturator externus muscles. The intertrochanteric crest is located in the posterior extracapsular region. Here, the posterior ascending cervical arteries breach the capsule and travel up the femoral neck, and the MFCA runs adjacent to the posterior capsule before passing through the lateral capsule in the posterior trochanteric fossa. The LFCA passes anterolaterally to the iliopsoas and then divides into several branches, one of which comprises the anterior ascending cervical branch to the femoral head and neck [1, 14]. The ascending cervical arteries pierce the capsule around the base of the femoral neck, with an average of 2 arteries anteriorly, 2 medially, 1.4 posteriorly and 1.1 laterally [1, 14]. The epiphyseal and metaphyseal lateral ascending cervical

artery branches all originate from a common arterial stem in the posterior trochanteric fossa that supplies the greater trochanter and femoral head and neck for the entirety of the growth and development period. Epiphyseal branches of the ascending cervical arteries supply the capital secondary ossification centre, mostly crossing the lateral aspect of the mid-femoral neck. All the ascending cervical arteries groups then converge to form a subsynovial ring on the femoral neck surface in the subcapital sulcus. From this ring, some vessels branch to the metaphysis to supply the capital secondary ossification centre, and eventually, the ossified femoral neck [1, 14]. Figure 2.12 shows the locations and connections of the arteries forming the extracapsular arterial ring.

 urgical Implications for the Femoral S Vascular Supply Avascular necrosis (AVN) of the femoral head can be a devastating iatrogenic complication following the treatment of different pediatric hip

2  Anatomy and Physiology of the Pediatric Hip

conditions. AVN is caused by disruption of the blood supply to the femoral head, resulting in osteonecrosis, or bone cell death. Therefore, knowledge of the femoral head vascular supply is critical when performing corrective or hip-­ preserving surgical procedures. The extracapsular ring comprised of retinacular branches of the MFCA and LFCA provide most of the bloody supply to both the proximal femur and greater trochanter [14]. These vessels anastomose in the piriformis fossa; consequently, due to their substantial vascular contribution, this anastomosis is particularly susceptible to insult during surgical drill entry into the femoral canal, or during femoral neck fracture [73, 74]. Gautier and colleagues later described key central (anterior or medial to the lesser trochanter) and peripheral (posterior or lateral to the lesser trochanter) anastomoses of the MFCA that may explain why certain cases of surgical, traumatic and non-traumatic hip dislocations develop AVN while others do not [67]. The most consistently present peripheral anastomosis was that of the deep branch of the MFCA and the IGA along the inferior border of the piriformis [67]. Grose and colleagues expanded upon the findings of Gautier and Ganz, identifying the anastomotic site between the MFCA and IGA to be extracapsular, immediately adjacent to the tendon of the obturator externus in the trochanteric fossa [75]. Their findings confirmed the postulation of Gautier and Ganz that IGA was capable of perfusing the femoral head in the absence of blood flow in the MFCA [67, 75]. During growth, the relative contribution of the MFCA to the overall femoral blood supply increases while that of the LFCA decreases [74]. Additionally, by 10 years of age, the number of vessels along the medial and anterior aspects of the femoral neck has reduced by 50%, presenting important considerations during femoral neck osteotomy procedures. Despite the recognized importance of intimate knowledge of the femoral head vascular supply for hip surgical procedures, there had been a lack of precise quantitative information regarding the capsular insertion and intracapsular course of vessels supplying the femoral head. Two recent studies, one using gross specimen dissection [70] and one using gadolinium-enhanced MR imaging

45

[76], have therefore examined the distribution of vascular foramina at the femoral head-neck junction in order to better define potential danger zones and prevent iatrogenic vascular damage during surgical interventions of the hip joint. The findings of these studies have provided a detailed, quantitative map of the arterial topographic anatomy of the femoral head-neck junction. The distribution of vascular foramina is particularly well-visualized when the femoral head is presented as a clock-face with the superior margin at 12 o’clock, anterior margin at 3 o’clock, inferior margin at 6 o’clock and posterior margin at 9 o’clock (Fig.  2.13). Lazaro and colleagues anatomically defined the quantitative locations of the insertions of the deep, ascending and transverse branches of the MFCA into the capsule, as well as the entry points into the articular rim of the femoral head [70]. The borders of the superior, inferior and anterior retinaculum of Weitbrecht (RW) were also defined. To completely preserve the vascular supply of the hip during surgical procedures, the posterior femoral capsular attachment and the superior and inferior RW should be maintained intact and thus should be considered vascular danger zones to avoid. Rego and colleagues also examined the topographic location of the vessels feeding the epiphysis, with particular regard to surgical procedures related to femoroacetabular impingement and cam lesion resection [76]. Using MR imaging techniques, this study found a predominance of arterial foramina from 10 to 12 o’clock, with the retinaculum extending from 1 to 10 o’clock (Fig.  2.13). Their findings suggest that critical areas for perfusion of the femoral head may overlap with the area of anterolateral cam deformity in FAI patients, and therefore presents a guide for the surgeon when selecting cuts for subcapital and intracapital osteotomies [76]. Specific safe zones for the depth of bone resection in these procedures were also elucidated. While both of these studies provide valuable insight to aid in the prevention of iatrogenic vascular comprise to the femoral head during surgical procedures about the hip, caution must be taken in assuming the generalizability of these results as they were based on 14 and 16 anatomical specimens, respectively. Additionally, the information gleaned in these investigations was from normal

E. K. Schaeffer and K. Mulpuri

46

Hip capsule 0° 270°

320

330

340

350

0

310

360° Scale

10 20

0

260

70

80

240

60

250

50

270

40

280

29

30

0

30

230

90

|

220

100

90°

0

0 21

11 12

0

20 0

130

140

150

160

170

180

190

180°

0

20

30

0

40

31

30

0

0

15

1

60 0 1

170 180 190 200

0 31 0 30

260 270 280 29 0 250 0

24

0

23

110

0

0

21

22

0 24

100

20

2

0

30

13

Mean borders of anterior retinaculum of weitbrecht (ARW) of 74-116˚ (35˚ to 135˚); ARW present in 64% (9/14) of hips

14

0

15

0

160

170

180

5

190

200

21 0

22

7 Mean borders of inferior retinaculum of weitbrecht (IRW) of 185-219˚ (166° to 245˚); IRW present in 100% of hips

0

Mean lesser trochanter location (216˚; 205˚ to 230˚)

4

Anterior

|

0

6

13

5

3

0

7

12

Mean entry point to the FHNJ 8 of transverse MFCA terminal 4 branches of 204˚ (145˚ to 244˚) [86% posterior, 14% anterior]

90

Mean entry point to the FHNJ 3 of ascending MFCA terminal branches of 246˚ (207˚ to 281˚)

9

14

290

40

Mean entry point to FHNJ of deep MFCA terminal branches 2 10 of 3390 (286˚ to 225˚) [80% posterior, 20% anterior]

80

280

30

1

|

270

20

90 100 1 10

260

10

12

70

11

0

350

80

250

340

70

8

33

60

0

0 32

2

60

Posterior

1

10

50

9

Mean insertion of the ascending branch of the MFCA at capsular attachment to the femur (244˚; 216˚ to 269˚)

Right Hip

50

10 Mean borders of superior retinaculum of weitbrecht (SRW) of 301-1° (2550 to 226°); SRW present

0

350

340

33

Anterior

12

11

0 32

Posterior

Superior

Mean insertion of deep branch of the MFCA at capsular attachment to the femur (327°; 310° to 355°)

6 Mean transverse branch (inferior retinacular artery) capsular insertion (177˚; 167˚ to 187˚)

Inferior

Fig. 2.13  Distribution of the vascular foramina at the femoral head-neck junction. The outer circle shows attachments of the femoral capsule and the inner circle shows the articular rim of the femoral head. MFCA = medial femoral

circumflex artery; FHNJ  =  femoral head-neck junction. Figure reproduced with permission of the Licensor through PLSclear from Lazaro et  al., 2015 [70] (PLS Reference Number 5670)

hips, and may not be entirely transferrable to the deformed hip undergoing surgery. Ganz’s work with surgical dissection studies has revolutionized the development of safe surgical dislocation of the femoral head techniques [67, 77–79]. Surgical dislocation of the femoral head

has a number of pathologic indications for use, including treatment for SCFE, femoral head reduction in Perthes disease, femoral head-neck osteochondroplasty for cam impingement, and for the fixation of acetabular fractures. Ganz’s improvements and modifications have helped sur-

2  Anatomy and Physiology of the Pediatric Hip

geons around the world to be more cognizant of the vascular supply when performing hip preservation procedures for a number of these different indications, resulting in a decrease in the incidence of AVN. However, it is vital to recognize that, even with optimized technique, there is a steep learning curve associated with these procedures that influences the rate of complications, including AVN. The importance of surgeon experience and comfort level with the surgical hip dislocation and other procedures involving the pediatric hip cannot be overstated. In support of this, a recent study reviewing proximal femoral varus derotation osteotomy in children with cerebral palsy showed that surgeon volume (number of procedures performed) was a strong predictor of surgical success [80]. Another study investigating the impact of surgeon experience on complications following periacetabular osteotomy did not find such an association, but did find a reduction in operative time with increasing surgeon experience [81]. Despite its importance, the definition, diagnosis, and classification of AVN remain controversial, with reported rates being highly variable across the literature. The identification and ­classification of AVN is confounded by a number of factors including severity, duration, and overall functional impact. Some cases of AVN are transient, and resolve spontaneously over time with little clinical consequence. However, some hips that have a mild and transient vascular insult early in development can have severe growth disturbances long-term, and factors differentiating these findings have not been delineated. More severe cases can result in permanent consequences, including hip pain due to arthrosis, limb length discrepancy, and gait disturbance; again however, not all cases of more pronounced and prolonged vascular disruption result in permanent functional consequences for the patient. Consequently, despite being a prominent iatrogenic complication associated with treatment of many pediatric hip conditions, a clearer understanding of AVN remains a critical outstanding issue. There is a large volume of literature now on AVN following hip reconstruction surgery for children with cerebral palsy. Potential relationships between the magnitude of surgery, amount

47

of surgical correction and certain pre-operative radiographic measures (NSA, Reimers Migration Percentage) have been suggested, although none of these have been substantially validated as true predictors or causal factors of AVN [82]. Similarly, for DDH, AVN is a frequently reported complication following both closed and open reduction; however, highly variable rates are reported across the literature, and true prognostic or predictive factors have yet to be firmly elucidated. Persistent low-level evidence has contributed to this issue, with numerous single-centre retrospective studies, small sample sizes in individual studies and a lack of a standardized definition or timeframe for AVN diagnosis limiting cross-study comparison. Surgeons need a detailed understanding of the vascular supply to the femoral head to help reduce incidence of AVN when treating pathologies of the pediatric hip.

Vascular Supply of the Acetabulum The blood supply to the acetabulum is completely established in the prenatal development phase and is comprised of two independent systems: a central axis originating from the acetabular artery that supplies the tri-radiate cartilage and a peripheral source involving the superior gluteal, inferior gluteal, internal pudendal and obturator arteries, the latter system forming a periacetabular vascular circle, or extra-articular peripheral ring [83]. The blood supply originating from the acetabular artery is formed by three major branches penetrating the cartilaginous nucleus at the centre of the acetabulum. The superior branch originates from the superior pedicle of the ligamentum teres and extends to the primary ossification centre of the ilium [1]. The antero-inferior branch originates from the artery of the ligamentum teres and extends to the superolateral margin of the pubis. The posteroinferior pedicle arises from the posterior branch of the artery of the ligamentum teres and extends to the ischial ossification centre [1]. In the peripheral system, the superior gluteal artery (SGA) arises from the posterior division of the internal iliac artery, and supplies the supe-

48

rior part of the acetabulum and the weight-bearing dome [1, 14]. The SGA is located in the intermuscular plane of the gluteus medius and minimus, tracking along the upper rim of the acetabulum toward the anterior superior iliac spine. The posterior branch of the obturator artery supplies most of the inferior acetabulum while the inferior gluteal artery (IGA) supplies the posterior acetabulum. The inferior gluteal artery has also been described to contribute to the blood supply of the femoral head through an anastomosis with the MFCA adjacent to the tendon of obturator externus in the trochanteric fossa [75]. On the anterior acetabulum, the SGA, the pubic branch of the obturator artery, and the ascending branch of the LFCA anastomose, provide collateralization between the different arterial systems. Similar to the clock face analogy used described for the femoral head, one can also picture the right acetabulum as a clock face to better visualize the locations of the vessels providing the acetabular blood supply. By convention, the transverse acetabular notch is commonly defined as the 6 o’clock position. With this image in mind, the superior and inferior gluteal arteries exit the pelvis at the greater sciatic notch (10 o’clock). These arteries diverge and supply the anterosuperior acetabulum from 10 to 4 o’clock, and the posterior acetabulum from 8 to 10 o’clock, respectively. At 6 o’clock, the obturator artery exits the pelvis via the obturator foramen and supplies the inferior acetabulum from 4 to 8 o’clock [84].

 urgical Implications for the Vascular S Supply of the Acetabulum The incidence of AVN of the acetabulum is very low, both from idiopathic and iatrogenic causes. Consequently, much of the surgical literature has focused on the vascular supply to the proximal femur. However, AVN of the acetabular fragment has been reported following rotational periacetabular osteotomies in adults and with Bernese periacetabular osteotomy (PAO) when the inferior cut is intra-articular, disrupting the acetabular branch of the obturator artery [84]. In 1992,

E. K. Schaeffer and K. Mulpuri

Damsin and colleagues defined the two independent arterial supply systems involved in acetabular vascularization, and assessed the risks of necrosis following PAO in childhood [85]. In their study, the acetabular fragments remained perfused following sectioning of the ilium just above the acetabulum and the ischial and iliopubic rami at their median portion. The authors did, however, note a susceptible zone at the anterior inferior iliac spine [85]. This particular zone had poor distribution of arterial branches from the anastomotic arch between the artery of the roof, the IGA, internal pudendal and obturator artery. The authors suggest that this poor distribution could contribute to defective development of the outer part of the acetabular roof, a growth disorder that can result following PAO [85]. In 2003, Beck and colleagues produced similar findings when performing the Bernese PAO [84]. Utilizing a modified Smith-Petersen approach, whereby all cuts are performed from within the pelvis, the acetabular fragment remains vascularized by the supra-acetabular and acetabular branches of the SGA, IGA and obturator artery. The degree of correction tolerated by the smaller blood vessels, however, is not clearly defined. Hempfing and colleagues reported a significant reduction in blood flow to the supra-acetabular region during the initial cuts of a Bernese PAO, secondary to disruption to the iliolumbar vasculature [86]. This reduction in blood flow proved to be reversible, however, provided fixation was placed within 30  min of the initial disruption. These findings prompted the authors to suggest ­leaving a bone bridge of 2–2.5 cm in the supra-acetabular region during periacetabular procedures such as the Bernese PAO in order to enable continued perfusion of the area [86]. Additionally, Seeley and colleagues note caution regarding the Hardinge direct lateral approach to the hip, primarily indicated in total hip arthroplasty and proximal femur fractures [83]. This approach risks injury to the superior gluteal vascular pedicle, which may result in abductor weakness and limping. Overall, far fewer studies have reported on the impact of the vascular supply of the acetabulum than the femoral head during surgical hip procedures. The majority of the surgeon’s focus

2  Anatomy and Physiology of the Pediatric Hip

in preoperative planning and during the procedure itself is on avoiding vascular insult to the femoral head. However, much like in the anatomical development of hip joint geometry, it may be important to consider the impact of the vasculature on both the femoral head and acetabulum in parallel rather than as separate entities due to the intimate relationship between them. Vascular insult to the proximal femur, and the resulting changes in femoral head geometry, have been shown to have an influence on acetabular morphology. In a radiological review of 155 children with Perthes disease, Joseph reported consistent acetabular changes throughout the course of Perthes disease that can influence the final outcome, including increased Sharp’s angle, acetabular radius, acetabular depth and ilium height, and persistent irregularity of the acetabular contour (Chap. 6) [86]. It was difficult to ascertain, h­owever, whether these changes could be attributed to vascular disruption in the acetabulum or vascular disruption in the femoral head. More likely, the changes were due to contributions from both sources. It remains unclear if an isolated vascular insult to the acetabulum would have an impact on a wellperfused femoral head. Consequently, a surgeon must approach preoperative planning and the surgical procedure with appropriate consideration and understanding of the anatomy and vasculature of both the acetabulum and the femoral head to minimize potential for complications. The orthopaedic surgeon must not underestimate the importance of careful consideration of the anatomy and physiology of the femoral head and acetabulum when planning and performing surgical hip procedures in order to optimize functional outcomes and mitigate potential long-term debilitating complications.

References 1. Chung SMK.  Hip disorders in infants and children. Philadelphia, PA: Lea and Febiger; 1981. 2. Strayer LM. The embryology of the human hip joint. Yale J Biol Med. 1943;16:13–26. 3. Delaere O, Dhem A.  Prenatal development of the human pelvis and acetabulum. Acta Orthop Belg. 1999;65(3):255–60.

49 4. Watanabe RS.  Embryology of the human hip. Clin Orthop. 1974;98:8–26. 5. Drachman DB, Sokoloff L.  The role of movement in embryonic joint development. Dev Biol. 1966;14(3):401–20. 6. Bowen JR, Kotzias-Neto A. Developmental dysplasia of the hip. 5th ed. Brooklandville, MD: Data Trace Publishing Company; 2006. 7. Lee MC, Eberson CP.  Growth and development of the child’s hip. Orthop Clin North Am. 2006;37: 119–32. 8. Jouve JL, Glard Y, Garron E.  Anatomical study of the proximal femur in the fetus. J Bone Joint Surg Br. 2005;14:105–10. 9. Ralis Z, McKibbin B.  Changes in the shape of the human hip joint during its development and their relation to its stability. J Bone Joint Surg Br. 1973;55(4):780–5. 10. Ulloa I. Embryonic and fetal development of the vascular system of the proximal end of the femur and acetabulum in man. Z Orthop. 1962;96:306–23. 11. Noback CR, Robertson GG.  Sequences of appearance of ossification centers in the human skeleton during the first five prenatal months. Am J Anat. 1951;89(1):1–28. 12. Weinstein SL, Mubarak SJ.  Developmental hip dysplasia and dislocation: part I.  AAOS Instr Course Lect. 2004;53:523–30. 13. Herring JA.  Tachdjian’s pediatric orthopaedics e-book. Texas Scottish Rite Hospital. Amsterdam: Elsevier Health Sciences; 2013. 14. Chung SMK.  The arterial supply of the developing proximal end of the human femur. J Bone Joint Surg Am. 1976;58(7):961–70. 15. Laage H, et  al. Horizontal lateral roentgenogra phy of the hip in children. J Bone Joint Surg Am. 1953;35(2):387–98. 16. Birkenmaier C, Jorysz G, Jansson V, Heimkes B.  Normal development of the hip: a geometrical analysis based on planimetric radiography. J Pediatr Orthop B. 2010;19(1):1–8. 17. Emmett J.  Measurements of the acetabulum. Clin Orthop. 1967;53:171–4. 18. Wakeley CPG. A note on the architecture of the ilium. J Anat. 1929;64:109. 19. Laurenson RD.  Development of the acetabular roof in the fetal hip. J Bone Joint Surg Am. 1965;47(5):975–83. 20. Cheynel J, Huet R. Studies of the pathological physiology of the hip. Rev Chir Orthop Reparatrice Appar Mot. 1952;38(3):279–86. 21. Whyte G, Sink EL, Coleman SH. The effects of acetabular depth on clinical outcomes after hip arthroscopy in patients with femoroacetabular impingement. J Hip Preserv Surg. 2016;3(S1):S19–20. 22. Reikeras O, Hinderaker T, Steen H.  Reduced acetabular depth in hip instability in the newborn. Orthopedics. 1999;22(10):943–6. 23. Loder RT, Mehbod AA, Meyer C, Meisterling M.  Acetabular depth and race in young adults: a potential explanation of the differences in the

50 prevalence of slipped capital femoral epiphysis between different racial groups? J Pediatr Orthop. 2003;23(6):699–702. 24. Dega W.  Anatomical and mechanical study of the fetal hip for the purpose of elucidating the etiology and pathogenesis of congenital dislocation. Chir Organi Mov. 1933;18(5):425. 25. Bost FC, Hagey H.  The results of treatment of congenital dislocation of the hip in infancy. J Bone Joint Surg Am. 1968;30(2):454–68. 26. Salter RB. Innominate osteotomy in the treatment of congenital dislocation and subluxation of the hip. J Bone Joint Surg Br. 1961;43:518–39. 27. Dwight T. The size of the articular surfaces of the long bones as characteristic of sex. Am J Anat. 1905;4:19–31. 28. Wanner MR, Loder RT, Jennings SG, Ouyang F, Karmazyn B.  Changes in femoral head size and growth rate in young children with severe developmental dysplasia of the hip. Pediatr Radiol. 2017;47(13):1787–92. 29. Rosenberg MR, Walton R, Rae EA, Bailey S, Nicol RO.  Intra-articular dysplasia of the femoral head in developmental dysplasia of the hip. J Pediatr Orthop B. 2017;26(4):298–302. 30. Blowers DH, Elson R, Korley E. An investigation of the sphericity of the human femoral head. Med Biol Eng. 1972;10:762–75. 31. Hammond BT, Charnley J. The sphericity of the femoral head. Med Biol Eng. 1967;5:445–53. 32. Okano K, Enomoto H, Osaki M, Takahashi K, Shindo H.  Femoral head deformity after open reduction by Ludloff’s medial approach. Clin Orthop Relat Res. 2008;466:2507–12. 33. Sankar WN, Neubuerger CO, Moseley CF.  Femoral head sphericity in untreated developmental dislocation of the hip. J Pediatr Orthop. 2010;30:558–61. 34. Bobroff ED, Chambers HG, Saroris DJ. Femoral anteversion and neck-shaft angle in children with cerebral palsy. Clin Orthop Relat Res. 1999;364:194–204. 35. Walmsley T. The neck of the femur as a static problem. J Anat Physiol. 1915;49:314–35. 36. Boese CK, Dargel J, Oppermann J, Eysel P, Scheyerer MJ, Bredow J, Lechler P.  The femoral neck-shaft angle on plain radiographs: a systematic review. Skeletal Radiol. 2016;45(1):19–28. 37. Walton NP, Wynn-Jones H, Ward MS, Wimhurst JA. Femoral neck-shaft angle in extra-capsular proximal femoral fracture fixation; does it make a TAD of difference? Injury. 2005;36(11):1361–4. 38. Fabry G, MacEwen GD, Shands AR.  Torsion of the femur: a follow-up study in normal and abnormal conditions. J Bone Joint Surg Am. 1973;55:1726–38. 39. Lloyd-Roberts GC.  Orthopaedics in infancy and childhood. London: Butterworths; 1971. 40. Sankar WN, Neubuerger CO, Moseley CF.  Femoral anteversion in developmental dysplasia of the hip. J Pediatr Orthop. 2009;29(8):885–8. 41. Lagenskiold A, Salenius P.  Epiphyseodesis of the greater trochanter. Acta Orhop Scand. 1967;38: 199–219.

E. K. Schaeffer and K. Mulpuri 42. Neumann DA.  Kinesiology of the hip: a focus on muscular actions. J Orthop Sports Phys Ther. 2010;40(2):82–94. 43. Netter FH.  Atlas of human anatomy. 4th ed. Philadelphia, PA: Saunders Elsevier; 2006. 44. Giorgi M, Carriero A, Shefelbine S. Effects of normal and abnormal loading conditions on morphogenesis of the prenatal hip joint: application to hip dysplasia. J Biomech. 2015;48:3390–7. 45. Carter D, Orr T, Fyhrie D. Influences of mechanical stress on prenatal and postnatal skeletal development. Clin Orthop Relat Res. 1987;219:237–50. 46. Sharir A, Stern T, Rot C. Muscle force regulates bone shaping for optimal load-bearing capacity during embryogenesis. Development. 2011;138:3247–59. 47. Burr D. Muscle strength, bone mass and age-related bone loss. J Bone Miner Res. 1997;12:1547–51. 48. Ford CA, Nowlan NC, Thomopoulos S, Killian ML.  Effects of imbalanced muscle loading on hip joint development and maturation. J Orthop Res. 2017;35:1128–36. 49. Bullough P. The role of joint architecture in the etiology of arthritis. Osteoarthr Cartil. 2004;12:2–9. 50. Onyemaechi NOC, Anyanwu EG, Obikili E, Ekezie J. Anatomical basis for surgical approaches to the hip. Ann Med Health Sci Res. 2014;4(4):487–94. 51. Hansen AD.  Anatomy and surgical approaches. In: Morrey BF, editor. Reconstructive surgery of the joints. 2nd edn. New York, NY: Churchill Livingstone; 1996. 52. McFarland B, Osborne G.  Approach to the hip: a suggested improvement on Kocher’s method. J Bone Joint Surg Br. 1954;36B:364. 53. Hardinge K. The direct lateral approach to the hip. J Bone Joint Surg Br. 1982;64:17–9. 54. Thieme. Atlas of anatomy. New  York, NY: Thieme; 2006. 55. Babst D, Steppacher SD, Ganz R, Siebenrock KA, Tannast M.  The iliocapsularis muscle: an important stabilizer in the dysplastic hip. Clin Orthop Relat Res. 2011;469:1728–34. 56. Dora C, Houweling M, Koch P, Sierra RJ. Iliopsoas impingement after total hip replacement: the results of non-operative management, tenotomy or acetabular revision. J Bone Joint Surg Br. 2007;89:1031–5. 57. Ganz R, Klaue K, Vinh TS, Mast JW. A new periacetabular osteotomy for the treatment of hip dysplasias: technique and preliminary results. Clin Orthop Relat Res. 1988;232:26–36. 58. Ward WT, Fleisch ID, Ganz R. Anatomy of the iliocapsularis muscle: relevance to surgery of the hip. Clin Orthop Relat Res. 2000;374:278–85. 59. Charnley J, Ferreiraade S.  Transplantation of the greater trochanter in arthroplasty of the hip. J Bone Joint Surg Br. 1964;46:191–7. 60. Masse A, Aprato A, Rollero L, Bersano A, Ganz R.  Surgical dislocation technique for the treatment of acetabular fractures. Clin Orthop Relat Res. 2013;471(12):4056–64.

2  Anatomy and Physiology of the Pediatric Hip 61. Testa NN, Mazur KU.  Heterotopic ossification after direct lateral approach and transtrochanteric approach to the hip. Orthop Rev. 1988;17:965–71. 62. Gaskill TR, Philippon MJ.  Surgical hip dislocation for femoroacetabular impingement. Am J Sports Med. 2012;40:NP1–2. 63. Koizumi W, Moriya H, Tsuchiya K, Takeuchi T, Kamegaya M, Akita T.  Ludloff’s medial approach for open reduction of congenital dislocation of the hip. A 20-year follow-up. J Bone Joint Surg Br. 1996;78:924–9. 64. Ferguson AB Jr. Primary open reduction of congenital dislocation of the hip using a median adductor approach. J Bone Joint Surg Am. 1973;55:671–89. 65. Zlotorowicz M, Czubak J.  Vascular anatomy and blood supply to the femoral head. Osteonecrosis. 2014;2:19–25. 66. Trueta J.  The normal vascular anatomy of the femoral head in adult man. Clin Othop Relat Res. 1997;334:6–14. 67. Gautier E, Ganz K, Krugel N, Gill T, Ganz R. Anatomy of the medial femoral circumflex artery and its surgical implication. J Bone Joint Surg Br. 2000;82(5):679–83. 68. Kalhor M, Beck M, Thomas WH, Ganz R. Capsular and pericapsular contributions to acetabular and femoral head perfusion. J Bone Joint Surg Am. 2009;91:409–18. 69. Senior HD.  The development of the arteries of the human lower extremity. Am J Anat. 1919;25(1):55–95. 70. Lazaro LE, Klinger CE, Sculco PK, Helfet DL, Lorich DG.  The terminal branches of the medial femoral circumflex artery: the arterial supply of the femoral head. Bone Joint J. 2015;97-B(9):1204–13. 71. Boraiah S, Dyke JP, Hettrich C, Parker RJ, Miller A, Helfet DL, Lorich DG.  Assessment of vascularity of the femoral head using gadolinium (Gd-DTPA)enhanced magnetic resonance imaging: a cadaver study. J Bone Joint Surg Br. 2009;91-B:131–7. 72. Crock HV. An atlas of the arterial supply of the head and neck of the femur in man. Clin Orthop Relat Res. 1980;152:17–27. 73. Zlotorowicz M, Szczodry M, Czubak J, Ciszek K. Anatomy of the medial femoral circumflex artery with respect to the vascularity of the femoral head. J Bone Joint Surg Br. 2011;93-B:1471–4. 74. Schottel PC, Hinds RM, Lazaro LE, et  al. The effect of antegrade femoral nailing on femoral head perfusion: a comparison of piriformis fossa and trochanteric entry points. Arch Orthop Trauma Surg. 2015;135:473–80. 75. Grose AW, Gardner MJ, Sussmann PS, Helfet DL, Lorich DG. The surgical anatomy of the blood supply

51 to the femoral head: description of the anastomosis between the medial femoral circumflex and inferior gluteal arteries at the hip. J Bone Joint Surg Br. 2008;90(1):1298–303. 76. Rego P, Mascarenhas V, Collado D, Coelho A, Barbosa L, Ganz R.  Arterial topographic anatomy near the femoral head-neck perforation with surgical relevance. J Bone Joint Surg Am. 2017;99:1213–21. 77. Ganz R, Slongo T, Siebenrock KA, Turchetto L, Leunig M.  Surgical technique: the capsular arthroplasty: a useful but abandoned procedure for young patients with developmental dysplasia of the hip. Clin Orthop Relat Res. 2012;470(11):2957–67. 78. Kalhor M, Gharehdaghi J, Schoeniger R, Ganz R. Reducing the risk of nerve injury during Bernese periacetabular osteotomy: a cadaveric study. Bone Joint J. 2015;97-B(5):636–41. 79. Shore BJ, Zurakowski D, Dufreny C, Powell D, Matheney TH, Snyder BD. Proximal femoral varus derotation osteotomy in children with cerebral palsy: the effect of age, gross motor function classification system level, and surgeon volume on surgical success. J Bone Joint Surg Am. 2015;97(24): 2024–31. 80. Novais EN, Carry PM, Kestel LA, Ketterman B, Brusalis CM, Sankar WN.  Does surgeon experience impact the risk of complication after Bernese Periacetabular osteotomy? Clin Orthop Relat Res. 2017;475(4):1110–7. 81. Hesketh K, Leveille L, Mulpuri K. The frequency of AVN following reconstructive hip surgery in children with cerebral palsy: a systematic review. J Pediatr Orthop. 2016;36(2):e17–24. 82. Seeley MA, Georgiadis AG, Sankar WN. Hip vascularity: a review of the anatomy and clinical implications. J Am Acad Orthop Surg. 2016;24(8):515–26. 83. Beck M, Leunig M, Ellis T, et al. The acetabular blood supply: implications for periacetabular osteotomies. Surg Radiol Anat. 2003;25:361–7. 84. Damsin JP, Lazennec JY, Gonzales M, Guerin-Surville H, Hannoun L.  Arterial supply of the acetabulum in the fetus: application to periacetabular surgery in childhood. Surg Radiol Anat. 1992;14(3):215–21. 85. Hempfing A, Leunig M, Notzli HP, Beck M, Ganz R.  Acetabular blood flow during Bernese periacetabular osteotomy: an intraoperative study using laser Doppler flowmetry. J Orthop Res. 2003;21(6):1145–50. 86. Joseph B.  Morphological changes in the acetabu lum in Perthes’ disease. J Bone Joint Surg Br. 1989;71(5):756–63.

3

Biomechanics of the Hip During Gait Morgan Sangeux

Introduction Human bipedalism has assigned a major role to the hip joint, as the connection between the upper and lower parts of the body. The hip joint needs to confer both stability and mobility and its anatomical features reflect these two roles. As a ball and socket joint, the hip allows rotation in the three anatomical planes. The complementary shapes of the femur and the acetabulum, together with strong extra-capsular ligaments, provide passive stability in the erect posture. In addition, the hip is surrounded by uni- and bi-articular muscles able to generate movement, while at the same time stabilizing the joint under load. Some of the largest muscles in the human body cross the hip joint, most notably the gluteals and hamstrings. The muscle forces they generate are the predominant contributors to hip contact forces during activities of daily living, including during gait. In this chapter, the mechanics of how muscle forces and bony levers combine at the hip joint M. Sangeux (*) The Royal Children’s Hospital, Melbourne, VIC, Australia The Murdoch Children’s Research Institute, Melbourne, VIC, Australia The University of Melbourne, Melbourne, VIC, Australia Biomech-Intel, Marseille, France

© Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_3

to achieve movement will be explored. The chapter starts with the functional anatomy of the hip, which details variations in bone shape and the potential effect on muscle lever arms. A  more detailed account of the anatomy and physiology of the pediatric hip can be found in Chap. 2. The second section is dedicated to the biomechanics of the hip during gait. It details kinematics and kinetics modelling of the pelvis and the hip, describes the normal gait pattern and the utility of gait analysis to investigate hip pathologies. Additionally, hip joint contact forces during walking and other activities of daily living are discussed. Throughout the chapter, we use musculoskeletal modelling to illustrate the concepts, or compute theoretical results. The normative data for the gait patterns (kinematics and kinetics) illustrated in this chapter were captured on typically developing children in our gait laboratory.

Functional Anatomy This section focuses on the relationship between shape and function and describes the biomechanical role of key structures pertaining to the pediatric hip. We do not intend to provide a comprehensive overview of the anatomy of the hip joint (see Chap. 2), rather to focus on the structures with the most impact on its biomechanics.

53

M. Sangeux

54

Bony Structures The shape of the acetabulum and the femoral head are complementary. The two structures form a congruent ball (the femoral head) and socket (the acetabulum) joint. The articular surfaces of the acetabulum and femoral head are covered with hyaline cartilage and the joint contains synovial fluid; both act to reduce friction. Several geometric parameters describe the shape of the acetabulum and the femur. For example, the centre edge angle (lateral, or anterior), as well as the acetabular depth and coxa profunda [25], quantify the amount of coverage the acetabulum offers to the femoral head (See Chap. 2). The orientation of the acetabulum cup within the pelvis may be described with the Acetabulum Version (AV) and Acetabulum Inclination (AI) angles (Fig. 3.1). These parameters describe the amount of coverage the acetabulum offers to the femoral head, but also how this

coverage changes for different positions of the hip joint. The shape of the femur may be defined using three axes. The condylar axis defines the medio-­lateral axis of the distal femur, the neck axis defines the medio-lateral axis of the proximal femur, and the shaft of the femur defines the longitudinal axis. The angle between the shaft and neck axes is called Neck Shaft Angle (NSA) or Centrum-Collum-Diaphyseal angle (CCD), and the angle between the condylar and neck axes, in the plane perpendicular to the shaft axis, is called Femoral Neck Anteversion (FNA) (Fig. 3.2). NSA and FNA are both determinants of the percentage of the femoral head covered by the acetabulum for a given position of the hip joint. These geometric parameters also influence the lines of action of the peri-articular muscles (most notably, the hip abductors), and their capacity to produce a moment in a given anatomic plane.

AV AI Coronal view

Fig. 3.1  Anatomical definition of acetabular inclination (AI) and acetabular version (AV). The orientation of the acetabular rim plane is denoted in red, and the pelvis transverse plane (coronal view) and the pelvis sagittal plane (transverse view) are denoted in black. Interestingly, the

Transverse view

measurement is sometime described from the perpendicular to the rim plane (in dashed red) and the pelvis sagittal plane (coronal view) and coronal plane (transverse view) (in dashed black). The picture was created using 3DSlicer [14] and CT-scan data from Hara and colleagues [17]

3  Biomechanics of the Hip During Gait

55

children aged 4–17 years old, and utilising low-dose bi-plane X-ray measurements, did not find a significant trend for change in neck shaft angle with age [45], finding the average NSA to be 130° (SD: 5°). “For children aged 4 to 17 years, femoral neck shaft angle has not been shown to change with age”

NSA

FNA

Coronal view

Transverse view

Fig. 3.2  The neck shaft angle (NSA) and femoral neck anteversion angle (FNA) are presented in the coronal and transverse views respectively. The axis of the femoral neck is denoted in red, the axis of the most posterior aspects of the condyles in green and the axis of the shaft of the femur in blue. The picture was created using 3DSlicer [14] and CT-scan data from Hara and colleagues [17]

Reports on the change in acetabular version and inclination with age are scarce, although recent studies found that acetabular version was smaller in skeletally immature children [2, 20]. Average values for acetabular version in normal adult populations were reported to be a mean of 20° (range: 6–33°) [50] and 23° (SD: 7°) [19] anteverted (pointing anteriorly) for Chinese and American samples, respectively. Average values for acetabular inclination were reported to be a mean of 54° (range: 46–60°) and 56° (SD: 4°) for these samples, respectively [19, 50]. Mild sexual dimorphism was measured with mean adult female values increased by around 2° for acetabular version and inclination [19, 50]. In adults, the average neck shaft angle is just under 130° [9, 16]. Increased values for neck shaft angle are otherwise known as coxa valga, and decreased values as coxa vara. A recent study in 508

The average femoral neck anteversion is approximately 10–15° [23, 46]. When the FNA is negative (i.e. the condylar axis is rotated externally with respect to the neck axis) the femur is said to be retroverted. There is some evidence that FNA tends to decrease with age [21, 43]. However, these findings were based on physical examination measurements rather than medical imaging and FNA measurements between medical imaging and physical examination have been shown to differ [10, 34]. Currently, the gold standard method to measure acetabular AI and AV, and femoral NSA and FNA, is to use axial medical imaging, for example computed tomography (CT) or magnetic resonance imaging (MRI). These modalities provide native 3-dimensional information and thus avoid the potential difficulties of trying to derive these values using 2-dimensional projections. However, some measurement protocols do not fully utilise the 3-dimensional information. For example, it is common practice to measure FNA by superimposing the neck and condylar axes as seen in 2D axial sections of the femur, i.e. perpendicular to the axis of the scanner table. Regrettably, the portion of the neck visible in these 2D sections may be reduced, and even more so if NSA is increased (coxa valga). In addition, and in the presence of severe hip flexion contracture, the axis of the femur shaft may be quite different than the axis of the scanner table. It is important to note that these limitations are due to the measurement protocol, rather than due to the medical images. Medical imaging software such as 3DSlicer [14] allows visualization of the volumetric image in all directions. The possibility to reslice (i.e. create images in planes with arbitrary orientation, different from the axial plane) allows for the implementation of 3D measurement protocols from standard axial ­ imaging, e.g. Fig. 3.3 and [36].

M. Sangeux

56 Oblique slice Coronal slice

when the hip is in extension and slackened in flexion. As a result, the ligaments ensure passive resistance to hip hyperextension [26]. The ligaments also restrict internal or external rotation towards the end of the flexion-extension range [49].

Muscles and Moments

Transverse slices

Coronal slice

Fig. 3.3  Measuring femoral neck anteversion and neck shaft angles using 3D information from CT-scans. Adapted from [36] with permission. Points correspond to HJC (Hip Joint Centre, centre of the femoral head), GT (Greater Trochanter, intersection of the neck axis with the greater trochanter), PS (Proximal Shaft, centre of the diaphysis at the level of the lesser trochanter), DS (Distal Shaft, centre of the intercondylar fossa), MC/LC (Medial/Lateral Condyles, most distal aspects of the femur condyles)

Passive Soft Tissues The stability of the hip joint is strengthened by the surrounding passive soft tissues. Passive structures with major biomechanical roles include the labrum, capsule and three extracapsular ligaments. The labrum is a fibrous structure attached to the rim of the acetabulum (see Chap. 2). Mechanically, the labrum not only deepens the acetabulum but may facilitate a ‘suction effect’ on the femoral head [11]. The iliofemoral, pubofemoral and ischiofemoral ligaments connect the bones of the pelvis (respectively the iliac, the pubis and the ischium) to the proximal femur, at the level of the intertrochanteric line (see Chap. 2). The three ligaments wrap around the neck of the femur and are stretched

Forces produced by muscle contractions are converted to moments at the joint by bony levers. The moment is defined as the force (muscle action) multiplied by the distance of force application (the lever, or moment, arm) from the fulcrum of movement. In other words, the capacity of a muscle to generate a moment is directly proportional to the length of its lever arm with respect to the joint centre of rotation. Muscle lever arms are expressed in the anatomical planes, to facilitate clinical interpretation; for example, to identify muscles or muscle groups responsible for producing hip flexion moment vs. muscles responsible for producing hip abduction moment (Fig. 3.4). For the hip, the geometric properties of the proximal femur, particularly with respect to the axial (i.e. FNA) and coronal (i.e. NSA) planes, are the primary determinants of muscle lever arms. “Forces produced by muscle contractions are converted to moments at the joint by bony levers”. The moments associated with the primary periarticular muscles, or muscle groups, responsible for joint motion in the flexion-extension, abduction-adduction, and internal-external rotation planes are shown in Figs. 3.5 and 3.6. Since the force and lever arm of a muscle is dependent on joint position (see Fig.  3.4), we plotted the moment and lever arm of key muscles during hip flexion. The graphs in Figs. 3.5 and 3.6 highlight that some muscles have a reversed role ­depending on the position of the hip in the sagittal plane. For example, the gluteus medius is an abductor of the hip throughout the range of motion in the sagittal plane but, depending on the degree of flexion, it may act as either an external or an internal rotator (in flexion and extension, respectively). Similarly,

3  Biomechanics of the Hip During Gait

57

FP FGM Psoas

Gluteus medius anterior fibres

dGM = 0

dGM

dP

dP = 0

Psoas abduction moment = 0

Gluteus medius flexion moment = 0

Coronal view

Sagittal view

Fig. 3.4  The hip in 65° flexion in the coronal (left) and sagittal (right) planes. The anterior fibres of the gluteus medius can produce an abduction moment but 0 flexion moment because its lever arm in the sagittal plane is 0, no

matter the force FGM. The situation is opposite for the psoas muscle; its lever arm in the coronal plane is 0 and cannot produce an abduction moment, but does produce a flexion moment in the sagittal plane

Hip flexion(+)/extension(-)

100

100

Moment (Nm)

Moment (Nm)

50 0 –50 –100 –150 –200 –20

Moment (Nm)

100

0

60 20 40 Hip flexion (°)

80

100

50 0 –50

–100 –20

0

20 40 60 Hip flexion (°)

80

100

Hip internal(+)/extension(-) rotation Gluteus maximus Hamstrings Gluteus medius Tensor fascia latae Adductor magnus Other adductors Piriformis Iliopsoas Rectus femoris

50

0

–50 –20

Hip abduction(+)/adduction(-)

0

60 20 40 Hip flexion (°)

80

100

Fig. 3.5  Maximum moment generated by the periarticular hip muscles in the three anatomical planes, as a function of hip flexion. The graphs highlight that the capacity of the muscles to generate a moment at the hip may depend on the position of the hip joint. Different muscles, or different fibres of the same muscle, were grouped according to their primary function for clarity (see Table 3.1). The graphs were produced using OpenSim 3.3,

and the generic model provided to study gait [12]. The data used in the model were obtained from several cadaveric experiments using multiple subjects [53]. The values obtained for the moment and lever arm in the figures above are representative of a generic adult measuring 1.70  m and weighing 75  kg. Femoral neck anteversion (FNA) and neck shaft angle (NSA) in the generic model were measured at 18° and 125°, respectively

M. Sangeux

58 Hip flexion(+)/extension(-)

5 0 –5 –10 –15 –20 –20

Moment arm (cm)

15

0

60 20 40 Hip flexion (°)

80

Hip abduction(+)/adduction(-)

20 Moment arm (cm)

Moment arm (cm)

10

10 0 –10 –20 –20

100

0

80

100

Hip internal(+)/external(-) rotation Gluteus maximus Hamstrings Gluteus medius Tensor fascia latae Adductor magnus Other adductors Piriformis Iliopsoas Rectus femoris

10 5 0 –5 –10 –15 –20

20 40 60 Hip flexion (°)

0

20 40 60 Hip flexion (°)

80

100

Fig. 3.6  Moment-arm of the periarticular hip muscles in the three anatomical planes, as a function of hip flexion. The graphs highlight that the lever arm of the muscles may vary with the position of the hip joint. Different muscles, or different fibres of the same muscle, were grouped for clarity (see Table 3.1). The data used in the model were obtained

from several cadaveric experiments using multiple subjects [53]. The values obtained for the moment and lever arm in the figures above are representative of a generic adult measuring 1.70 m and weighing 75 kg. Femoral neck anteversion (FNA) and neck shaft angle (NSA) in the generic model were measured at 18° and 125°, respectively

the adductors may produce an extension moment when the hip is in extension, and flexion moment when the hip is in flexion. Functional roles for each of the periarticular muscles/muscle groups acting on the hip are described in Table 3.1.

Table 3.1  Functional roles for each of the periarticular muscles/muscle groups acting on the hip Legend Gluteus maximus Hamstrings

Main role Hip extensor, external rotator Semimembranosus Hip extensor, adductor Semitendinosus Biceps femoris Gluteus All fibres Hip abductor, medius extensor, and internal/external rotatora Gluteus All fibres Hip abductor, flexor minimus and internal/ external rotatora Tensor fascia Hip abductor latae Adductor All fibres Hip adductor, magnus extensor Hip adductor, Other Adductor brevis flexor/extensora adductors Adductor longus Gracilis Piriformis Hip external rotator Iliopsoas Iliacus Hip flexor Psoas Rectus Hip flexor femoris

“Some muscles have a reversed role depending on the position of the hip in the sagittal plane”. Musculoskeletal models are required to calculate the muscle lever arms during complex movements. Most models assume the action of a muscle may be represented through a single line of action, and the muscle force-length (and velocity) is obtained using a Hill type model [47]. An advantage of musculoskeletal modelling is the ability to explore how the lever arms of the muscles evolve when the bony structures are deformed. For example, Fig.  3.7 was produced using a deformed model with either increased femoral neck anteversion (increased by 30° compares to the generic model, i.e. FNA of 48° instead of 18°) or increased femoral neck shaft angle (increased by 20° compares to the generic

Grouping All fibres

Depending on hip flexion

a

3  Biomechanics of the Hip During Gait

59

Hip flexion(+)/extension(–)

Hip abduction(+)/adduction(–) 15 Moment arm (cm)

Moment arm (cm)

10 0 –10 –20 –30 –20

0

20

40

60

80

100

Hip flexion (°)

Moment arm (cm)

15

10 5 0 –5 –10 –20

0

20

40

60

80

100

Hip flexion (°)

Hip internal(+)/external(–) rotation Gluteus maximus Gluteus medius Iliopsoas

10 5 0

Increased FNA Increased NSA

–5 –10 –15 –20

0

20

40

60

80

100

Hip flexion (°)

Fig. 3.7  Effect of increased femoral neck anteversion (FNA: 48°, dotted line) and increased neck shaft angle (NSA: 145°, dashed line) on the lever arm of the gluteus

maximus, gluteus medius, and the iliopsoas compared to normal (solid) FNA (18°) and NSA (125°)

model, i.e. NSA of 145° instead of 125° in the original generic model). The graph highlights the major impact of FNA and NSA to the abduction lever arm of the gluteus medius and therefore its capacity to produce an abduction moment.

right step followed by a left step. A right stride commences when the right foot touches the ground and combines a left step followed by a right step. Average left and right steps may have different lengths if the gait is asymmetrical, but average left and right strides must have the same length, as long as the subject is walking in a straight line. The stride may also be split into the stance and swing phases. The stance phase is the period where the foot is in contact with the ground, typically the first 60% of the total stride. The swing phase is the period where the same foot is off the ground (but the contralateral foot is on the ground), “swinging” through to move the limb forward towards foot contact (typically 40% of the total stride length) (Fig. 3.8). The stance phase may be further split into three additional sub-phases depending on whether one (single) or two (double) limbs are supporting the body. The single limb support phase on the ipsilateral side corresponds to the

 iomechanics of the Hip During B Gait Gait Terminology Walking is achieved through the repetition of steps, from alternating left and right lower limbs. However, clinical gait analysis uses the stride as the elementary gait cycle. The stride commences at one foot contact (usually heel strike) and fi ­ nishes at the next ipsilateral foot contact (Fig. 3.8). A stride combines two steps. A left stride commences when the left foot touches the ground and combines a

M. Sangeux

60

Foot contact

Foot off

Second double support

Swing

Opposite Foot contact

Opposite Foot off

Foot contact

First double support

Stance

Fig. 3.8  One left stride (in red) and two right strides (in blue). Associated events and phases are denoted for the left stride. Color-filled feet are in contact with the ground, outlined feet are not. Adapted from [6] with permission

swing phase on the contralateral side. A transition “double support” phase (typically 10% of the gait cycle) occurs as the ipsilateral limb stance phase ends and the contralateral limb stance phase begins. There are further sub-divisions of the gait cycle, but these are not necessary to detail the kinematics and kinetics of the hip joint. The interested reader is referred to the classic text by Jacqueline Perry [30].

Kinematics Clinical gait analysis primarily focus on the interpretation of the kinematics of the lower limb [6]. Kinematics is the study of the motion of a body, or a system of bodies, without consideration given to its mass or the forces acting on it1. There are many ways to describe a movement mathematically. In gait analysis, movement is expressed as rotations around anatomical axes of the body segments. It may also be described as taking place within the anatomical plane of a limb segment (i.e. sagittal, coronal, and axial planes). To calculate the kinematics of the hip, one must therefore first define

Definition by http://www.thefreedictionary.com

1 

the anatomical axes and planes of the pelvis and femur segments (Fig. 3.9). The orientation of each limb segment is described with Euclidean coordinate systems. By convention in gait analysis, we label the posterior-­ anterior axis (perpendicular to the coronal plane) as X, the right-left axis (perpendicular to the sagittal plane) as Y, and the inferior-superior axis (perpendicular to the transverse plane) as Z [6]. The definition of a Euclidean coordinate system requires these axes to be mutually perpendicular. “Kinematics is the study of the motion of a body, or a system of bodies, without consideration given to its mass or the forces acting on it”. During a clinical gait analysis assessment, anatomical landmarks defining each segment are identified by qualified personnel (typically a physiotherapist or a clinical scientist), and reflective markers are placed at these landmarks. The three-dimensional position of these markers in space are captured by a digital, multi-camera system and subsequently analysed. The pelvis segment is defined by four anatomical landmarks: the left and right anterior superior iliac spines (ASIs) and the left and right posterior superior iliac spines (PSIs) which

3  Biomechanics of the Hip During Gait

61

Zpelvis

PSIs ≈14°

LASI

RASI

Ypelvis RHJC

ASIs Xpelvis

Zfemur

RKNE

RKJC

Yfemur

Coronal view

Xfemur

Sagittal view

Fig. 3.9  Kinematics modelling for the pelvis and right femur segments in a standing subject. The pelvis segment is defined by four anatomical landmarks: the left and right anterior superior iliac spines (ASIs) and the left and right posterior superior iliac spines (PSIs) which define the transverse plane of the pelvis. The right-left axis (Y), defined from the right to the left ASIs is the primary axis. The secondary axis (X*, superimposed with X in the figure) is defined from the mid-point between the right and left PSIs to the mid-point between the left and right ASIs. The inferior-superior axis (Z) is defined from the cross-­ product (Y × X*). The posterior-anterior axis (X) is defined from the cross-product (Y × Z). X* and X are superimposed on this idealised model. It may not be strictly true all the time. Although X and X* lie in the

same transverse plane, only X is guaranteed, by construction, to be perpendicular to both Y and Z and therefore satisfy the definition of the Euclidean coordinate system. The femur segment is defined from three anatomical landmarks: the hip joint centre (HJC), and the medial and lateral knee epicondyles. The knee joint centre (KJC) is defined as the mid-point between the epicondyles. The primary axis (Z, longitudinal axis, perpendicular to the transverse plane) is defined as the vector from KJC to HJC and the secondary axis (Y*) is defined as the vector between the medial and lateral epicondyles. The cross-­ product (Z × Y*) makes the posterior-anterior axis (X, perpendicular to the coronal plane), and the cross-product (Z × X) determines the medio-lateral axis (Y, perpendicular to the sagittal plane)

define the transverse plane of the pelvis. The primary axis of the pelvis segment, the right-left axis (Y), is defined from the right to the left ASIs. The inferior-superior axis (Z) is defined from the cross-product of this axis with an axis (X*) defined from the mid-point between the right and left PSIs to the mid-point between the

left and right ASIs. The posterior-anterior axis (X) is defined from the cross-product of Y and Z. It is important to note that although X* and X lie in the same transverse plane only X is guaranteed, by construction, to be perpendicular to both Y and Z and therefore satisfy the definition of the Euclidean coordinate system (Fig. 3.9).

M. Sangeux

62

The centre of the hip joint is defined within the pelvis segment and is assumed to correspond both to the centre of the acetabulum and the centre of the femoral head. Since it is impossible to directly access the hip joint centre (HJC), regression equations have been devised to estimate the coordinates of the HJC within the pelvis segment, using anthropometric measurements. Anatomic measurements including lower limb length, pelvic width and pelvic depth have been used as predictors of the position of the HJC. Medical imaging has been used to determine the position of the coordinates of the HJC with respect to the pelvis segment as a function of these predictors. The first regression equations published tended to provide markedly different predictions of the position of the HJC (up to 3 cm apart), with poor results [7, 22, 24]. More recent regression equations [17, 18], provide converged position estimates and improved results, with accuracy around 1.5 cm [32, 39, 40]. The femur segment is defined from three anatomical landmarks: the HJC, the medial and lateral knee epicondyles. The knee joint centre (KJC) is defined as the mid-point between the epicondyles. The primary axis (Z, longitudinal axis, perpendicular to the transverse plane) is defined as the vector from KJC to HJC and the secondary axis (Y*) is defined as the vector between the medial and lateral epicondyles. The cross-product of Z and Y* makes the posterior-­ anterior axis (X, perpendicular to the coronal plane), and the cross-product of Z and X determines the medio-lateral axis (Y, perpendicular to the sagittal plane) (Fig. 3.9). In clinical gait analysis, the hip joint angles describe the three sequential rotations (Cardan angles) required to transform the orientation of the pelvis segment into that of the femur segment [35]. The sequence commences with hip flexion-­ extension around the medio-lateral axis of the pelvis, then hip adduction-abduction around a posterior-anterior axis, and finishes with hip internal-external rotation around the longitudinal axis of the femur (Fig. 3.10). By convention, hip flexion, adduction and internal rotation are positive angles and hip extension, abduction and external rotation are negative angles. It is difficult to interpret the kinematics of the hip in isolation from the pelvis. Since the pelvis

1-Flexion 2-Adduction

3-Internal rotation

Fig. 3.10  Rotation sequence (Cardan angles) for the hip joint in gait analysis, represented as its equivalent hip joint coordinate system [52]

is considered the root of the lower-limbs, its parent segment is not another limb segment but simply the 3D coordinate system of the laboratory. The most intuitive way to resolve the orientation of the pelvis in Cardan angles for a clinician is to commence with pelvis rotation around the inferior-­superior axis (i.e. X-Y plane) of the laboratory, then account for pelvis obliquity (Y-Z plane) and to finish with pelvis tilt around the medio-lateral axis of the pelvis (X-Z plane) [5]. Left, or right, pelvic angles are provided for the left, or right, stride respectively. By convention, pelvis anterior tilt, obliquity with ipsilateral side up, and rotation with ipsilateral side forward, are positive angles. Lastly, it may be useful to report the orientation of the femur segment with respect to the laboratory (rather than with respect to the pelvis, i.e. hip kinematics). In this case, we report the angles formed by the longitudinal axis of the femur (Z)

3  Biomechanics of the Hip During Gait

projected in the sagittal and coronal planes of the laboratory, and the medio-lateral axis of the femur (Y) in the transverse plane of the laboratory.

 ypical Gait Kinematics T There are no major differences between the gait kinematics of children and that of adults. The typical gait pattern is estimated to have matured around 5–7 years old [44]. However, the stride to stride kinematics variability during gait decreases with age, and is increased in children aged less than 16 years of age compared to adults [38]. “The stride to stride kinematics variability during gait decreases with age, and is increased in children aged less than 16 years of age compared to adults”. At the pelvis, transverse plane (i.e. X-Y plane) rotation presents the largest range of movement during gait (average amplitude: 13°). The pelvis is rotated forward at initial contact and rotates backwards until 50% of stance phase has been completed, before rotating forward again to start swing phase. Pelvis obliquity is neutral at initial contact, then rises on the ipsilateral side during the first double support period, then decreases during the remainder of stance, and rises again to neutral during swing (Fig. 3.11). At the same time, the contralateral side has the opposite motions. The pelvis is anteriorly tilted for the entire stride (average: 14°; Fig. 3.10), and the range of tilt variation during gait is minimum (average amplitude: 1.3°). Incidentally, the average angle between the transverse plane of the pelvis (from ASIs and PSIs) and the radiological/anatomical anterior pelvic plane (from ASIs and pubic symphysis) was also 14° in a recent study2 [25]. This would mean that the ASIs are directly superior to the pubic symphysis during normal gait for the average gait pattern, and the average pelvis shape. The hip joint (rather than the pelvis) presents the largest range of movement in the sagittal (flexion-extension) plane (average amplitude: 44°). The hip is flexed (average: 36°) at initial contact and extends until reaching maximum The value in the publication is 104° but it corresponds to the complementary angle, therefore 90° need to be subtracted leading to 104 − 90° = 14°.

2 

63

extension (average: 7° extension) towards the end of stance, before flexing into swing phase. Hip flexion reaches a maximum towards the end of swing (average: 37°) and decreases slightly for the last 10% of the stride (Fig. 3.11). Hip motion in the adduction-abduction (Y-Z or coronal) plane presents a similar pattern to that of pelvis obliquity. The hip is neutral at initial contact, abducts throughout mid stance, then crosses into adduction during the first double support period, achieving maximum adduction just after toe off. Hip rotation is relatively constant and slightly internally rotated for most of the gait cycle (Fig. 3.11). Kinematics of the femoral segment resembles that of hip flexion-extension, albeit translated towards increased extension in relation to pelvic tilt. As a result, the projection of the femur in the sagittal plane oscillates around the vertical almost symmetrically during the gait cycle. The range of movement of the femurs in the sagittal plane (average amplitude: 45°) is the major determinant of stride length. For example, it may be difficult to believe that subjects with hip arthrodesis have no range of movement at the hip joint when looking at video footages in the sagittal plane (Fig.  3.12). However, when the kinematics of the pelvis are measured during gait analysis, the contribution of pelvic tilt to the excursion of the femurs in the sagittal plane is substantial and clear (Fig. 3.11). Gait analysis has been shown to be an accurate tool for measuring the effect of hip joint pathologies on hip movement during gait. Reports utilizing gait analysis for pediatric patients with slipped capital femoral epiphysis (SCFE) [37], torsional deformities [42], and cerebral palsy [30], for example, have detailed the pathokinematics of the hip for these conditions.

Kinetics Kinetics is the study of the net forces, moments, and power during the gait cycle which are complimentary measures, interpreted in conjunction with kinematics, to provide a complete picture of a child’s walking pattern. Normal hip kinetics in the sagittal plane is governed by the transition between an internal extension (extensors) moment for the first part of the stance phase to an internal

M. Sangeux

64 Pelvis tilt anterior(+)/posterior(-)

60

20 0 –20

0

Angle (°)

Pelvis obliquity upward(+)/downward(-)

0

50

100

Pelvis pro(+)/re(-) traction

0

50

100

30 20 10 0 –10 –20 –30

30 20 10 0 –10 –20 –30

40 20 0 –20

0

100

50

0

Hip adduction(+)/abduction(-)

Angle (°)

100

0

50

100

Hip rotation internal(+)/external(-)

Angle (°)

50

Angle (°)

Angle (°) Angle (°)

30 20 10 0 –10 –20 –30

20

Femur flexion(+)/extension(-)

60

–20 0

30 20 10 0 –10 –20 –30

40

Angle (°)

40

Angle (°)

Angle (°)

60

Hip flexion(+)/extension(-)

0

50

100

30 20 10 0 –10 –20 –30

30 20 10 0 –10 –20 –30

50

100

Femur adduction(+)/abduction(-)

0

50

100

Femur internal(+)/external(-)

0

50

100

Fig. 3.11  Pelvis (left column), Hip (middle column), and Femur (right column) kinematics during gait in typically developing children (mean: black, 1 SD: grey band), and kinematics of one child with hip arthrodesis (red). The normative data were collected in our laboratory from 35

children and adolescents (mean age: 11 years SD: 3 years), using a state of the art motion capture system and force platforms [38, 48]. We used the conventional gait model [6] and processing was performed using Plug-In-Gait (Nexus, VICON, UK)

Fig. 3.12  Sagittal view of the left limb of an adolescent with hip arthrodesis. Without gait analysis, it would be difficult to identify that most of the range of movement of the femur in the sagittal plane comes from pelvic tilt,

rather than hip flexion (see red curves). These images were taken from screen captures of a video taken during gait analysis hence the image quality is reduced

3  Biomechanics of the Hip During Gait

65

flexion moment during the terminal part of stance (Fig. 3.13). In the coronal plane, there is typically an internal abduction (abductor) moment during stance, most notably during the period of single limb support. Moments generated about the hip in the transverse plane are minimal. “Kinetics involves the study of the net forces, moments, and power during the gait cycle which are complimentary measures, interpreted in conjunction with kinematics, to provide a complete picture of a child’s walking pattern”. Joint power is a scalar quantity calculated as the net moment times the joint angular velocity. Power generation occurs when the net joint moment and the joint angular velocity are in the same direction, i.e. the moment tends to amplify joint movement. Power absorption occurs when

the net joint moment and the joint angular velocity are in opposite directions, i.e. the moment tends to resist joint movement. Joint power may be difficult to interpret because it is difficult to formally attribute the moment, and power generated or absorbed to a single structure, or even a set of structures. Power generation may arise from the action of muscles contraction, or gravity, and power absorption may arise from the resistive action of muscles contracting eccentrically or lengthening passively, tendons lengthening, or any other passive soft tissues lengthening/ stretching (e.g. ligaments or joint capsule). There is a substantial amount of power generation during the first and second double support phases. During the first double support phase, power generation is likely due to the concentric contraction of the hip extensors. During the second double support phase, power generation is

Hip moment extensors(+)/flexors(-)

3

3 2 Moment (Nm)

Moment (Nm)

2 1 0 –1 –2

0

20

40

60

80

0

–2

100

Hip rotators moment external(+)/internal(-)

0

20

40

60

80

100

80

100

Hip power 3 2 Power (W)

2 Moment (Nm)

1

–1

3

1 0 –1 –2

Hip moment abductors(+)/adductors(-)

1 0 –1

0

20

40

60

80

100

–2

0

20

40

60

Fig. 3.13  Sagittal, coronal, and transverse plane hip moments (internal), and hip power, in our normative dataset

M. Sangeux

66

likely due to the concentric contraction of the hip flexors. There is mostly power absorption during single limb support, likely due to the eccentric contraction of the hip abductors.

 ltered Kinematics and Kinetics A of the Hip Joint Pathologies affecting the hip joint, or its periarticular muscles, will likely affect the kinematics and kinetics at the hip during gait. Pathologies resulting in the weakening of the hip abductor muscles (e.g. spina bifida, cerebral palsy) may be revealed by pelvis drop (Trendelenburg’s sign), increased hip adduction, and reduced coronal plane hip moment during single limb support. Reduced coronal plane hip moments may also be identified visually by lateral flexion of the trunk towards the supporting limb (i.e. the “abductor lurch”). By shifting towards the supporting limb, the mass of the trunk modifies the position of the centre of mass of the body so that it aligns better with the hip joint centre in the coronal plane (see next sub-­ section, Joint Contact Forces and Fig.  3.15), which leads to the reduced coronal plane hip moment measured during gait. “Pathologies resulting in the weakening of the hip abductor muscles (e.g. spina bifida, cerebral palsy) may be revealed by pelvis drop (Trendelenburg’s sign), increased hip adduction, and reduced coronal plane hip moment during single limb support”. Some pathologies may lead to functional weakening of the abductor muscles. For example, the presence of coxa valga will reduce the effective length of the femoral neck in the coronal plane and therefore the abduction lever arm of the gluteals. Increased femoral neck anteversion also results in an effective reduction of the abduction lever arm of the gluteals (Fig. 3.7). However, in this case the abduction lever arm may be restored if the hip is rotated internally, leading to the lateralisation of the greater trochanter. Consequently, increased internal hip rotation during stance is considered a positive sign of increased femoral neck anteversion affecting gait function in chil-

dren with cerebral palsy, or idiopathic torsional deformity [3, 28, 29, 31]. Femoral derotation osteotomy is often performed in children with cerebral palsy to correct for increased femoral neck anteversion [4]. “Increased internal hip rotation during stance is considered a positive sign of increased femoral neck anteversion in cerebral palsy, or idiopathic torsional deformity”. In addition to hip abductor weakness, children with cerebral palsy may present with increased muscle tone, and skeletal or muscle deformities (see Chap. 20). For example, increased hip flexion throughout gait, or lack of hip extension in terminal stance may be related to contracture, or spasticity, of the hip flexors [33]. It is important to note that impairments not directly related to structures surrounding the hip may affect kinematics of the hip and the pelvis. For example, increased external tibial torsion may lead to a compensatory increase in hip internal rotation, weakness of the plantarflexors may lead to crouch gait and increased hip flexion, and leg length discrepency or scoliosis may lead to increased pelvis obliquity [33]. It is therefore essential to assess the entire lower limb, through gait analysis and a standardized physical examination, to differentiate between true impairments and secondary compensations affecting the hip. “It is essential to assess the entire lower limb, through gait analysis and a standardized physical examination, to differentiate between true impairments and secondary compensations affecting the hip”.

Joint Contact Forces The joint contact force (sometimes called ‘loading’) is the mechanical stimulus experienced within the joint. It is the force to consider when discussing the loading of the bones or cartilage. Standard clinical gait analysis does not provide information on joint contact forces, which should not be confused with the net forces, from kinetics data. Kinetics data are calculated using a process called inverse dynamics. Inverse dynamics calculates the joint

3  Biomechanics of the Hip During Gait

forces and moments satisfying the equation of motions given the joint kinematics and the forces acting over the entire lower limb, i.e. gravitational, inertial and ground reaction forces. Inverse dynamics does not provide the joint contact forces because it does not consider the forces internal to the lower limb, most notably the muscle forces. Figure 3.14 highlights the difference between the hip joint contact force and the net force from inverse dynamics. The figure presents the hip joint contact force (normalised by bodyweight) during gait as measured by Bergman and colleagues in elderly persons post total hip arthroplasty [8], against the results of inverse dynamics in our normative dataset [48]. The hip joint contact forces were 2.5 times larger than the net forces derived from inverse dynamics, for which periarticular muscle forces are not considered. A simple diagram in the coronal plane allows us to better understand muscle contributions to the joint contact force (Fig. 3.15). During single limb support, a hip abduction moment is generated to balance the weight of the thorax and contralateral lower limb (Fig. 3.15). It is possible to estimate the force required by the abductor muscle group to balance the body in the coronal plane (assuming quasi-static equilibrium) if one knows the moment arms of both the centre of mass and the hip abductor muscles, as well as the force due to body weight. Static equilibrium 300

Joint contact force Joint force from inverse dynamics

Force (%BW)

250 200

FHip (femur→pelvis)

Fa

FCOM dCOM da

Fig. 3.15  Diagram presenting the force acting on the pelvis from the gluteus medius and minimus abductor muscles (Fa) and gravity (FCOM) and the reaction required at the hip to assure static equilibrium during single limb support. Fa and da represent the overall force and moment-arm of the gluteus medius and minimus muscle fibres

(i.e. when the body is at rest) imposes that the sum of the moments around the hip joint centre is equal to zero according to the equation: Fa × d a −∗ FCOM × dCOM = 0 with Fa and FCOM respectively being the forces due to the contraction of the abductors and that of supraacetabular bodyweight, respectively3. Assuming the moment arms of Fa and FCOM with respect to the hip joint centre are such that dCOM = 2 × da, the equation above may be simplified to: Fa = 2 × FCOM Therefore, the force developed by the abductor muscles needs to be twice that of the bodyweight, due to the shorter moment arm associated with the abductors. Once we know the force of the abductors, and assuming no other muscles crossing the hip are active, we can estimate the hip joint contact force

150 100 50 0

67

0

10 20 30 40 50 60 70 80 90 100

Fig. 3.14  Magnitude of the hip joint contact force from Bergmann and colleagues [8] and derived from inverse dynamics, which only include forces external to the musculoskeletal system (gravitational and inertial), i.e. does not take muscle forces into account

By convention, the moment of a force that rotates the segment clockwise is negative.

3 

M. Sangeux

68

FHip. Indeed, static equilibrium also imposes that the sum of the forces is equal to zero. Therefore: Fa + FCOM = 3 × FCOM = FHip



This theoretical result illustrates the experimental results of Bergman and colleagues [8] (Fig. 3.15) where the maximum hip joint contact force during gait occurs at the beginning of the single limb support phase, and is equal to 2.6 bodyweight (i.e. FCOM). Bergman and colleagues have been investigating hip joint contact force for more than 30 years, their experimental data is freely available from the Orthoload website (https://orthoload.com/database/). Recently, this same research group measured hip joint contact forces in a range of daily living activities [8], Table 3.2. The data were collected in adults post-total hip arthroplasty, so may not be directly comparable with the hip joint contact forces experienced in children. However, the relative differences between activities may still be informative. The study results demonstrated that

when cycling at 40  rpm and 90  W, the hip joint contact force was equivalent to body weight (BW), whereas sitting down or standing up doubled the forces. The maximum hip contact forces among the various activities were also measured for jogging, with peak values of more than four times the BW.  Overall, these values further highlight the major contribution of muscle forces to joint reaction forces. Hip joint reaction forces during various activities are detailed in Table 3.2. There are limited means to measure joint contact forces and muscle forces in  vivo, and none are non-invasive. Therefore, musculoskeletal modelling is used in biomechanics research to estimate muscle forces, and subsequently estimate joint contact forces [27]. There are still some obstacles to the routine use of musculoskeletal modelling for clinical practice. The main issues relate to the necessity to create subject-­ specific models to provide accurate estimate of muscle and joint contact forces [1, 15, 28, 29]. “Musculoskeletal modelling is used to estimate muscle forces, and subsequently estimate joint contact forces”.

Table 3.2  The average (across subjects) maximum hip joint contact force during daily living activities from [8] Activities Cycling Sitting down Standing up Squatting Walking Walking up stairs Walking down stairs Jogging

Average peak (%BW) 99% 185% 217% 231% 262% 303% 313% 417%

Final Note The hip joint allows a wide range of movement in the three anatomical planes. Although this chapter focused on the biomechanics of the hip during gait, it is important to note that participation in everyday activities involves more at the hip than what is needed to walk. Figure 3.16 summarises the hip range of movement

BW body weight Sagittal plane

Coronal plane

Transverse plane

Walking

Walking

Walking

Cycling

Cycling

Cycling

Sitting

Sitting

Sitting

Squatting

Squatting

Squatting

PROM

PROM

PROM

–20

0

Extension

20 40 60 80 100 120 (°)

Flexion

–40

–20

0

Abduction

Fig. 3.16  Maximum (pale) and average (bright) hip joint range of movement during PROM (passive range of movement), Squatting, Sitting, Cycling (includes mount-

20

(°)

Adduction

–60 –40 –20

External

0

20

40

60 (°)

Internal

ing and dismounting) and walking. Data from Reese and Bandy [55], Hara et al. [54], and Adam et al. [51]

3  Biomechanics of the Hip During Gait

during various activities: passive physical examination, sitting, cycling (including the hip movement required to get in and out of a saddle), walking and squatting to pick up an object from the floor. Figure 3.16 highlights that the hip during walking operates at opposite extremes of the passive range of movement than most of the other activities, sitting, squatting, and cycling. When planning for surgical reconstruction, the surgeon should consider the function of the hip in other activities, not only during walking.

Classic Literature Perry, J. (1992). Gait Analysis Normal and Pathological Function. New Jersey, SLACK Incorporated. Jacqueline Perry pioneered the use of gait analysis to investigate musculoskeletal problems and determine treatment plans. Her classical textbook provides a comprehensive description of normal gait (primarily kinematics and electromyography), and many examples of pathological gait with interpretation. Baker, R. (2013). Measuring Walking: A Handbook of Clinical Gait Analysis. Cambridge, Mac Keith Press. In 2003, Richard Baker and Jillian Rodda produced a CD-ROM titled “All you ever wanted to know about the conventional gait model, but were afraid to ask”. The handbook published by Richard Baker in 2013 provides all you ever wanted to know about measuring walking, including both theoretical and practical aspects. The book will be particularly useful for those who start a new clinical gait analysis service, and wish to implement best practices. Pandy, M. G. and T. P. Andriacchi (2010). Muscle and joint function in human locomotion. Annu Rev Biomed Eng 12: 401–433. Marcus Pandy and Thomas P. Andriacchi provide a comprehensive overview of musculoskeletal modelling as a means to estimate muscle and joint function during gait and running. The review highlights some of the answers musculoskeletal modelling provided to fundamental questions, such as why humans choose to switch between walking and running at a specific speed.

69

Key Evidence Bergmann, G., A.  Bender, J.  Dymke, G. Duda and P. Damm (2016). Standardized Loads Acting in Hip Implants. PLoS One 11(5): e0155612. This article is the culmination of 30  years of research by Bergman and colleagues in measuring hip joint loadings in  vivo, using instrumented hip prostheses. The article is supplemented with a comprehensive dataset, including hip joint contact force data measured in ten subjects for various activities. Arnold, A. S., A. V. Komattu and S. L. Delp (1997). Internal rotation gait: A compensatory mechanism to restore abduction capacity decreased by bone deformity? Developmental Medicine and Child Neurology 39(1): 40–44. This article explored the relationship between femoral neck anteversion, lever arm of the hip abductor muscles and hip rotation during gait. Arnold et al. used musculoskeletal modelling to propose a convincing mechanical explanation for the increased hip rotation kinematics pattern observed in children with excessive femoral anteversion. Delp, S.  L. and F.  E. Zajac (1992). Force-­ Generating and Moment-Generating Capacity of Lower-Extremity Muscles before and after Tendon Lengthening. Clinical Orthopaedics and Related Research (284): 247–259. This article is one of the earliest attempts to use musculoskeletal modelling to answer practical orthopaedic question. The authors estimated the effect of lower limb muscle tendon lengthening on the maximum isometric force, and subsequently the joint moment, muscle can produce before and after the simulated lengthening procedure. The authors showed that lengthening the Achilles tendon leads to a dramatic reduction in soleus maximum force and ankle moment. At the hip, the lengthening of the semimembranosus tendon leads to a large reduction in force and hip extension moment, whereas lengthening the semitendinosus tendon has very small effects [13].

M. Sangeux

70

Take Home Messages

• Passive hip joint structures such as the labrum, and hip extracapsular ligaments play a major biomechanical role in hip stability. • Hip muscles act in different planes and their role may change depending on the angle of hip flexion-extension. • Shape parameters of the femur, the neck shaft angle and the femoral neck anteversion, have a significant effect on muscle roles and lever arms. • Three-dimension gait analysis is essential to the understanding of hip kinematics, especially to evaluate compensation at the pelvis and other joints of the lower limbs due to hip pathologies. • Muscle forces are the main contributors to hip joint contact forces.

Acknowledgments  I am grateful for the work of the teams behind the free and open-source software 3DSlicer and OpenSim [12, 14], which allowed me to produce many of the simulations and illustrations in this chapter.

References 1. Ackland DC, Lin YC, Pandy MG. Sensitivity of model predictions of muscle function to changes in moment arms and muscle-tendon properties: a Monte-Carlo analysis. J Biomech. 2012;45(8):1463–71. 2. Albers CE, Schwarz A, Hanke MS, Kienle KP, Werlen S, Siebenrock KA. Acetabular version increases after closure of the triradiate cartilage complex. Clin Orthop Relat Res. 2017;475(4):983–94. 3. Arnold AS, Komattu AV, Delp SL.  Internal rotation gait: a compensatory mechanism to restore abduction capacity decreased by bone deformity? Dev Med Child Neurol. 1997;39(1):40–4. 4. Bache C, Selber P, Graham H.  The management of spastic diplegia. Curr Orthop. 2003;17:88–104. 5. Baker R. Pelvic angles: a mathematically rigorous definition which is consistent with a conventional clinical understanding of the terms. Gait Posture. 2001;13(1):1–6. 6. Baker R, Leboeuf F, Reay J, Sangeux M. The Conventional Gait Model - Success and Limitations. In: Müller B, Wolf SI, Brueggemann G-P, Deng Z, McIntosh A, Miller F, et al., editors. Handbook of Human Motion. Cham: Springer International Publishing; 2017. p. 1–19.

7. Bell AL, Pedersen DR, Brand RA.  A comparison of the accuracy of several hip center location prediction methods. J Biomech. 1990;23(6):617–21. 8. Bergmann G, Bender A, Dymke J, Duda G, Damm P.  Standardized loads acting in hip implants. PLoS One. 2016;11(5):e0155612. 9. Boese CK, Dargel J, Oppermann J, Eysel P, Scheyerer MJ, Bredow J, Lechler P.  The femoral neck-shaft angle on plain radiographs: a systematic review. Skeletal Radiol. 2016;45(1):19–28. 10. Botser IB, Ozoude GC, Martin DE, Siddiqi AJ, Kuppuswami S, Domb BG.  Femoral anteversion in the hip: comparison of measurement by computed tomography, magnetic resonance imaging, and physical examination. Arthroscopy. 2012;28(5):619–27. 11. Bsat S, Frei H, Beaule PE.  The acetabular labrum: a review of its function. Bone Joint J. 2016;98-B(6):730–5. 12. Delp SL, Anderson FC, Arnold AS, Loan P, Habib A, John CT, Guendelman E, Thelen DG.  OpenSim: open-source software to create and analyze dynamic simulations of movement. IEEE Trans Biomed Eng. 2007;54(11):1940–50. 13. Delp SL, Zajac FE.  Force-generating and moment-­ generating capacity of lower-extremity muscles before and after tendon lengthening. Clin Orthop Relat Res. 1992;(284):247–59. 14. Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin JC, Pujol S, Bauer C, Jennings D, Fennessy F, Sonka M, Buatti J, Aylward S, Miller JV, Pieper S, Kikinis R. 3D Slicer as an image computing platform for the quantitative imaging network. Magn Reson Imaging. 2012;30(9):1323–41. 15. Gerus P, Sartori M, Besier TF, Fregly BJ, Delp SL, Banks SA, Pandy MG, D’Lima DD, Lloyd DG.  Subject-specific knee joint geometry improves predictions of medial tibiofemoral contact forces. J Biomech. 2013;46(16):2778–86. 16. Gilligan I, Chandraphak S, Mahakkanukrauh P.  Femoral neck-shaft angle in humans: variation relating to climate, clothing, lifestyle, sex, age and side. J Anat. 2013;223(2):133–51. 17. Hara R, McGinley J, Briggs C, Baker R, Sangeux M.  Predicting the location of the hip joint centres, impact of age group and sex. Sci Rep. 2016;6:37707. 18. Harrington ME, Zavatsky AB, Lawson SEM, Yuan Z, Theologis TN.  Prediction of the hip joint centre in adults, children, and patients with cerebral palsy based on magnetic resonance imaging. J Biomech. 2007;40(3):595–602. 19. Higgins SW, Spratley EM, Boe RA, Hayes CW, Jiranek WA, Wayne JS. A novel approach for determining three-dimensional acetabular orientation: results from two hundred subjects. J Bone Jt Surg Am. 2014;96(21):1776–84. 20. Hingsammer AM, Bixby S, Zurakowski D, Yen YM, Kim YJ. How do acetabular version and femoral head coverage change with skeletal maturity? Clin Orthop Relat Res. 2015;473(4):1224–33. 21. Jacquemier M, Glard Y, Pomero V, Viehweger E, Jouve JL, Bollini G.  Rotational profile of the lower limb in 1319 healthy children. Gait Posture. 2008;28(2):187–93.

3  Biomechanics of the Hip During Gait 22. Kainz H, Carty CP, Modenese L, Boyd RN, Lloyd DG.  Estimation of the hip joint centre in human motion analysis: a systematic review. Clin Biomech (Bristol, Avon). 2015;30:319. 23. Koerner JD, Patel NM, Yoon RS, Sirkin MS, Reilly MC, Liporace FA.  Femoral version of the general population: does “normal” vary by gender or ethnicity? J Orthop Trauma. 2013;27(6):308–11. 24. Leardini A, Cappozzo A, Catani F, Toksvig-Larsen S, Petitto A, Sforza V, Cassanelli G, Giannini S.  Validation of a functional method for the estimation of hip joint center location. J Biomech. 1999;32:99–103. 25. Lewis CL, Laudicina NM, Khuu A, Loverro KL. The human pelvis: variation in structure and function during gait. Anat Rec (Hoboken). 2017;300(4):633–42. 26. Myers CA, Register BC, Lertwanich P, Ejnisman L, Pennington WW, Giphart JE, LaPrade RF, Philippon MJ. Role of the acetabular labrum and the iliofemoral ligament in hip stability: an in vitro biplane fluoroscopy study. Am J Sports Med. 2011;39(Suppl):85S–91S. 27. Pandy MG, Andriacchi TP.  Muscle and joint function in human locomotion. Annu Rev Biomed Eng. 2010;12:401–33. 28. Passmore E, Graham HK, Pandy M, Sangeux M. The effect of femoral and tibial torsion on muscle and joint function during walking in typically developing children. Gait Posture. 2016a;49:41. 29. Passmore E, Pandy M, Graham HK, Sangeux M. What is the respective effect of joint position and bone shape on clinical musculoskeletal modelling? An EOS study. Gait Posture. 2016b;49(Suppl 2):26. 30. Perry J. Gait analysis normal and pathological function. Thorofare, NJ: SLACK Incorporated; 1992. 31. Robin J, Graham HK, Selber P, Dobson F, Smith K, Baker R.  Proximal femoral geometry in cerebral palsy: a population-based cross-sectional study. J Bone Jt Surg B. 2008;90(10):1372–9. 32. Sangeux M.  On the implementation of predictive methods to locate the hip joint centres. Gait Posture. 2015;42(3):402–5. 33. Sangeux M, Armand S. In: Canavese F, Deslandes J, editors. Kinematic deviations in children with cerebral palsy orthopedic management of children with cerebral palsy: a comprehensive approach. New York, NY: Nova Science Publishers; 2015. 34. Sangeux M, Mahy J, Graham HK. Do physical examination and CT-scan measures of femoral neck anteversion and tibial torsion relate to each other? Gait Posture. 2014a;39(1):12–6. 35. Sangeux M, Marin F, Charleux F, Ho Ba Tho MC. In vivo mechanical properties of the anterior cruciate ligament. Comput Methods Biomech Biomed Engin. 2007;10(suppl):35–6. 36. Sangeux M, Pascoe J, Graham HK, Ramanauskas F, Cain T.  Three-dimensional measurement of femoral neck anteversion and neck shaft angle. J Comput Assist Tomogr. 2015a;39(1):83–5. 37. Sangeux M, Passmore E, Gomez G, Balakumar J, Graham HK.  Slipped capital femoral epiphysis, fixation by single screw in situ: a kinematic and

71 radiographic study. Clin Biomech (Bristol, Avon). 2014b;29(5):523–30. 38. Sangeux M, Passmore E, Graham HK, Tirosh O. The gait standard deviation, a single measure of kinematic variability. Gait Posture. 2016;46:194–200. 39. Sangeux M, Peters A, Baker R. Hip joint centre localization: evaluation on normal subjects in the context of gait analysis. Gait Posture. 2011;34(3):324–8. 40. Sangeux M, Pillet H, Skalli W. Which method of hip joint centre localisation should be used in gait analysis? Gait Posture. 2014c;40(1):20–5. 41. Sangeux M, Rodda J, Graham HK. Sagittal gait patterns in cerebral palsy: the plantarflexor-knee extension couple index. Gait Posture. 2015b;41(2):586–91. 42. Schwartz MH, Rozumalski A, Novacheck TF. Femoral derotational osteotomy: surgical indications and outcomes in children with cerebral palsy. Gait Posture. 2014;39(2):778–83. 43. Staheli LT, Corbett M, Wyss C, King H.  Lower-­ extremity rotational problems in children. Normal values to guide management. J Bone Joint Surg Am. 1985;67(1):39–47. 44. Sutherland DH, Olshen R. The development of mature walking. London: Mac Keith Press; 1988. 45. Szuper K, Schlegl AT, Leidecker E, Vermes C, Somoskeoy S, Than P. Three-dimensional quantitative analysis of the proximal femur and the pelvis in children and adolescents using an upright biplanar slot-scanning X-ray system. Pediatr Radiol. 2015;45(3):411–21. 46. Tayton E.  Femoral anteversion. J Bone Jt Surg B. 2007;89(10):1283–8. 47. Thelen DG, Anderson FC, Delp SL.  Generating dynamic simulations of movement using computed muscle control. J Biomech. 2003;36(3):321–8. 48. Tirosh O, Sangeux M, Wong M, Thomason P, Graham HK. Walking speed effects on the lower limb electromyographic variability of healthy children aged 7-16 years. J Electromyogr Kinesiol. 2013;23(6):1451–9. 49. van Arkel RJ, Amis AA, Jeffers JR. The envelope of passive motion allowed by the capsular ligaments of the hip. J Biomech. 2015;48(14):3803–9. 50. Zhang H, Wang Y, Ai S, Chen X, Wang L, Dai K.  Three-dimensional acetabular orientation measurement in a reliable coordinate system among one hundred Chinese. PLoS One. 2017;12(2):e0172297. 51. Adam P, Beguin L, Grosclaude S, Jobard B, Fessy MH.  Functional range of motion of the hip joint. Rev Chir Orthop Reparatrice Appar Mot. 2008;94(4):382–91. 52. Baker R. Globographic visualisation of three dimensional joint angles. J Biomech. 2011;44(10):1885–91. 53. Delp SL, Hess WE, Hungerford DS, Jones LC. Variation of rotation moment arms with hip flexion. J Biomech. 1999;32(5):493–501. 54. Hara D, Nakashima Y, Hamai S, Higaki H, Ikebe S, Shimoto T, Hirata M, Kanazawa M, Kohno Y, Iwamoto Y.  Kinematic analysis of healthy hips during weightbearing activities by 3D-to-2D model-­to-­image registration technique. Biomed Res Int. 2014;2014:457573. 55. Reese NB, Bandy WD.  Joint range of motion and muscle length testing. London: Elsevier Health Science; 2016.

Part II Developmental Dysplasia of the Hip

4

Developmental Dysplasia of the Hip in Young Children Stuart L. Weinstein and Joshua B. Holt

Introduction Developmental dysplasia of the hip (DDH) is an allencompassing term used to describe the wide spectrum of disorders of development of the hip that manifests in various forms and at different ages. The longstanding terminology of congenital dysplasia of the hip was initially described by Dupuytren in 1832 when he noted the classic features of shortening of the thigh, lack of abduction, prominence of the greater trochanters, and associated abductor lurch [1]. This original term has generally been replaced by developmental dysplasia or DDH since the 1980s after it was initially introduced in the 1960s as the former implies the existence of the disorder at the time of birth, while the latter more appropriately conveys the spectrum of associated hip pathology typically being described [2–4]. Failure to maintain the appropriate spatial positioning of the femoral head in relation to the acetabulum results in the clinical findings of instability, subluxation, dislocation, or more inclusively dysplasia. As the newer terminology implies, DDH often evolves over time as the structures of the hip are normal during embryogenesis but gradually become abnormal. Importantly, progression of S. L. Weinstein (*) · J. B. Holt Department of Orthopaedics and Rehabilitation, University of Iowa Hospitals and Clinics, Iowa City, IA, USA e-mail: [email protected]; Joshua-holt@ uiowa.edu

© Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_4

deformity is dynamic and is capable of getting better or worse as the child develops, depending on the multidisciplinary care provided [5]. As such, DDH will be used herein to describe any structural abnormality of the hip resulting from an uncharacteristic relationship between the femoral head and the acetabulum, not associated with a discrete primary insult. The general exception to this rule is the case of “teratologic dislocation of the hip”, in which the femoroacetabular relationship is abnormal before birth and presents as an irreducible hip with limited range of motion during the newborn period. Teratologic dislocation is commonly associated with neuromuscular syndromes, such as arthrogryposis or myelodysplasia. This is a rare unique entity and will be discussed separately. “Importantly, progression of deformity is dynamic and is capable of getting better or worse as the child develops…depending on the multidisciplinary care provided [5]”.

Pathophysiology Normal Growth and Development During embryonic development, the acetabulum and the femoral head develop from the same primitive mesenchymal cells [6–8]. Around the seventh week of gestation a cleft develops in the precartilaginous cells which progressively distinguish the acetabulum and femoral head. The hip joint is fully 75

76

S. L. Weinstein and J. B. Holt

tissues are in continuity with the joint capsule inserting just above it. The articulating surface of the acetabular cartilage is covered with articular cartilage while the opposite side is a typical growth plate. Through interstitial growth within the cartilage and appositional growth beneath the perichondrium, the acetabular cartilage continues to slowly grow throughout childhood. The importance of this growth center becomes all too evident after aggressive periosteal stripping or improper osteotome placement results in growth disturbance. Fig. 4.1  Embryonic hip. Note that the acetabulum and femoral head develop from the same mesenchymal cells before the development of a cleft at about 7 weeks gestation, which eventually divides the acetabulum and femoral head into separate structures. (Used with permission)

formed by the 11th gestational week [7–9] (Fig.  4.1). In the normal hip, acetabular development continues during intrauterine life by way of labral growth and the femoral head remains deeply seated in the acetabulum [10]. This relationship is largely maintained by the surface tension of synovial fluid. Even after incising the joint capsule, dislocation of a normal infant’s hip is extremely difficult [11, 12]. This highlights the common misconception that DDH is the consequence of increased capsular laxity. Normal growth and development of the hip continues in the newborn period through a genetically determined relationship between the acetabular and triradiate cartilages and the proximal femur. Of primary importance is the maintenance of this relationship, including a centered, well-­positioned femoral head within the developing acetabulum [9, 13–16].

Growth of the Proximal Femur The entire proximal end of the femur is composed of cartilage in the infant. Three key growth centers are present in the proximal femur: the physeal plate, the greater trochanter, and the femoral neck isthmus [18] (Fig. 4.2). The trochanter and proxi-

Acetabular Growth and Development The acetabular cartilage complex is a three-­ dimensional structure representing the confluence of the ilium, ischium, and pubis. The outer twothirds of the cup-shaped cavity is formed from acetabular cartilage and remains covered in hyaline cartilage throughout development. The lateral portion of this cartilage is homologous with epiphyseal cartilage of the skeleton [17]. Projecting from the margin of the acetabular cartilage is a fibrocartilaginous extension of tissue, the acetabular labrum. Adjacent to the acetabular cartilage, labral

Fig. 4.2 The proximal femur of an infant demonstrating three physeal plates. This includes the growth plates of the greater trochanter, the proximal femoral physis (physeal plate), and the femoral neck isthmus, a reflection of the previous common origin of the other two. (Used with permission)

4  Developmental Dysplasia of the Hip in Young Children

mal femur enlarge by appositional cartilage cell proliferation with the proximal femoral ossification center appearing between the fourth and seventh month of life [18]. Ossification of this cartilaginous anlage continues at a decreasing rate until skeletal maturity when only a thin layer of articular cartilage remains. Influenced by a combination of forces including resting muscle tension, active muscular pull, and normal weight bearing—in conjunction with synovial fluid nutrition and femoral blood supply—the growth and development of the proximal femur is dependent on a balance between the three growth centers. Normal growth and the adult shape of the proximal femur can be altered by changes in any of these factors [18–22]. “Influenced by a combination of muscle forces and normal weight bearing  - in conjunction with synovial fluid nutrition and femoral blood supply - the growth and development of the proximal femur is dependent on a balance between three growth centers”. During infancy, the trochanteric and epiphyseal growth plates are connected by a small cartilaginous isthmus along the lateral border of the femoral neck. This growth cartilage is a reflection of their previous common origin and contributes to the expanding lateral width of the femoral neck until maturity. Any disturbance in growth (e.g. post-traumatic physeal arrest or stimulation secondary to post-inflammatory hyperemia) involving one or more of these three growth centers can alter the eventual shape of the proximal femur. Approximately 30% of the overall growth and length of the femur is determined by the proximal femoral physeal plate contributions. Any damage to, or alterations in, the physeal blood supply has the potential to disrupt its normal growth, often resulting in varus deformity of the proximal femur. This typical deformity occurs due to an imbalance in growth between the slowed epiphyseal growth plate and continued growth at the trochanteric and femoral neck growth plates [23]. This manifests as an abnormal articular trochanteric distance, measured from the tip of the greater trochanter to the superior articular surface of the femoral head. Conversely, the greater trochanteric physis is categorized as a traction apophysis whose normal appositional growth is

77

dependent on abductor muscle pull, resulting in a continuation of growth despite alterations in femoral head and neck physeal activity. As such, “trochanteric overgrowth” is actually normal trochanteric growth in the presence of “undergrowth” of the proximal end of the femur [21].

 eterminants of Shape and Depth D of the Acetabulum Both clinical studies in humans with unreduced dislocations and experimental studies in animals have clearly demonstrated that the continued presence of a spherical femoral head is the main stimulus for the development of the typical concave shape of the acetabulum [13, 17, 24–26]. Failed development of adequate acetabular depth and area was demonstrated after excision of the femoral head in rats. Additionally, despite normal innominate bone length and triradiate cartilage histology, the acetabular cartilage is prone to atrophy and degeneration without the normal stimulus of a reduced femoral head [13]. In fact, normal interstitial and appositional growth within the acetabular cartilage, new periosteal bone formation in the adjacent pelvic bones, and a reduced, spherical femoral head, must all occur in concert for the normal depth of the acetabulum to be obtained during development [13, 14]. Further deepening occurs as three secondary ossification centers develop at puberty [17]. The first of these centers is the os acetabulum which develops into the thick cartilage that separates the acetabulum from the pubis and forms the anterior wall at about 8 years of age. The second is the acetabular epiphysis which functions as the epiphysis of the ilium and forms a major portion of its superior edge. It also appears at the age of 8 and fuses to the remainder of the acetabulum at about 18 years. The third center is a small epiphysis in the ischial region that appears around 9  years and fuses at 17 years [13, 14, 17, 27] (Fig. 4.3). The coordinated growth of the proximal femoral, acetabular, and triradiate cartilages—and adjacent bones—are together responsible for normal acetabular growth and development. Alterations in this finely balanced, genetically determined growth—coupled with abnormalities

S. L. Weinstein and J. B. Holt

78

Acetabular Epiphysis

Labrum

Ilium

Os Acetabuli te Triradia

Epiphysis Pubis Ischium Transverse Acetabular Ligament ©JSchoenecker2018

Fig. 4.3  Diagram of the right innominate bone. Note the os acetabulum within the acetabular cartilage adjoining the pubis. The acetabular epiphysis adjoining the ilium and the epiphysis adjoining the ischium comprise the other two growth centers within the acetabular cartilage. The origin of the ligamentum teres (*) is shown on the non-articular medial wall

in the intrauterine environment, vascular supply, and relatively gentle forces persistently applied—contribute to the pathogenesis of hip dysplasia [28–35]. “Alterations in the growth of the proximal femoral, acetabular, and triradiate cartilages - coupled with abnormalities in the intrauterine environment, vascular supply, and joint forces  - contribute to the pathogenesis of hip dysplasia”.

Hip Dislocation in the Newborn DDH in the newborn appropriately describes a broad spectrum of growth abnormalities of the proximal femur and/or acetabulum, ranging from

mild dysplastic changes to severe pathologic changes. The more severe pathologic changes are typically seen in teratologic dislocations. The most common pathoanatomical change in the newborn with DDH is the neolimbus described by Ortolani [20, 36], present in up to 98% of DDH cases that occur perinatally. The neolimbus represents a hypertrophied ridge of acetabular cartilage in the superior, posterior, and inferior aspects of the acetabulum [12, 37]. Pressure of the femoral head or neck on this hypertrophied tissue often results in a trough or groove in the acetabular cartilage. The palpable sensation of the femoral head gliding in and out of the acetabulum over this ridge of acetabular cartilage produces the Ortolani sign (see Clinical Presentation section) [20, 36]. “The neolimbus represents a hypertrophied ridge of acetabular cartilage in the superior, posterior, and inferior aspects of the acetabulum [12, 37]”. With stabilization of the anatomic relationship between the acetabulum and femoral head, these pathologic changes are typically reversible in the newborn with DDH, as evidenced by the 95% success rate of treating DDH with simple devices such as the Pavlik harness and the von Rosen splint [38– 40]. Conversely, an estimated 2% of newborns with DDH have idiopathic teratologic dislocations in the antenatal period not associated with a neuromuscular condition or other syndromes. The pathoanatomy, clinical findings, and treatment success are different in these cases than for typical DDH.

Acetabular Development in DDH Although it is generally accepted that early identification and successful treatment of DDH in the newborn period typically results in normal hip development, acetabular development may be irreversibly affected and fail to adequately develop even in early-diagnosed and appropriately treated cases. As discussed previously, the primary stimulus for normal growth and development is maintenance of the appropriate relationship between the femoral head and the acetabulum during growth [13, 25, 26]. As soon as subluxation or dislocation

4  Developmental Dysplasia of the Hip in Young Children

has been identified, the femoral head must be reduced as soon as possible and this reduction maintained. With increasing delay in diagnosis, normal growth and development may not occur. If a concentric reduction is achieved and maintained at an early stage, however, the acetabulum has great potential for resumption of normal growth and development [41–43]. The age at which a dysplastic hip can return to “normal” after reduction remains controversial [40, 42–50]. The ability of the acetabular cartilage to resume normal development post reduction is multifactorial, depending on the age at which reduction is obtained, its intrinsic growth potential, and the presence of normal proximal femoral geometry. In patients with treated DDH, accessory centers of ossification have been found to contribute to acetabular development in up to two thirds of cases, compared with only 2–3% of normal hips. These accessory centers appear 6–10  months after reduction within the peripheral acetabular cartilage [41–44, 51]. They are thought to be caused by persistently abnormal joint forces imparted to the “neolimbus” by the displaced femoral head and/or neck, or from damage incurred from repeated attempts at unsuccessful closed reduction. Evidence of accessory centers of ossification should be sought on post-reduction radiographs to determine whether acetabular development is progressing. This should factor into the decision about proceeding with surgical intervention for residual acetabular dysplasia. Although accessory ossification centers may coalesce to form a normal acetabulum, their presence may indicate injury to the cartilage and does not assure normal acetabular development.

Natural History Course of DDH in Newborns The natural history of untreated DDH in the newborn remains an enigma. The rate in which an unstable hip spontaneously reduces, becomes dislocated, subluxated, or dysplastic remains unknown. Barlow reported that over 60% of

79

infants born with any sign of clinical instability (e.g. Ortolani positive) would normalize within the first week of life without treatment and that this number increases to 88% by the third month of life [52]. Similarly, Pratt and colleagues reported that 15 of 18 “dysplastic” hips at age less than 3  months were radiographically normal without treatment at a mean follow-up of 11 years [53]. In contrast, Coleman reported that only 22% of 23 dysplastic hips at age less than 3  months were normal at 3-year follow-up, while 26% were dislocated, 13% subluxated, and 39% remained dysplastic [54]. Further complicating the discussion and classification of instability is the consideration of hips that are clinically stable but have abnormal ultrasonographic findings. Since it is not possible to predict which hips will normalize, which hips will dislocate, and which hips will remain reduced yet dysplastic, all newborns with hip instability on clinical exam should be treated. The decision to treat infants with somewhat abnormal ultrasounds despite normal clinical examination is somewhat more controversial. Most providers, including the authors, typically err on the side of overtreatment and would recommend treatment in the setting of markedly abnormal ultrasound despite a normal clinical exam. “Since it is not possible to predict which hips will normalize, which hips will dislocate, and which hips will remain reduced yet dysplastic, all newborns with hip instability on clinical exam should be treated”.

Course of DDH in Adults As in children, the natural history of untreated DDH in the adult is variable. In the setting of complete dislocation, outcomes depend largely on the development of a false acetabulum and bilaterality [28, 55–58]. When a false acetabulum is absent or poorly formed, patients have a greater than 50% chance of a good outcome and often maintain good range of motion with little functional disability; compared to less than 25% chance of a good outcome with a well formed false acetabulum [55]. Hips

80

with well-developed false acetabula are more likely to develop radiographic evidence of degenerative joint disease and experience poor clinical outcomes. Patients with bilateral dislocations often complain of lower back pain related to hyperlordosis of the lumbar spine [27, 55, 59–61]. In the case of unilateral complete dislocation, problems relating to limb-length inequality, ipsilateral painful knee deformity, compensatory scoliosis, decreased agility, and gait disturbance, are common. Valgus deformity at the knee develops secondary to flexion-adduction deformity at the hip, often leading to attenuation of the medial collateral ligament and degenerative arthritis of the lateral compartment [27, 55, 58].

Course of Dysplasia and Subluxation After the neonatal period, the term hip dysplasia can be used to describe either anatomic or radiographic abnormalities. Anatomic dysplasia refers to inadequate development of the acetabulum and/or femoral head often resulting in abnormal contact (i.e. subluxation or dislocation) between the proximal femur and acetabulum [62]. Unlike anatomic dysplasia, which includes the full spectrum of hip displacement, an important distinction is made between radiographic dysplasia and radiographic subluxation. In radiographic dysplasia, the Shenton line remains intact [55, 63, 64] but the acetabulum shows increased obliquity and loss of concavity. Radiographic subluxation refers to a hip with a disruption of the Shenton line and the femoral head is superiorly, laterally, or superolaterally displaced from the medial wall of the acetabulum. Although these radiographic findings should be described independently as their natural histories are different, secondary degenerative changes at a later stage may lead to conversion from radiographic dysplasia to radiographic subluxation [59, 61, 64–68]. Radiographic subluxation, resulting from untreated or incompletely treated DDH, invariably leads to degenerative joint disease and clinical instability [27, 56, 59, 64, 66, 69–71].

S. L. Weinstein and J. B. Holt

“Radiographic subluxation, resulting from untreated or incompletely treated DDH, invariably leads to degenerative joint disease and clinical instability”. In the adolescent and young adult population, post-DDH dysplasia should be distinguished from hip dysplasia unrelated to DDH of the newborn. Although multiple studies evaluating hip arthroplasty demonstrate a high prevalence of early degenerative joint disease secondary to adolescent hip dysplasia there is still much to learn about this entity. It is believed, however, that uncorrected dysplasia from DDH in the newborn consistently leads to early degenerative joint disease. Although it has been suggested that the center-­ edge (CE) angle is predictive of degenerative joint disease, there is currently no good evidence that supports a definitive correlation between radiographic measures of dysplasia (e.g. CE angle, acetabular index, migration percentage, etc.) and disease progression [51, 64]. The rate of joint deterioration and subsequent clinical disability is, however, directly related to the presence and severity of subluxation and the age at the time of diagnosis [59, 64–66, 68, 72–75]. Once pain associated with degenerative arthritis begins, it typically progresses rapidly over a period of months [55]. In reality, any deviation from the normal radiographic findings (i.e. well-­ developed teardrop, normal femoral neck-shaft angle, intact Shenton line, downsloping sourcil, and well-developed Gothic arch) of a mature pelvis may lead to degenerative joint disease over time, although a definitive correlation has not been demonstrated [55, 56, 64–68, 76]. Physical signs are often not present in children with radiographic hip dysplasia. The diagnosis is commonly made incidentally or only after the patient develops symptoms related to the dysplasia [28, 29, 51, 68, 77]. Approximately half of patients with degenerative joint disease secondary to hip dysplasia have evidence of radiographic dysplasia in the asymptomatic contralateral hip [68, 78–80]. The severity of subluxation has been shown to correlate with peak periods of pain by Wedge and Wasylenko [55]. They demonstrated that patients with the most

4  Developmental Dysplasia of the Hip in Young Children

severe subluxation developed pain in the second decade of life, those with moderate subluxation developed pain in the third and fourth decades, and those with minimal subluxation developed pain in the fifth decade. Severe degenerative changes on radiographs typically become evident approximately 10  years after the onset of symptoms. A completely dislocated hip typically causes symptoms much later than a subluxated hip [60, 81, 82].

 esidual Femoral and Acetabular R Dysplasia After Treatment The goal of treatment of DDH is to have a radiographically confirmed normal hip at skeletal maturity in order to prevent degenerative joint disease in the future. Attempts must be made to correct any amount of hip subluxation identified. Additionally, attempts should be made to correct acetabular dysplasia in the preadolescent phase as this eventually leads to degenerative joint disease even in the absence of subluxation [31, 32]. The treatment of hip dysplasia identified in the adolescent phase remains controversial and will be discussed in the following chapter.

Epidemiology Causes of DDH Developmental dysplasia of the hip is multifactorial. Genetic and ethnic factors play a clear role as the incidence of DDH is very low among children of African and Chinese descent and as high as 50 per 1000 live births among Native Americans and Lapps [33, 54, 55, 62, 83–94]. A positive family history of DDH has been reported in up to one in three children with DDH [34, 83, 94]. Phenotypic aspects including femoral acetabular anteversion, primary acetabular dysplasia, various degrees of joint laxity, or a combination of all may contribute to the development of DDH [33, 37]. Antenatal mechanical factors and certain neu-

81

romuscular conditions have been demonstrated to profoundly influence the genetically determined intrauterine growth and development of the fetal hip. These factors include breech position, oligohydramnios and myelomeningocele [7, 8, 95, 96].

Risk Factors and Incidence As a multifactorial condition, clear cause and effect has not demonstrated by any single factor, however, several authors and meta-analyses have highlighted associations between DDH and breech positioning, a positive family history, sex, and first-born status [97, 98]. First born children are more likely to be affected with DDH than subsequent children in white populations [11, 28, 29, 99–102]. The theory that the “crowding phenomenon” plays a role in the pathogenesis of DDH is supported by the higher incidence of DDH in twin pregnancies and first born children as the mother’s unstretched abdominal muscles and primigravid uterus may subject the fetus to prolonged periods of increased pressure and abnormal positioning. As a result, the fetus is forced up against the mother’s spine and pelvis, limiting fetal hip abduction. This positioning theory is further supported by the predilection for left hip involvement in DDH.  Most often, this hip is forced into adduction against the mother’s sacrum in one of the most common fetal positions [11, 35, 54]. Additionally, other intrauterine molding abnormalities associated with “crowding”—including torticollis, metatarsus adductus, and oligohydramnios—are all associated with DDH [11, 54]. DDH is more common in girls (80%) and among children delivered in the breech presentation. Twice as many girls are born breech and 17–23% of children with DDH had a breech presentation, compared to only 2–4% of the general population [28, 29, 77, 103]. Further, over half of breech presentations are first-born children, again suggesting a positive association between sex, breech positioning, first-born status, and DDH [28, 29, 77].

S. L. Weinstein and J. B. Holt

82

The postnatal environment has also been suggested to influence the development of DDH. Most clearly, societies that swaddle newborns with hips maximally adducted and extended in the immediate postnatal period have a very high incidence of DDH [53–55, 70, 104, 105]. The effects of capsular laxity on the development of DDH have been debated. While newborns with DDH do have capsular laxity, it is more likely that this is the result of hip instability rather than the cause. Investigators have argued in favor of laxity as a cause of DDH by citing the fact that secondary reversible acetabular dysplasia can be created in animals by primarily producing ligamentous laxity [25, 26, 34, 37, 103, 106]. Conversely, others have shown that the acetabulum is most shallow at birth and that with the normal laxity of the newborn hip, it is at the highest risk of dislocation during this period [106–108]. Hip capsule laxity has been demonstrated in normal infant hips and may even allow mild instability. During dynamic imaging examinations of Ortolani negative stillborn hips, despite mild contrast dye pooling on arthrogram and the presence of capsular laxity during ultrasonography, all were found to be pathologically normal. This is in distinct contrast to the postmortem findings associated with infants with positive Ortolani signs which showed a cartilaginous ridge separating the hip socket into two sections, inverted labrum, and degeneration of the acetabular cartilage [12, 14, 36, 69, 91, 109]. Further, DDH is not a characteristic feature of conditions associated with hyperlaxity such as Down, Ehlers-Danlos, and Marfan syndromes [36]. Patients at high-risk for development of DDH can be identified by taking into account the epidemiologic, ethnic, and diagnostic criteria as described above. Health providers should be alert to the possibility of DDH in patients presenting with risk factors known to be associated with the diagnosis, most importantly a positive family history and breech positioning. “Health providers should be alert to the possibility of DDH in patients presenting with risk factors known to be associated with the diagnosis; most importantly a positive family history and breech positioning”.

Risk Factors for DDH

1. Primary risk factors (a) Positive family history or ethnic background (Native American, Laplander) (b) Breech presentation 2. Secondary risk factors (a) Female gender (b) Torticollis (c) Metatarsus adductus (d) Oligohydramnios (e) Persistent hip asymmetry

Clinical Presentation Early Diagnosis An expert and artful examination of the infant’s hips is a key diagnostic tool in the early diagnosis of DDH. It must occur in a controlled setting and environment by an experienced examiner. A relaxed child should be positioned on a firm surface if possible; otherwise the parent’s lap will suffice. Movement of the hip from a reduced to a subluxated or dislocated position is a delicate maneuver and relies on the light touch of the examiner. The tight fit between the femoral head and the acetabulum is lost in the newborn with DDH. This is manifest clinically as the ability of the examiner to make the femoral head slide in and out of the acetabulum, with a palpable sensation commonly known as the Ortolani sign [12, 20, 36, 37]. Although some have tried to distinguish the sensation of manual reduction of a dislocated or subluxated hip (Ortolani sign) from that of dislocating or subluxating a located hip (Barlow sign), the authors prefer to refer to any hip that can be subluxated or dislocated and subsequently reduced as being Ortolani positive, as initially described by Ortolani [36, 52] (Fig. 4.4). It is the opinion of the authors that distinguishing between hips that can be reduced from a dislocated position from those that can be dislocated from a reduced position is of no clinical importance as the clinical treatment of these hips is identical.

4  Developmental Dysplasia of the Hip in Young Children

a

83

b

c Adduct

Barlow Test

Out

In

d

e

f Abduct

Ortolani Test

Out

In ©JSchoenecker2018

Fig. 4.4  Diagram of the Barlow (upper) and Ortolani (lower) tests. (a) With the infant supine the examiner gently holds the leg in neutral resting position and the femoral head is located within the acetabulum. (b) The hip is gently adducted. (c) With a posteriorly directed force the hip is dislocated as the examiner feels the femoral head glide over the posterior aspect of the acetabulum. (d) While the

infant rests supine and the examiner holding the leg in neutral position the hip is resting in the dislocated position. (e) The combination of gentle abduction of the hip and (f) anteriorly directed pressure on the greater trochanter will reduce the femoral head back into the acetabulum as the examiner feels a palpable “clunk” of the femoral head gliding back into the acetabulum

“Distinguishing between hips that can be reduced from a dislocated position from those that can be dislocated from a reduced position is of no clinical importance as the clinical treatment of these hips is identical”.

French and then Italian literature as a palpable sensation was, unfortunately, translated into English as a “click”. Unlike the palpable “clunk” described by LeDamany and Ortolani, high-pitched soft tissue related clicks are commonly experienced when examining the hips of normal newborns. These are often transmitted from the iliopsoas, the trochanteric region, or the knee, and generally have no diagnostic significance [110–113]. Unfortunately, a pervasive, poor understanding of the underlying pathoanatomy of DDH and its diagnosis by clinical examination has led to over diagnosis and over treatment of infants [114–117]. Contrary to the Ortolani test, whereby a dislocated hip is reduced into the acetabulum, the Barlow maneuver is a provocative maneuver in which the

In his original description in 1912, LeDamany referred to the palpable sensation of the hip gliding in or out of the acetabulum as the signe de ressaut and later, in 1936, Ortolani described it as the segno dello scotto, providing a description of the pathogenesis of this exam finding [36, 107, 109]. Ortolani named the hypertrophied ridge of acetabular cartilage responsible for the finding as the neolimbus. This key diagnostic sign first described in

S. L. Weinstein and J. B. Holt

84

hip is flexed and adducted and the femoral head is palpated to subluxate or dislocate over the acetabular ridge under gentle axial pressure [52]. Some providers have suggested that a reduced hip that is dislocatable (Barlow positive) is more stable than a hip that is dislocated and is reducible on examination (Ortolani positive) and therefore may spontaneously stabilize. The authors, however, agree with the original work of LeDamany and Ortolani who described the palpable sensation of both subluxating or dislocating a reduced hip or reducing a subluxated or dislocating as pathologic and make no distinction in treatment between them. Idiopathic teratologic (i.e. irreducible) dislocations are extremely rare in newborns. Irreducible hips are most often associated with neuromuscular conditions, such as myelodysplasia, arthrogryposis, or with syndromes; accounting for only 2% of DDH cases in large series [12, 54, 62, 118, 119]. The secondary adaptive changes that are often present in these cases are more typical of those seen in the late-diagnosed DDH case.

Late Diagnosis If the diagnosis of DDH is not made early, secondary adaptive changes will reliably develop [84]. The most common physical finding associated with late-diagnosis is limited abduction, the clinical manifestation of a shortened adductor longus associated with hip subluxation or dislocation [62]. Additional findings include apparent femoral shortening, called the Galeazzi sign, asymmetric gluteal, thigh, or labial folds, high riding greater trochanter (Klisic test), and limb length inequality [5, 120, 121]. When bilateral dislocations are present, patients often exhibit a waddling gait and hyperlordosis of the lumbar spine. “The most common physical finding associated with late-diagnosis is limited abduction, the clinical manifestation of a shortened adductor longus associated with hip subluxation or dislocation”. Normal hip joint growth and development are impaired if DDH is not detected at an early stage. Older patients at the time of detection, particu-

larly those detected beyond 6  months of age, often do not respond to simple treatment methods such as the Pavlik harness and instead require more substantial interventions including a closed or open reduction (requiring general anesthesia).

Essential Clinical Tests

1. Skillful performance of Ortolani and Barlow maneuvers 2. Assessment for symmetric hip abduction 3. Assessment for Galeazzi sign and limb length asymmetry 4. Assessment for symmetric gluteal, thigh, and labial folds

Imaging Utility of Ultrasound in Hip Dysplasia Ultrasonography as a diagnostic or screening exam in DDH has gained worldwide popularity. Although some controversy exists, its cost-­ effectiveness as a universal screening tool for DDH has yet to be clearly demonstrated [122– 132]. The earliest advocate of the use of ultrasonography in orthopaedics was Graf in Austria in the 1970s and many proponents of ultrasound argue that it should be used in all newborns as a standard screening tool for DDH [133–137]. Ultrasound use in DDH is unique in that it can provide both a morphologic assessment and a dynamic evaluation [138–143]. The morphologic assessment includes an evaluation of the anatomic characteristics of the hip joint including the determination of two primary angles: the α-angle and β-angle (Figs.  4.5 and 4.6). The α-angle is a measure of the slope of the superior aspect of the bony acetabulum while the β-angle describes the cartilaginous component more laterally. The dynamic aspect of the examination, as popularized by Harcke and colleagues, provides a real-time assessment of the hip during range of motion and stress examinations (e.g. Ortolani and Barlow maneuvers) [144]. The anatomic classification of infant hip dysplasia, initially

4  Developmental Dysplasia of the Hip in Young Children

85

a

b

Fig. 4.5  Hip types based on ultrasonographic results, according to the Graf classification. (a) Correct hip ultrasound images must fulfil certain criteria. These criteria are divided into two usability checklists: (i) the identification of 6 landmarks (numbered in the figure): (1) chondro-­osseous junction, (2) femoral head, (3) synovial fold, (4) joint capsule, (5) labrum, (6) cartilagenous part of the acetabular roof, (7) bony roof, and (8) the plane (i.e. ilium, which must look like a straight line); (ii) the identification of three landmarks (marked with yellow arrows) are essential to establish the standard plane which is required for accurate measurement of infant hip ultrasound: the lower limb of the bony roof (usually it is the brightest and largest lower end of the bony roof), and the midportions of both the Ilium and the labrum. If any of these points are missing or not clearly shown, the sonogram is worthless and should not be used. The only exception is when the joint is decentered (dislocated) (Graf IV). (b) When ultrasound images are accepted based on the criteria mentioned above, the angles can be accurately measured. It is important

to outline the labrum, mark its centre, and identify the ‘turning point’ (marked with a yellow arrow). The turning point—confusingly, also called the ‘bony rim’—is the most lateral point of the concave bony socket. It is the turning point from concavity to convexity. It is essential to look for the turning point from inner to outer (i.e. distal medial to proximal lateral). There is frequently a small acoustic shadow (purple dashed arrow) just medial to the turning point. Using the turning point as a reference, three independent lines are drawn as follows: the baseline runs tangential to the outer surface of the plane (Ilium), where the cartilaginous roof meets the Ilium. The bony roof line runs tangentially from the lower limb to the turning point. The cartilage roof line is drawn from the turning point to the center of the labrum. The angle between the bony roof line and the base line is the α-angle, whereas the angle between the cartilage roof line and the base line is the β-angle. Note that these lines are often do not meet at the same point. (c) Hip types based on ultrasonographic results, according to the Graf classification. (Courtesy Sattar Alshyrda)

S. L. Weinstein and J. B. Holt

86 Type Alpha angle (α) > 60°

I

50–59°

II

D

43–49°

III

< 43°

IIc

Ia

> 55°

Ib

Normal hip (at any age). This grade is further divided into (la; β 55°). The significance of this subdivision is not yet established. Patient dose not need follow-up. If the child is < 3 months. This may be physiological and does not need treatment; however, Follow up is required.

< 77° Stable Unstable

< 77° > 77°

IIIa IIIb IV

Descriptions

< 55°

< 77°

IIa IIb

43–49°

Beta angle (β)

< 43°

> 3 months, delayed ossification. Critical zone, labrum not everted. This is further divided into stable and unstable by provocation test. This is the first stage where the hip becomes decentred (subluxed). it used to be called IId, but for the above reason, it is a stage on its own now. Dislocated femoral head with the cartilaginous acetabular roof is pushed upwards. This is further divided into Illa and IIIb depending on the echogenicity of the hyaline cartilage of the acetabular roof (usually compared to the femoral head) which reflects the degenerative changes. Dislocated femoral head with the cartilaginous acetabular roof is pushed downwards.

Fig. 4.5 (continued)

a

b

c

d

Fig. 4.6  Hip ultrasound. (a) Anatomy of a normal newborn demonstrated. Note the sharp edge of the ilium and the echogenic triangular labral fibrocartilage (*). (b) Anatomy of a newborn with dysplasia. Note the rounded edge of the ilium. (c) Normal anatomy with alpha (α) and

beta (β) angles shown as the angles created by base line (white line along ilium), the inclination line (red line along labrum) and the roof line (yellow line along acetabular roof). (d) Abnormal anatomy of a newborn with dysplasia shows increased beta angle and decreased alpha angle

4  Developmental Dysplasia of the Hip in Young Children

introduced by Graf, is widely accepted and remains commonly used throughout the world today. In this system, hips are classified into four types and several subtypes according to various factors including the bony and cartilaginous components of the acetabulum, percentage of acetabular coverage, and shape of the superior bony rim. Hips are then designated as type I (a mature hip with α-angle >60°) through type IV (a dislocated hip with α-angle 0.20, acetabular depth to width index 40% being first order relatives. First order relatives of adolescent/adult diagnosed patients had a twofold increase in incidence of hip replacement by age 65 compared to infant/ DDH first order relatives (50% vs. 22%). However, first order family members of infant/ DDH patients were four times more likely to have DDH themselves (59% vs. 16%). “Rates of acetabular dysplasia vary widely by gender and cultural origin with several well accepted patient characteristics being predictive of increased risk during infancy. Once a dysplastic hip is present, the necessity of surgical treatment is directly related to the relative severity of symptoms”.

Clinical Presentation Young children with hip dysplasia with or without subluxation typically do not present with any complaints of hip pain or any apparent functional limitations. More likely, children in the first 6–8 years of life will have relatively normal hip function despite radiographic evidence of notable hip dysplasia. However, with growth and increased body mass during and post puberty, older children and adolescents with acetabular deficiency will variably become symptomatic, particularly in the presence of hip joint subluxation. Signs include functional hip joint associated fatigue and often a subtle, though progressive, gluteus medius weakness limp (Trendelenburg). This early onset weight bearing pain typically is located laterally (trochanteric) and occurs secondary to chronic gluteus medius fatigue. Later, anterolateral groin pain secondary to chondrolabral strain and or injury variably develops. In these relatively older patients, groin pain is characteristically described as an anterior hip centered discomfort often with catching or snapping. Groin pain typically occurs with hip motion such as in pivoting,

J. G. Schoenecker et al.

142

twisting, running, when arising from a sitting position, and on initiating walking. On physical examination, the gait of most of the younger children will appear to be normal. On careful observation of older children and or adolescents, there may be an abductor limp and/ or a positive Trendelenburg sign on single leg stance examination. The hip abductors may be weak on resistance testing. Passive hip range of motion is initially normal and often increased in all planes, secondary to both relative joint laxity and a variably deficient anterolateral acetabulum. Later as the hip disease progresses, range of motion can become restricted, associated with the onset of painful impingement at the acetabular chondrolabral junction.

Essential Clinical Findings

• Residual radiographic hip dysplasia is typically asymptomatic in younger children • With growth and development, patients with hip instability initially note abductor fatigue and associated limping; with progression, anterior groin pain can occur • On exam, gait in the younger patient will appear normal; with time the Trendelenberg test becomes positive • Hip range of motion initially will be normal or slightly increased in flexion, internal rotation and abduction

Imaging Hip joint dysplasia in the child occurs secondary to abnormal growth of either the acetabulum, femoral head, or both, in a non-dislocated hip joint. The shape of the acetabular anlage and its functional capacity to support weight bearing is evaluated with a standing anterior posterior (AP) radiograph. Typically the acetabular development will be deficient anterolaterally. In addition, the absence (or presence) of hip joint subluxation (i.e. Shenton’s line intact or not) is assessed. The contour of the anterior and posterior edges of the proximal femurs are visualized on the supine

frog lateral radiographs. In older children and adolescents, further visualization of the anterior acetabulum can be seen with the standing false profile lateral radiograph [4]. Persistent hip subluxation after the age of 6 years portends a guarded prognosis for maximal acetabular development for any given hip dysplasia [9]. If subluxation is noted on the standing AP pelvic radiograph, a functional view (AP with hip abducted and internally rotated) is obtained to assess if the femoral head positionally reduces back into the true acetabulum. Knowing that positional reduction of femoral head subluxation is possible, is very helpful in the preoperative planning of joint reconstruction surgery.

Essential Imaging

• Plain radiographs: For measurement of AI, LCEA: Tonnis sourcil angle, teardrop morphology, and femoral deformity • MRI: Used on occasion to assess development of the cartilaginous acetabulum and/or version • MRI arthrogram (MRA): For assessment of chondrolabral integrity in adolescents • CT: For either assessment of version (femoral and acetabular) or for concomitant FAI

Normative values for acetabular index (AI) [14], lateral center edge angle (LCEA) [14, 34], Tonnis sourcil angle [4, 14] are helpful in decision-making and are summarized in Tables 5.1 and 5.2. We also assess the anterior CEA [4, 14] in our older patients. The acetabular tear drop morphology can be useful in assessing acetabular development in late infancy/early childhood Table 5.1  Suggested acetabular index (AI) guideline values [14] Age 1 3 6

Acetabular index 10,000/mm3, and 20 (ex 94) had an ESR > 20 mm/h. Only two patients had long-term hip pathology. The author commented that if there was definite radiological abnormality at presentation, or ‘if the child’s symptoms are prolonged’, long-term complications were much more likely.

340

Key Evidence

J. S. Huntley

establish causality over effect i.e. activation of the interferon system might occur secondary to There have been a number of useful pragmatic TS, or indeed to another insult (non-viral) which narrative reviews of hip TS [1, 32, 43, 54, 65]. simultaneously caused TS. Landin LA, Danielsson LG, Wattsgard C. There is some evidence concerning epidemiol1987. Transient synovitis of the hip. Its inciogy [12, 13, 18], a systematic review on the dence, epidemiology and relation to Perthes’ course, recurrence rate and outcome of TS [9], disease. J Bone Joint Surg Br 69:238–42. This one double-blind randomised controlled trial for treatment with ibuprofen [15], multiple studies prospective 5  year study (1978–1982) followed on predictive algorithms [45, 46, 48–51, 66, 67], children (1–13 years) with a clinical diagnosis of and two brief systematic reviews of studies dis- hip “transient synovitis”, in a well-defined poputinguishing between transient synovitis and sep- lation in Malmo, Sweden. Importantly there was tic arthritis [39, 47]. Synopses of key studies are agreement that all children with limp/other hip symptoms should be referred for emergency given below. Leibowitz E, Levin S, Torten J, Meyer R. assessment. All 294 episodes (in 275 children) 1985. Interferon system in acute transient had: (1) a history of 1–7 days pain with limp or synovitis. Arch Dis Child 60:959–62. This study reluctance to weightbear, (2) restricted hip movesought to compare the ‘interferon system’ of 20 ment, (3) anteroposterior and frog-lateral radiochildren (13 boys, 7 girls; age (years) mean 5.3, graphs, and (4) full blood count and ESR.  If range 1–17.9) with a clinical diagnosis of tran- admitted, throat swabs and urine samples were sient synovitis (but 2 had symptoms for 6–8 weeks cultured, and additional bloods taken for serolbeforehand; 6 (30%) had recurrence), with non-­ ogy and CRP, with skin traction usually applied. matched ‘controls’ (74 healthy individuals of all Incidence by age was an inverted “V”, peaking at ages and both sexes; blood assayed ‘over the past 5 years, with an average annual rate of 20.3/10,000 few years’). They examined (1) plasma interferon (0.2%), and a boy:girl ratio of 2.6:1. The incilevels, (2) in vitro production of interferons α and dence of recurrence was 400/10,000/year (4%) γ by peripheral mononuclear cells, (3) antiviral i.e. a 20-fold risk compared with that of a first state of peripheral mononuclear cells (an assay episode. There was an October–November preinvolving cell infection with vesicular stomatitis ponderance, held to suggest an infectious (as virus, and subsequent supernatant addition to opposed to traumatic) aetiology. In 10 cases indicator fibroblast monolayers). The data indi- (3.7%), Perthes disease was diagnosed cated (1) higher in vivo blood interferon levels in 1–5 months after initial symptoms; on review of the TS group, and (2) a greater proportion of sub- the initial radiography, specific (i.e. not present in jects (TS as opposed to ‘control’) with “cells in TS patients who did not develop Perthes an antiviral state”, this being 14/18 (78%) com- ­condition) radiological findings (increased apparpared with 10/74 (13%), and (3) no significant ent joint space/decreased epiphyseal size/subdifference for in vitro production of either inter- chondral lucency) were present in 8/10. In the feron α or γ. There are major methodological 5  year period, 24 patients were diagnosed with concerns—in particular (a) case definition (two Perthes disease; of the 10 cases of Perthes disease patients with extended symptoms, and no ultra- initially diagnosed as “transient synovitis”, nine sound definition of effusion/synovial thickening), had been immobilised in traction, requiring lonand (b) the controls being non-matched, and not ger (mean 6  days, range 1–13) to recover than performed in contemporaneous assay. The con- other (mean 3.3 days, range 1–7) traction cases. clusion that the ‘findings suggest that the aetiol- The authors raised the possibility that increased ogy of transient synovitis is an acute viral intra-­articular pressure due to extension-splintinfection’ is an interesting possibility, but remains ing/traction might have harmed the 2 cases withspeculation, especially given that even if the out initial radiographic changes who subsequently association were confirmed, it would hardly developed Perthes’ disease.

12  Transient Synovitis

Bickerstaff DR, Neal LM, Brennan PO, Bell MJ. 1991. An investigation into the etiology of irritable hip. Clin Pediatr 30:353–6. This study involved a prospective series of 111 children (boy:girl, 2.2:1) presenting with ‘acute hip pain’, April 1987 to January 1988. Investigations included blood tests (full blood count, ESR, ASOT, rheumatoid factor), throat swabs, midstream urine (culture), hip ultrasound and X-rays. Of 111, 26 and 85 were treated as in- and out-patients, respectively. Thirty-three children (30%) only reported antecedent infection (25 ex 30 were upper respiratory in nature); only 7 (6%) had a ‘significant episode of trauma’; in 8 (7%) had presenting temperature > 37.5 °C. No haematological investigation was significantly different from normal. Of the 111, 79 (71%) had an effusion on ultrasound— this persisted in 6, and 2 of these later developed radiographic features of Perthes disease. The authors compared their data to the pooled reports of 1843 cases of ‘irritable hip’ reported in the literature (the mean age from the comparator group being 5.9 years, range 3–10 years, with a boy:girl ratio of 2:1). They advocated a ‘more selective use of investigations, based on clinical suspicion’, finding that the common investigations were usually not helpful in establishing aetiology, or management, of the irritable hip. Fischer SU, Beattie TF. 1999. The limping child: epidemiology, assessment and outcome. J Bone Joint Surg Br 81B:1029–34. This prospective 6  month study (1 January to 30 June, 1996) followed 243 children (age 1–14  years), from presentation at the emergency department with acute limp (painful or painless)/gait abnormality/refusal to walk or weight-bear/acute hip/ groin/thigh/knee or lower-leg pain, in a well-­ defined population in Edinburgh, Scotland. Patients were excluded if they had either (1) longstanding orthopaedic or neurological conditions, or (2) overtly traumatic conditions. A protocol for initial investigations included temperature, full blood count and ESR.  Radiographs were performed if indicated after history and exam; if hip abnormality was suspected a hip ultrasound was performed—transient synovitis was only diagnosed if there was

341

synovial thickening, an effusion or both. ‘Irritable hip’ was defined as hip abnormality on examination but not on imaging. All notes were re-­ evaluated at 18–21  months post-presentation. The incidence rate of limp was 18/10,000 (0.18%), with a boy:girl ratio of 1.7:1 (152:91). The commonest diagnoses were ‘irritable hip’ (39/243; 16%) and transient synovitis (23.5%), together comprising 39.5%. However, in a further 72/243 (30%) no final diagnosis was achieved—though these patients recovered fully. Five patients had Perthes disease, with an incidence after ‘irritable hip’/transient synovitis of 2%. A table of differential diagnoses was constructed. Using this basic assessment protocol, most patients presenting with acute atraumatic limp were discharged from the department (94%), and only 25% required management outwith the emergency department. Kocher MS, Zurakowski D, Kasser JR. 1999. Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm. J Bone Joint Surg Am 81:1662–70. This study involved a retrospective review of children (n = 168; after 114 atypical group patients were excluded: immunocompromise, renal failure, neonatal sepsis, postoperative hip infection, rheumatological disease, Perthes’ disease, osteomyelitis) evaluated at a tertiary care hospital over a 17  year period (1979–1996) with true SA (n = 38), presumed SA (n = 44) and TS (n = 86), all of whom had undergone blood investigations (full blood count and ESR), blood cultures and aspiration of the hip joint fluid (for (1) white cells > 50 × 103/mm3, and (2) culture). Univariate and multiple logistic regression analyses were used to probe the differences between the groups, identifying a set of four independent multivariate predictors: (1) history of fever, (2) non-weight-­ bearing, (3) ESR  >  40  mm/h, and (iv) WBC > 12 × 103 cells/mm3. For a patient in this dataset, the probability of SA is given by the number of positive predictors: 0 (0.2%), 1 (3%), 2 (40%), 3 (93.1%), 4 (99.6%). Kermond S, Fink M, Graham K, Carlin JB, Barnett P. 2002. A randomised clinical trial: should the child with transient synovitis of the

342

J. S. Huntley

hip be treated with nonsteroidal anti-­ tic concurrence, although there was a slight shift inflammatory drugs? Ann Emerg Med for patients in this second dataset—the proba40:294–9. This small (n = 36) double blind ran- bility of SA given by the number of positive predomised placebo-controlled trial compared ibu- dictors was: 0 (2%), 1 (9.5%), 2 (35%), 3 (73%), profen (n  =  17) versus placebo (n  =  19) 4 (93%). (Attention is drawn to the figure of for children [aged 1–11  years; median 5  years 9.5% for just one predictive factor, compared to (ibuprofen group) and 7  years (placebo group)] 3% in the original study). with a clinical diagnosis of hip transient synoviLuhmann SJ, Jones A, Schootman M, tis, whether or not this was ultrasound positive Gordon JE, Schoenecker PL, Luhmann JD. [effusion on initial ultrasound: 11/17 (65%; ibu- 2004. Differentiation between septic arthritis profen) and 17/19 (89%; placebo)]. The exclu- and transient synovitis of the hip in children sion criteria were (1) age less than 12 months or with clinical prediction algorithms. J Bone greater than 12 years, (2) symptoms longer than Joint Surg 86:956–62. Retrospective view 1 week, (3) fever >38 °C in the emergency depart- assessing Kocher’s diagnostic criteria on a conment, (4) WBC  >  15  ×  103  cells/mm3, (5) secutive series of 163 patients (165 hips; boys, ESR > 25 mm/h, (6) pathology on hip radiograph, girls; 1992–2000, at a single institution) who (7) hospital admission required, (8) known hip underwent hip arthrocentesis for evaluation of pathology, (9) ibuprofen contraindicated. There irritable hip. Results were 20 (true SA), 27 (prewere two possible confounders in terms of het- sumed SA), 118 (TS). Notably, 18 patients with erogeneities between the groups: (a) age distribu- synovial fluid white cell count  B

Freq (%) 4–17

Oligo- (oJIA) (a) Persistent (b) Extended

JIA subtype Systemic (sJIA)

ANA –

Acute (7%) Usually symptomatic

Rare ( B

2–11

Psoriatic (PsA)

13  Juvenile Idiopathic Arthritis and the Hip 349

J. S. Huntley et al.

350

and longevity. Very little robust literature exists to offer guidance in terms of decisions or outcomes, nor to guide surgeons in operative strategy, implant choice and future planning. “Total hip arthroplasty (THA) is rarely undertaken in children, especially before skeletal maturity, but it is the key surgical option”.

Pathophysiology Juvenile idiopathic arthritis (JIA) is an umbrella term encompassing a heterogeneous group of inflammatory conditions affecting articular cartilage, (and variously other extra-articular systems), characterised by persistent arthritis in any joint for ≥6  weeks [14]. The cytokine tumour necrosis factor alpha (TNFα) was implicated in rheumatoid arthritis, after being identified at significant levels in the majority of synovial fluid samples from Rheumatoid Factor (RhF)-positive patients [15], and subsequently in synovial cells following arthrocentesis of JIA patients [16]. Oligo- and poly- arthritis are cell-mediated autoimmune diseases, with many T-helper-1 (Th-­1) cells in the inflamed synovium of affected joints, together with increased levels of Th-1 cytokines [17]. In JIA, the affected synovial membranes are hypertrophied and vascularised. There are activated T- and B- cells, plasma cells and macrophages. Host cells, including activated osteoclasts and fibroblasts, mediate bone and cartilage erosion, involving a network of cytokines (particularly IL-1, IL-6, TNF-α; but also IL-21, IL-23 and IL-17) and matrix metalloproteinases [17]. For the systemic subtype, the systemic features (fever, evanescent rash, hepato-­ splenomegaly), association with HLA alleles [18] and absence of auto-antibodies (or autoreactive T-cells) suggest that this subtype is a distinct entity, sharing more with autoinflammatory than autoimmune disease per se [1]. A link was shown to raised interleukin(IL)-6 (Benedetti et  al. [19, 20]), which provided the rationale for the observed clinical efficacy of tocilizumab [21].

Interleukin-1 (IL-1) has been similarly implicated [22] but the role of IL-1 inhibition in sJIA is in its early phase, not least because of the heterogeneity of response. Three IL-1 antagonists—anakinra [22, 23], canakinumab [24] and rilonacept [25]—appear to be effective and ‘safe’ at short-term follow-up [26]. The hip is rarely involved as a sentinel joint. In early disease, synovial hypertrophy and effusion cause pain and muscle spasm. Subsequently fibrosis and periarticular changes can cause fixed flexion and adduction contractures [13].

Natural History There have been several long-term studies of JIA progression into adulthood [29–34]. These are awkward to compare because of different outcome measures and inconsistent definitions of disability/remission [34]. Furthermore, treatment regimens have changed dramatically over this timescale. In the Norwegian cohort, Flato and colleagues [29] found that continued elevation of ESR in the first 6  months after presentation predicted persistent active disease at 15 years. Disease status is often given by the European League against Rheumatism (EULAR) criteria, in which there are four groups ([35], after [36, 37]): 1. Active—increasing number of active joints irrespective of drug treatment. 2. Stable—stable number of joints but requiring drug therapy. 3. Inactive—no evidence of active synovitis and/ or active extra-articular features without drugs for 2  mm displacement of both anterior and posterior pelvic ring fractures (Fig. 16.8). They claimed that there is a relation between fracture type and length of hospital stay as well as transfusion requirement and the need for pelvic ring fixation. Type III-B injuries were more likely to receive blood products than type III-A fractures and therefore this modification would be predictive for significant morbidity and death. Ogden [17] classified pediatric pelvic disruptions into either avulsion injuries, stable pelvic ring fracture patterns, unstable pelvic ring fracture patterns or triradiate cartilage injuries. Stable pelvic fracture patterns include: (1) infolding of the iliac wing, (2) iliac crest apophysis avulsion or (3) isolated ischiopubic rami frac-

tures. Unstable pelvic fractures include: (1) unstable rami fractures accompanied with displacement of the ischiopubic fragment from the symphysis and (2) anterior symphyseal diastasis or pubic rami fractures and posterior sacroiliac joint disruption. Other used classification systems are those used for adult pelvic trauma which are Tile’s/AO classification [11] and Young and Burgess classification system [8]. Tile’s classification system depends upon the concept of pelvic stability, force direction and the resultant pathoanatomy. The pelvis is divided into anterior arch located anterior to the acetabular surface and posterior arch located posterior to the acetabular surface. Pelvic fractures are generally divided in to Type A fractures which are stable fractures, Type B which are partially stable and Type C unstable pelvic fractures (Table 16.2 and Fig. 16.9). Young and Burgess classification [8] is based on the mechanism of injury and is used to alert the treating surgeon to the potential associated injuries and resuscitation requirements. Pelvic ring injuries are classified into lateral compression LC injuries, anteroposterior compression APC, vertical shear VS and combined mechanisms CM. Both LC and ACP injuries are the commonest encountered pelvic ring injuries and each is subdivided into 3 subsets based on increasing severity of injury. The reliability of using Tile and Young— Burgess classification systems in adult pelvic trauma is questionable [33, 34]. While Furey et al [33] reported moderate interobserver agreement of using Tile’s and Young-Burgess classifications and Kappa scores of 0.46 (0.39–0.52) and 0.47 (0.42–0.52) respectively, Gabbe et al [34] found

Table 16.1  Torode and Zieg classification Type I

Avulsion fractures

Type II Type III

Iliac wing fractures Simple ring fractures

Type IV

Ring disruption fractures

Avulsion of the growth plates around the pelvis like ischial tuberosity, AIIS, iliac crest apophysis. Iliac wing infolding or disruption of the iliac crest apophysis Pubic rami or the disruption of the pubic symphysis without clinical instability of another pelvic segment – Bilateral pubic rami fractures (straddle fracture). – Fractures involving pubic rami or disruption of the pubic symphysis with associated posterior pelvic ring fracture or disruption of the sacroiliac joint. – Fractures involving anterior pelvic ring and the acetabular segment.

a4

c2

a3

c1

Fig. 16.8  Modified Torode and Zieg classification system. (a) type I avulsion injuries, (a1) Avulsion injury of the anterior superior iliac spine ASIS, (a2) Avulsion of the anterior inferior iliac spine, (a3) Avulsion of the ischial tuberosity, (a4) Avulsion of the

a2

a1

©JSchoenecker2018

lesser trochanter. (b) type II iliac wing fractures. (c) Type III Simple ring fractures, (c1) Stable anterior ring fractures—Pubic rami fracture, (c2) Stable anterior and posterior pelvic ring disruptions. (d) type IV Unstable pelvic ring disruptions

d

b

420 M. Kenawey

16  Pediatric Pelvic Injuries

421

Table 16.2  Tile’s classification Type A A1 A2 A3 Type B B1 B2 B3 Type C C1 C2 C3

Stable pelvic ring injury Fractures not involving the ring (avulsion fractures, iliac wing fractures). Iliac wing or anterior arch injury caused by direct blow Transverse sacral fracture Partially stable (incomplete disruption of the posterior arch) Open book injury (external rotation) Lateral compression injury (internal rotation) Bilateral Unstable (complete disruption of the posterior arch) Unilateral Bilateral with one side type B and one side type C Bilateral type C

that the interobserver reliability of both systems was poor and Kappa scores ranged from 0.12 to 0.17 and 0.09 to 0.14 respectively. On the other hand, Torode and Zieg classification is a simple system that was developed for use in pediatric pelvic trauma. Silber et al [30] evaluated the interobserver reliability of using Torode and Zieg classification system and taking treatment decisions based on plain radiographs with or without CT scans. The agreement scores were almost perfect with plain radiographs and the addition of CT scans didn’t significantly change the classification or management plans. Our experience with pediatric pelvic trauma and the use of Tile’s system and Torode and Zieg classification showed as well better intra and interobserver agreement for Torode and Zieg classification compared to Tile’s and Young-­ Burgess classifications and this agreement was not affected by the addition of CT studies to plain pelvic radiographs, however as trauma surgeons, we do need Tile and Young-Burgess classifications to guide our treatment choices (unpublished data).

Clinical Evaluation Evaluation of a child with suspected pelvic ring disruption should start with the ATLS algorithm to provide a systematic and organized approach for resuscitating such critically injured patients [35]. Following the primary survey and the resuscitative phase, a careful physical examina-

tion is conducted to evaluate areas of ecchymosis, superficial lacerations or closed degloving lesions (Morel-Lavallée lesions) around the pelvic area. Morel-Lavellee lesions are particularly important because of their closed nature in the presence of other distracting injuries in polytrauma patients [36]. Despite being closed injuries, bacterial colonization has been reported in up to 46% of sampled lesions which was independent from the time of injury with the increased risk of surgical site infection with the operative treatment of associated pelvic and acetabular injuries [37]. In addition, the presence of open wounds around the pelvis and perineum, bleeding from the urethral meatus (indicative of bladder or urethral injury), or rectal bleeding (which might be caused by spikes from the pelvic bones), are also determined [28, 32]. The incidence of deep pelvic infection in cases with open pelvic fractures and rectal trauma was 50% in the series of Song et  al [38] and the diagnosis of an open pelvic trauma is associated with an overall mortality of about 45% [39]. Careful distal neurovascular assessment is paramount. Detailed neurologic examination of the lumbosacral plexus is required in every patient with unstable pelvic ring injury in particular with the presence of sacral fractures [10]. The bony pelvic landmarks should be palpated and manual examination of the pelvis stability should be performed but with care not to cause agonizing pain to the patient and for the fear of stimulating intra-pelvic bleeding.

b2

c2

b1

c1

c3

b3

a3

©JSchoenecker2018

Fig. 16.9  Tile’s Classification system (Table 16.2). (a) Type A stable pelvic injuries. (b) Type B partially stable (incomplete disruption of the posterior arch). (c) Type C unstable (complete disruption of the posterior arch)

a2

a1

422 M. Kenawey

16  Pediatric Pelvic Injuries

Imaging Plain pelvic radiographs remain the gold standard in the initial assessment of children with pelvic ring injuries [14, 27, 30]. ATLS guidelines recommend obtaining a plain anteroposterior (AP) pelvic radiograph following primary survey to identify significant pelvic fractures [35]. The full adult radiographic series to assess pelvic ring injuries includes AP, inlet and outlet pelvic views (Fig. 16.10) [7, 11, 12]. AP pelvic radiographs would show anterior pelvic ring injuries with relative reasonable accuracy, how-

423

ever the posterior pelvic ring is mostly obscured by abdominal gases (Fig. 16.11). Another useful projection is the inlet view which is obtained by directing the X-ray beam from the head at an angle of 60° towards caudal direction. It is useful in the assessment of the integrity of the posterior pelvic ring including sacroiliac joint, the posterior part of the ilium and sacrum. Sacroiliac joint SIJ ligamentous disruption with widened SIJ space, dorsal displacement at the level of the SIJ or trans-iliac fracture dislocations of the SIJ, and sacral impaction and displaced sacral fractures could be easily visualized in the inlet view

a

b

c

Fig. 16.10  Full plain radiographic pelvic series of 8-year-old boy who had left side pelvic fracture Tile’s C. (a) AP pelvic view, (b) outlet view and (c) inlet view

424

Fig. 16.11  From the AP view we can see a vertical migration of the left hemipelvis with fractured right pubic ram  +  pubic symphysis disruption by pubic apophysis avulsion from the left pubic bone (arrow)

M. Kenawey

Fig. 16.12 Inlet pelvic view of the same boy in Figs. 16.10 and 16.11 and it shows more clearly the dorsal and vertical displacement of the left hemipelvis at the level of the sacroiliac joint (thick arrows) as well as pubic apophysis avulsion in the anterior ring (thin arrow)

Fig. 16.13 Outlet pelvic view of the same boy shows the vertical displacement of the left hemipelvis as well as a fleck of bone avulsed from the left iliac wing (white arrow) which represent avulsion injury of the iliac crest apophysis. Outlet view is useful in the assessment of the transverse processes of L5 (black arrows) which is avulsed with vertically displaced pelvic injuries due to the attachment of the iliolumbar ligament

(Fig. 16.12). Moreover, it allows visualization of pubic symphysis for symphyseal diastasis and rotational displacement of the pelvic ring. On the other hand, outlet view is obtained by directing the X-ray beam from the foot cephalad at an

angle of 45° (Fig. 16.13). The outlet view is very helpful in the assessment of the vertical stability and displacement of the hemipelvis and proximal migration of the iliac wing. It is also useful in the visualization of the entire sacrum, sacral

16  Pediatric Pelvic Injuries

425

Table 16.3  Essential imaging studies for pelvic ring fractures? Imaging modality Plain radiographs

Anteroposterior view Inlet view

Outlet view

CT scans

– Anterior pelvic ring disruptions – Grossly unstable and displaced pelvic ring injuries – Widened and dorsally displaced sacroiliac joint. – Posterior iliac fractures and crescent iliac fracture – Sacral fracture (sacral impaction fracture). – Anterior rotational symphyseal diastasis. – Vertical instability of the pelvis – Sacral fractures and their relation to the sacral foramina – Posterior iliac fractures – Avulsion of the transverse process of L5 – Assessment of the posterior pelvic ring including sacral injuries and its relation to sacral foramina and the central canal as well as sacroiliac joint integrity in addition to any associated acetabular trauma.

foramina and SIJs on both sides, and for identifying fractures which encroach the sacral foramina and the central canal [10]. Another useful sign of vertical pelvic instability that can be identified in the outlet view is the avulsion injury of the transverse process of L5 (Table 16.3) [40]. Abdominal and pelvic CT revolutionized our trauma protocols and they are now integral part of our routine imaging assessment of blunt abdominal and pelvic trauma. CT scans offer better visibility of complex pelvic and acetabular anatomy and the posterior pelvic ring including the sacroiliac joints and sacrum (Fig. 16.3). The use of intravenous contrast material allows simultaneous assessment of vascular and abdominal organ trauma. CT cystogram is useful in the evaluation of associated urethral and bladder injuries [41]. About a third of posterior ring injuries would be missed on initial plain radiographs in adults [42], while in pediatric patients, the sensitivity of detecting posterior ring injuries is even lower, ranging from 33% in case of sacral fractures to 50% for sacroiliac joint injuries [43]. There are several reasons for this [30]: (1) the pediatric bony pelvis is small and is partly cartilaginous with several growth centers with varying ossification sequences, making it difficult to recognize true fracture lines [17, 44], (2) the presence of thick periosteum limits fracture displacement, accentuating the elastic recoil of displaced pelvic injuries [16, 17]; and (3) the prevalence of apophyseal avulsions around the pelvis associated with

symphyseal and sacroiliac joint injuries are difficult to be identified in plain radiographs [15, 16, 45]. For these reasons, the use of trauma CT scans are advocated to better delineate the injury patterns affecting the posterior pelvic ring. As a result, outlet and inlet pelvic views are no longer ordered in favor of CT. In addition to defining the pathoanatomic lesions and injury patterns affecting the anterior and posterior pelvic ring areas, pediatric peculiar associated injuries like the iliac apophysis and pubic apophysis avulsions should be documented. Iliac apophysis avulsions can be diagnosed by identification of the “Thurston-Holland fragment” of attached bone using axial CT; often represented by a small bony avulsion from the most posterior attachment of the iliac apophysis at the area of the posterior superior iliac spine (PSIS) (Fig.  16.3). Plain radiographs may also demonstrate these small avulsions from the posterior iliac crest (Fig.  16.13). Pubic apophysis avulsion can be identified using either plain radiographs or CT scans; represented by a bony avulsion from the symphyseal side of the pubic bone (Fig. 16.2) [15, 16, 45]. Despite the improvements in visualization of posterior pelvic structures by the use of CT, Silber et al [30] found no advantage in adding CT scans to the routine AP pelvic radiograph for classification (using the Torode and Zieg system), nor did its use change the final treatment decision. It is therefore questionable if the identification of non-displaced fracture lines posteriorly

M. Kenawey

426

a

b

c

Fig. 16.14 (a) Plain pelvic radiograph and (b) axial CT cuts of ten-year-old male who had right fracture pelvis type C1 with right trans-foraminal comminuted sacral fracture with associated ipsilateral foot drop and anterior

pubic symphysis disruption. (c) Fixation was by double ilio-iliac bridge plating posteriorly and symphyseal plating anteriorly

would justify routine CT scans in pediatric pelvic trauma. Finally, MRI use in adult pelvic trauma is very limited and few studies have reported their experience with this imaging modality in the identification of ligamentous injuries of the pelvic floor and the SIJ [46, 47]. Recently, we started to use MRI in highly unstable and displaced pediatric pelvic ring trauma to delineate the posterior soft tissue pathology and study its variation compared to the known adult pathology [15]. To our knowledge, no studies have tried to explore the role of MRI in pediatric pelvic trauma (Fig. 16.4).

operative treatment (88% vs 79%, P = 0.09) and this was statistically significant in cases with fracture displacement >15  mm (84% vs 50%, P = 0.04).

 anagement of Pelvic Ring M Disruptions

Resuscitation and Management of Non-orthopedic Associated Injuries Enormous forces are required to disrupt the stability of the pelvic ring and therefore it should not be considered an isolated injury, but should be managed in the spectrum of polytrauma prinNon-operative Management ciples [11]. Initial management should stick to the ATLS principles to control acute life-­ threatening conditions. Traumatic brain injuries Management of Avulsion Injuries Around the Pelvic Ring as well as concomitant abdominal and pelvic organ injuries causing significant exsanguination Pelvic apophyseal avulsion fractures are treated are important causes of mortality in pediatric pelnon-operatively with protected weight-bearing vic ring injuries [32]. Urologic injuries are the for 4–6 weeks followed by physical therapy and next common to traumatic brain injuries and they gradual return to sports activities by 6–12 weeks might be due direct trauma caused by pubic sym[1, 5, 6]. In the study of Schuett et al [5], 5 out of physis disruption or pubic rami fractures. The 228 (2%) pelvic avulsion injuries had non-union, immediate treatment of children with pelvic ring 2 only were symptomatic non-unions and frac- fractures doesn’t differ significantly from adults ture displacement >20 mm was the main risk fac- particularly when the patient is hemodynamically tor independent of fracture type. In a meta-analysis unstable. In the series of Ismail et  al [32], only by Eberbach et al [6] reporting 596 patients in 14 one out of 41 hypotensive children died as a studies, they found that rates of excellent out- result of fatal pelvic fracture related exsanguinacome was higher following surgical treatment for tion. This emphasized the importance of controlpelvic apophyseal avulsions compared to non-­ ling hemorrhage from thoracic and abdominal

16  Pediatric Pelvic Injuries

related sources before resorting to more extensive steps to control pelvic related bleeding. Application of external fixators and possible embolization of major pelvic arterial bleeding, are reasonable therapeutic options for a child who has ongoing bleeding, but only after all other potential visceral injury has been excluded or controlled. “Initial management should stick to the ATLS principles to control acute lifethreatening conditions”. Conservative treatment is the treatment of choice of most of pediatric pelvic ring injuries. Choices of non-operative treatment might be: (1) weight bearing as tolerated, (2) partial weight bearing, (3) non-weight bearing, (4) pelvic sling for symphysis diastasis, (5) hip spica, (6) skeletal traction by a pin in the distal femoral metaphysis—10–15 pounds (vertical shear injuries), (7) traction through hip joint in lateral compression injuries.

Operative Management With increasing awareness of the long term negative outcome and the impact of disabilities and morbidity caused by persistent pelvic deformity and asymmetry, more indications of operative stabilization of pediatric pelvic ring injuries are evolving [15]. Ogden [17] and Kruppa [26] emphasized the importance of aggressively dealing with displaced posterior pelvic ring injuries to decrease long term back pain and sacroiliac pain. A recent article was published from our institution studying the indications and techniques of operative fixation of unstable pediatric ring injuries in 29 patients, and our indications were displaced SIJ disruptions and trans-iliac SIJ fracture/dislocation (crescent iliac wing fracture) in 15 out of 29 patients (52%), while pubic symphysis diastasis, trans-sacral fractures, displaced trans-triradiate cartilage injuries and trans-iliac fractures were the next [15]. Few published studies discussed the fixation techniques of unstable pediatric pelvic ring injuries [15, 18, 24–26]. Keshishyan et al [25] used a

427

custom made hinged external fixator in 12 children while Oransky et al [18], Smith et al [24], and Kruppa et al [26] reported the use of different forms of internal fixation in combination with external fixators. Kruppa et  al [26] treated 16 patients with unstable displaced pelvic ring injuries (Tile’s B2, B3 and C2) operatively. Their fixation choices included pubic symphysis plating or plate fixation of rami fractures (n  =  7), pubic ramus screws (n  =  3), external fixators (n = 4), IS screws (n = 9), iliac plating (n = 3), and tension band plating (n  =  1). Oransky et  al [18] used combinations of plates and screws, K-wires and external fixation in 8 children. To simplify the operative management of unstable pelvic ring; fixation choices and techniques should be discussed separately for the anterior and posterior pelvic structures. However, because of the rarity of such injuries and the low number of published studies reporting on the treatment modalities specific for pediatric pelvic injuries, there is insufficient evidence to recommend for or against specific fixation choice or technique (Table 16.4).

 nterior Pelvic Ring Reduction A and Fixation Techniques Anterior pelvic ring includes the areas of the pubic symphysis, pubic rami and the acetabular segment and the triradiate cartilage according to the maturity status of the pelvic bones. Failure of the anterior pelvic ring may be broadly through ligamentous disruption of the pubic symphysis, unilateral or bilateral pubic rami fractures, trans-­ acetabular fracture or trans-triradiate cartilage injuries or a combination of any of those [16, 20]. We have discussed earlier the difference of the injury patterns that might be encountered in young children with immature pelvis compared to older children and adolescents with mature pelvic bones [15, 16]. Fixation techniques of the anterior pelvic ring depends on the encountered pathology (Table 16.3). Due to the poor healing of the pubic symphysis ligamentous disruptions, anterior symphyseal plating is indicated in adults [48].

M. Kenawey

428

Table 16.4  The fixation options according the encountered anterior and posterior pelvic ring disruption mode Posterior Pathology Fixation options Trans-­symphyseal instability Symphyseal plate (choice in adults) Anterior supra-acetabular external fixator (choice in children due to the high incidence of pubic apophysis avulsion) Trans-pubic instability Anterior supra-acetabular external fixator Pubic ramus screw Trans-acetabular instability Internal fixation Trans-iliac instability Iliac wing plating Iliac wing screw fixation Trans-sacral fractures Iliosacral screw fixation (choice) Ilio-iliac posterior bridge plating (comminuted sacral fractures) Direct posterior sacral plating Triangular internal fixation (spino-pelvic dissociation–comminuted sacral fractures) Trans-iliac fracture/ Anterior SIJ plating dislocation SIJ Posterior iliac plating (choice in displaced injuries) Antegrade or retrograde lateral compression LC screws (choice) Iliosacral screws (in selected cases) Ligamentous SIJ diastasis Iliosacral screw (choice) Anterior sacroiliac plating Trans-sacral SIJ Iliosacral screw (choice) Anterior sacroiliac plating

a

b

Fig. 16.15  Pre and Post-operative X-ray (left and right respectively) for the same patient in Fig. 16.4 and fixation with pelvic brim plate and IS screw

Anterior symphyseal plating might be used as well in pediatrics (Fig.  16.14) [15, 17]. Other internal fixation options for pubic rami fractures are pubic ramus plating, anterior pelvic brim plating and pubic ramus screws. Pubic ramus plating is a small local plate that spans the superior pubic ramus [26]. Anterior pelvic brim plate is a longer plate that is extended along the pelvic brim (ilio-pectineal line) to the supra-acetabular

area and this can be used to fix: (1) displaced basal pubic rami fractures, (3) trans-acetabular or trans-triradiate cartilage injuries and (3) any combination of the previous injuries (Fig. 16.15) [49]. The last internal fixation tool in the anterior ring area is the pubic rami screws whether antegrade or retrograde pubic ramus screw fixation and similarly the percutaneous anterior and posterior acetabular column screw fixation [50].

16  Pediatric Pelvic Injuries

429

a

b

c

d

Fig. 16.16 Intra-operative fluoroscopic views of the technique of Schanz screws insertion for supra-acetabular external fixator. (a) The teepee view showing the tear drop of the ilium (black arrow). (b) an oblique axial CT cut showing the safe bony corridor of the supra-acetabular Schanz screw insertion starting from the AIIS (arrow) to

the posterior ilium. (c) Anteroposterior pelvic view shows the optimal placement of the Schanz screw reaching just proximal to the greater sciatic notch to engage the posterior ilium and (d) The teepee views shows its position exactly in the iliac tear drop

External fixation is another valuable tool for the fixation of the anterior pelvic ring in pediatrics [51]. Half pins of the external fixators can be inserted through the anterior third of the iliac crest (posterior to the anterior superior iliac spine) or through the supra-acetabular area just above the anterior inferior superior iliac spine [51]. Supra-acetabular half pins are far more stable than iliac crest pins because the path of the half pins in the iliac crest route is very short, ending at the extremely thin iliac fossa and in

our personal experience have very poor purchase and grip particularly in young and small children [52]. In contrast, supra-acetabular pins are inserted in the dense supra-acetabular bony stock where there is a safe corridor of bone parallel to the iliopectineal line and extends from the anterior inferior iliac spine AIIS to the posterior ilium and the posterior inferior iliac spine PIIS and is about 3 × 15 cm in adults (Fig. 16.16) [51]. Anterior supra-acetabular external fixator is our personal fixation choice

430

for anterior ring disruptions particularly in young children with immature pelvic bones. It is technically easy, a minimally invasive option and can be used even in pubic symphysis diastasis, because in young children, it is mostly a pubic apophyseal avulsions with very good healing potential rather than pure ligamentous failure. Moreover, in our personal experience regular cortical screws used with different anterior plating techniques in children with immature bony pelvis, have weak bony grip in the area of the soft pubic body and rami [15]. In our series of operative fixation of unstable pediatric pelvic injuries, we used supra-­acetabular external fixators for anterior ring fixation in 21 out of 29 patients (eight had pubic symphysis disruptions) [15].

 osterior Pelvic Ring Reduction P and Fixation Techniques Posterior pelvic ring is the key determinant of the whole pelvic ring stability and therefore, is the principle indicator for the need for operative stabilization [9, 53]. Components of the posterior pelvic ring are the iliac wings, sacroiliac joints and the sacrum. Posterior pelvic ring can fail through the ilium, the sacrum or the sacroiliac joint. Sacroiliac joint disruptions are the most common mode of posterior pelvic ring failure in young children [16]. There are three subtypes of SIJ injuries: the purely ligamentous SIJ diastasis, trans-iliac fracture/dislocation of the SIJ (the crescent fracture in the Young-Burgess LC II and III) and the trans-sacra fracture/dislocation SIJ. Fixation choices and techniques are dependent on the injury pattern and pathology affecting the posterior ring (Table 16.3). The most commonly used fixation devices in the posterior ring are the iliosacral IS screws and the lateral compression LC screws. Iliosacral screws can be used in different pattern of SIJ instability and in sacral fractures while LC screws are only used in lateral compression LC injuries—grades II and III according to Young—Burgess classification system [15]. We used IS screws as early as the 6th year of age. This technique is safe and very effective in pediatrics as well as in adults. We have used IS

M. Kenawey

screws for posterior ring fixation in 13 out of 29 pediatric patients who had operative stabilization of unstable pelvic injuries [15]. Scolaro et al [54] used IS screws in 67 pediatric patients between the age of 7 and 17-years. No nerve palsies were reported in both series. The main difficulties with the use of this fixation technique are related to: (1) obtaining the safe anatomic corridor, (2) reduction of displaced SIJ and (3) the fixation strength through the soft iliac wing. IS screws are inserted under strict fluoroscopic control as the case in adults. We should be careful because of the size of the bony channels is significantly smaller than adults and because of the ossification centers of the sacrum. We should be aware of the contraindication of the technique, in particular the presence of sacral congenital malformations (Fig.  16.17). IS screws might be inserted in the prone or the supine position but in cases that would need closed reduction of SIJ, we do prefer the prone position as the hemipelvis is usually dorsally and cranially displaced. One last remark is the unique intra-operative difficulty that we reported in our series with the use of IS screws which is piercing the soft iliac wing in 3 patients despite the use of washers underneath the screw heads and we had to take them out in 2 patients and the screws were reinserted through the holes of plates (Fig. 16.18) [15].

Fig. 16.17  Sixteen-year old boy who sustained this pelvic ring injury and there is partial sacralization of the left side of L5 (white arrow) which is a relative contraindication for IS screw fixation of the posterior pelvic ring

e

b

Fig. 16.18  Six-year-old boy who had bilateral fracture pelvis type C3 due to runover accident. (a) Pre-operative plain X-ray shows right SIJ ligamentous disruption and left trans-iliac SIJ fracture/dislocation and anteriorly there was pubic rami fracture + pubic symphysis disruption. (b–e) Show our steps to insert two IS screws on the right side. (b) Two guide wires were inserted in the first sacral segment. (c) Insertion of the distal

d

a

screw. (d and e) Insertion and tightening of the two screws and their heads pierced the iliac wing to lie completely within the SIJ space despite the use of washers. (f) The screws were taken out and then reinserted through the 3.5 mm holes of the reconstruction plate

f

c

16  Pediatric Pelvic Injuries 431

M. Kenawey

432

The second most commonly used posterior fixation tool are the LC screws. Starr et  al [55] devised the use of LC screws as a minimally invasive fixation option in cases of lateral compression pelvic injuries LC-2 and LC-3 according to Young—Burgess classification system. Lateral compression injuries grades 2 and 3 typically cause a trans-iliac fracture dislocation SIJ on the affected side with a posterior crescent iliac fracture (Fig. 16.19). In fact, this dorsal crescent iliac fragment is the stable part of the iliac wing which is still in attachment to the sacrum and axial skeleton while the whole remaining hemipelvis is dislocated from the sacrum through this trans-iliac intra-articular fracture, separating the dorsal crescent fragment from the remaining ventral part of the ilium (Fig. 16.19). The idea of the LC screws is to use the supra-acetabular bony corridor to connect the iliac wing to the posterior crescent frag-

a

b

c

d

Fig. 16.19 (a) Plain pelvic radiograph and (b) axial CT cuts of 14-year-old female who had fracture pelvis type C1 right side with posterior trans-iliac SIJ fracture/ dislocation and anterior pubic rami fractures. (c) ORIF using retrograde LC screws and supra-acetabular exter-

ment and it can be inserted in an antegrade direction in the supine position or in a retrograde direction in the prone position (Figs.  16.19 and 16.20). The choice of the position of the patient and route of insertion is dependent on the ability to reduce the fracture displacement in a closed fashion. The posterior approach and retrograde insertion is used when there is a need for open reduction of the trans-­iliac fracture dislocation by approaching the SIJ directly from posterior to help reducing the iliac wing to the posterior crescent iliac fragment. The main difficulty with the use of LC screws is when an open reduction is required in young children with immature bony pelvis. The bony crescent fragment that appears to be sizable in preoperative CT scans would be actually small and covered with the thick growing posterior iliac apophysis and the purchase of screw fixation in this small fragment would not be optimal [15].

nal fixator. (d) Three-dimensional CT reconstruction which shows that the dorsal crescent fragment is the stable fragment (arrows) while the remaining whole hemipelvis is dislocated from the SIJ

16  Pediatric Pelvic Injuries

a

433

b

c

Fig. 16.20 (a) Plain pelvic radiograph and (b) axial CT cuts of 18-year-old male with left side fracture pelvis type C1, posterior trans-iliac SIJ fracture/dislocation, anterior pubic rami fractures, and left femoral neck fracture. (c)

Closed reduction was achieved and fixation using percutaneous antegrade LC screws and anterior supra-acetabular external fixator

Another valuable internal fixation method of the posterior ring is the use of ilioiliac bridge plating or tension band plating (Fig. 16.14). The best indications for this type of fixation in pediatric patients are: (1) in cases of comminuted sacral fracture in which the risk of nerve injury with iliosacral screw fixation is high and (2) in cases of sacral fractures where iliosacral screw fixation is contraindicated or no adequate fluoroscopic imaging could be obtained. Ilioiliac bridge plating is a simple technique and is minimally invasive alternative for complicated posterior pelvic ring fixation [56].

and the pin is mounted directly to the power drill. The half pin should be directed approximately 10–20° cephalad and 20° to the medial or inner aspect of the pelvis towards the SIJ. With drilling, the pin would find its correct path within the bony corridor between the tables of the ilium. The direction and position of the half pin is checked on the fluoroscopic AP view to ensure that our pin passes just proximal to and beyond the greater sciatic notch to get the best bone stock and to engage the posterior ilium and on the fluoroscopic obturator view (or Teepee view) to ensure that the half pin lies exactly within the tear drop of the ilium (Fig. 16.16). The same steps are repeated on the other side and external fixator clamps are then mounted. The half pin on the injuries hemipelvis can be used as a joy stick for the reduction of any pubic symphysis diastasis or displaced pubic rami fractures and the half pins are then connected with the rod.

Surgical Techniques of Selected Procedures  upra-Acetabular External Fixator S The patient is placed in the supine position on a radiolucent orthopedic table. Eight to 10 mm skin incision is made about a finger breadth distal and lateral to the anterior superior iliac spine. A Kelly clamp or a small hemostat is then used to bluntly spread the soft tissue down to the AIIS.  In our experience, we would insert 5  mm half pins in children younger than 8-years and 6 mm half pins in children who are 8-years or older. We would insert the appropriate drill sleeve and the trocar is then fixed against the anterior bony margin just proximal to the AIIS to avoid penetrating the anterior capsule of the hip joint and intra-­articular placement of the half pin. We like to use self-­ drilling half pins to decrease the insertion time,

Iliosacral Screws The technique can be performed in both the supine and prone positions. The patient is positioned on a radiolucent orthopedic table. We must confirm the adequacy of fluoroscopic outlet and inlet views as well as the lateral lumbosacral view which will be used to guide the insertion of the IS screws before preparing and draping the patient. The difficulty would be mainly in obtaining adequate outlet view in the supine position and to get an adequate inlet view in the prone position usually because of the column of the operating table. With adequate outlet view, the superior pubic rami should be seen

M. Kenawey

434

against the foramina of S2 while in the inlet view, the bodies of S1 and S2 should be overlapped (Fig.  16.21). We start our technique under the guidance of the lateral view of the lumbosacral junction and adequate lateral lumbosacral view is confirmed with the lumbosacral junction being at

the center of the fluoroscopic image and both greater sciatic notch exactly overlapped (Fig. 16.22). A guide wire is used to mark the center of S1 body superficially on the skin and then 10 mm skin incision is made, and a straight Kelly clamp is used to spread out the soft tissues down to

Fig. 16.21  Fluoroscopic outlet (left) and inlet (right) pelvic views for IS screw fixation

Fig. 16.22  Fluoroscopic lateral view of the lumbosacral area (leN) with the greater scia,c notches overlapped (arrows) and marking the middle of S1 body by the guide wire (right)

16  Pediatric Pelvic Injuries

the lateral surface of the iliac wing. The guide wire is then inserted by hand through the skin wound until it stops at the lateral iliac wing and its tip is ­positioned against the center of S1 body under fluoroscopic control (Fig. 16.22). Then the guide wire is hammered into the iliac wing. The fluoroscopy is then repositioned to get the outlet and inlet pelvic views and the guide wire is advanced with a power drill to lie within the safe corridor. In the outlet view, the guide wire should be placed between S1 foramina and L5/S1 disc and it should be central with the body of S1 in the inlet view to avoid posterior penetration of the neutral canal or injury of the anterior neurovascular structures ventral to the sacrum (Fig.  16.23). The appropriate sized cannulated drill is used to drill the path of the screw which should be inserted just beyond the center of S1 to get the best bone grip. With tightening of the screw, care should be taken to avoid the screw head from penetrating the soft iliac wing and with the first sign of such penetration, the screw should be taken out and reinserted through the holes of a plate (Fig. 16.18).

 ateral Compression Screws L Lateral compression screws are inserted either in an antegrade direction or through retrograde

Fig. 16.23  Fluoroscopic outlet (left) and inlet (right) pelvic views showing insertion of the threaded guide wire of IS screw fixation in between the S1 foramen and L5/S1

435

route in either the supine and prone positions respectively. Antegrade insertion is similar to the insertion of the half pins in the supra-acetabular external fixator technique (Fig. 16.20). Our technique uses the same skin incision as the supra-­acetabular external fixator and then we would open the cortex of the iliac wing with 3.2 mm drill bit. A 2.5 mm threaded guide wire is then inserted by hand through the cortical hole and hammered into the supra-acetabular bone stock in between the tables of the iliac wing in the same direction as the supra-acetabular half pin. With gentle hammering and without the use of power drill, the threaded guide wire will advance strictly in the medullary space of the ilium till the posterior ilium with lower risk of cortical penetration (Fig. 16.24). The obturator/outlet view is used to control the insertion of the guide wire through the correct path in the tear drop of the ilium and the iliac view is used to control the depth of insertion (Fig. 16.25). A cannulated drill bit is then used to drill the path of the screw and the screw is inserted. Screws as long as 150 mm are required to engage the posterior ilium and to pass beyond the trans-iliac intra-articular fracture to the area of posterior iliac crescent fragment (Fig. 16.26).

disc (the oval line and the arrow respectively) in the outlet view and in the middle of the body of S1 in the inlet view

436

M. Kenawey

Fig. 16.24  Fluoroscopic views for the steps of antegrade LC screw insertion. First (left) a 3.2 mm drill is used to open the anterior iliac cortex just proximal to AIIS. Then 2.5 mm threaded guide wire is hammered manually into

the supra-acetabular bony corridor aiming 20° caudally and 20° medially and it should pass proximal to the greater sciatic notch to engage the posterior ilium

Fig. 16.25  Fluoroscopic AP (left) and obturator or Teepe view (right) to check the position of the guide wire. On the AP view, it passes proximal to the greater sciatic notch and crosses the SIJ (arrow). In the Teepe view, the wire is

exactly in the tear drop of the ilium, and it appears that the guide wire is bent to adapt the curvature of the base of the ilium (arrow)

For retrograde insertion, a short posterior approach to the SIJ is made. Gluteus maximus fibers are elevated subperiosteally from the external surface of the iliac wing until we reach the trans-iliac fracture line (Fig. 16.27). The fracture site is then manipulated to reduce the displaced

hemipelvis with the help of two small and curve Hohmann retractors placed with in the fracture gap and with the help of longitudinal downward traction on the affected limb to reduce the vertical displacement. Following reduction, two threaded 2.5 mm guide wires are inserted through the PIIF

16  Pediatric Pelvic Injuries

437

Fig. 16.26  The fluoroscopic AP (left) and Teepe (right) views shows the LC screw in place

Fig. 16.27  Axial CT cut (left) and three-dimensional CT reconstruction (right) views of a 14-year-old male patient with fracture pelvis right side Tile’s C1 with posterior transiliac fracture dislocation SIJ.  The axial CT cuts shows the path of the retrograde LC screw. The three-­ dimensional reconstruction shows the trans-iliac intra-­

articular fracture line which is exposed through the posterior approach and two Hohmann retractors are inserted between the fracture surfaces (arrows) to reduce the displaced iliac wing before the insertion of the retrograde LC screws

and just proximal to it through the posterior ilium and the sciatic buttress and 7.3  mm cannulated screws can be inserted for fixation (Fig. 16.19).

ation. The patient is positioned in the prone position on a radiolucent orthopedic table. Bilateral short straight skin incisions are made centered over the posterior superior iliac spine and extending down to the level of the PIIS. The fascia over the gluteus maximus is incised in line with the incision and then the gluteus maximus is reflected subperiosteally downwards and laterally to

I lio-iliac Posterior Bridge (Tension Band) Plating Our preferred technique is to use double plating without any additional form of posterior ring fix-

M. Kenawey

438

expose the posterior interspinous zone of the ilium and the external surface of the posterior ilium just to allow placement of the last hole of the plate on the external iliac surface. There are two variation in plate placement: (1) between the interspinous zones of the ilium, either superficial or deep to the paraspinal muscles and (2) Submuscular at the level of S2 bridging to the innominate bones just cranial to the greater sciatic notch [57]. We do prefer to place the plates in the interspinous zone and deep to the paraspinal muscles. The submuscular channel used to pass the plates is created with the help of small Cobb elevator and then templates are used to determine the correct length and shape of the used plates. Our aim is to get a hole just on the posterior border of the ilium within the interspinous zone and then an extra-hole laterally on each side which is bent to lie on the external surface of the iliac wing (Fig. 16.28). We use two 3.5 small DCP or reconstruction plates of the same length and the plates are passed deep to the paraspinal muscles. First, the screws of the holes in the interspinous zones are inserted into the sciatic buttress and sequentially tightened to compress the plates against the ilia. Then the most lateral holes are used to insert short screws into the iliac wings which may or may not cross the SIJ (Figs. 16.14 and 16.28).

Fig. 16.28  Tension band ilio-iliac posterior bridge plating can be used with comminuted trans-foraminal sacral fracture. the plate is bent on both sides so a screw is inserted along the same path of the retrograde LC screws bilaterally (thick white arrows) and an additional hole on the external surface on the iliac wing is used to insert another short iliac wing screw on both sides (thin white arrows)

Classic Papers Torode I, Zieg D (1985) Pelvic fractures in children. Journal of pediatric orthopedics 5:76–84 Torode and Zieg [19] reviewed retrospectively the records of 141 patients who were admitted to the Hospital for Sick Children between 1971 and 1981 with the diagnosis of pelvic fracture. Avulsion injuries and acetabular fractures were excluded. The age range was 2–17 years-old and there were 85 males. The commonest injury mechanism was pedestrian hit by a car in 78% of the cases. Eleven deaths (8%) were recorded in this series with the majority due to head injury (n = 6), abdominal visceral injury and blood loss (n  =  3), combined head and chest crush injury (n = 1) and one high spinal cord injury with C1, C2 subluxation. In this work, Torode and Zieg devised their classification of pediatric pelvic ring injuries which would give some guide to the injury severity, prevalence of the associated injuries and the expected outcome. All patients in this series had types II, III and IV pelvic ring injuries. The authors then analysed the relation of the fracture type to the resuscitation requirements as well as to the other associated general and musculoskeletal injuries, overall morbidity and mortality. It was clear that there is strong difference in the resuscitation requirements, blood transfusion and death rates for types III and IV compared to type II injuries. While only 3 patients with type II pelvic ring injuries required blood transfusion and there were no recorded deaths, 15 and 16 patients had blood transfusion for types III and IV respectively and there were 2 and 5 deaths respectively. Associated neurologic, genitourinary, abdominal and musculoskeletal injuries were strongly correlated as well to the pelvic fracture type and severity. Neurologic injuries were documented in 81 patients and ranged from simple concussion and skull fractures to central neurologic damage and hemiplegia as well as peripheral nerve injuries. Most cases with neurologic injuries were recorded with types III and IV fracture groups the association between pelvic ring injury and concomitant head injury is consistent with the triad of injuries caused by pedestrian

16  Pediatric Pelvic Injuries

hit by a car and known as Waddell’s triad. Peripheral nerve injuries or the lumbosacral plexus were seen with type IV pelvic fractures due to the associated posterior ring disruption. The same findings were seen with genitourinary injuries where the rates of macroscopic hematuria and other genitourinary injuries including vaginal and perineal lacerations, urethral injuries and bladder ruptures were higher in types III and IV fractures. Laparotomies as an indicator for the severity of associated abdominal injuries were recorded in 3 patients with type II fractures, 8 type III and 15 type IV. Last, associated musculoskeletal injuries were correlated as well to the pelvic fracture types with progression of 39% of patients with type II injuries to 56% with type IV and the commonest were long bones fractures of the tibia and fibula. Regarding the results of treatment of individual fracture types; Type II (18 children) fractures were treated conservatively with uniformly good results. Two patients had delayed ossification of the iliac apophysis but follow up at 2 years showed no significant difference between the two sides. In type III injuries (n = 70), two cases had delayed union of pubic rami fractures. Patients with wide symphyseal diastasis in this group had good prognosis due to the fact that in children this takes place through pubic apophysis avulsion (separation of cartilage/bone interface) with avulsion of a periosteal sleeve. Healing would involve filling the defect with new bone from the growth plate and from the periosteal sleeve. Patients with type IV fractures (n = 40), 3 children had operative stabilization. Eight patients had non-united pubic rami, 3 premature triradiate closure (two required Salter’s innominate osteotomy), one patient had malunited acetabular fracture and ankylosed hip, and one sacroiliac joint fusion without consequences. The authors found marked residual pelvic deformity in 5 patients but couldn’t find adverse effects on the long-term result and 4 patients with significant pelvic deformity had excellent clinical result, however this is not in agreement with the more recent result which clearly demonstrated the unfavorable long-term consequences of persistent pelvic deformities.

439

Key Evidence Keshishyan RA, Rozinov VM, Malakhov OA, Kuznetsov LE, Strunin EG, Chogovadze GA, Tsukanov VE (1995) Pelvic polyfractures in children. Radiographic diagnosis and treatment. Clinical orthopaedics and related research 320:28–33. Keshishyan et al [25] compared retrospectively two cohorts of pediatric patients with pelvic ring injuries; 31 children who were treated conservatively with immobilization and 12 patients who had surgical stabilization using custom designed hinged external fixator and the follow up ranged from 6 months to 11 years. They studied the effect of residual pelvic deformity on the functional outcome and proposed a method to quantitatively assess residual pelvic deformity by measuring the diagonal distances between the lower border of the sacroiliac joints to the middle point of the medial aspect of the contralateral acetabulum and they found a maximal difference of 4 mm in normal children (Fig. 16.7). Poor outcome was reported in 33% of the patients in the conservative group who complained of continuous pain, significant gait disturbance and >10% limitation of the hip range of motion. All patients who had poor outcome suffered from anterior and posterior pelvic ring disruptions and there was a correlation between residual pelvic deformity as measured by Keshishyan index and the late functional outcome. Keshishyan index was 6–11  mm in the good outcome group, 8–15  mm in the satisfactory group and 13–25 mm in those patients with poor outcome. In the operative group, the range of Keshishyan diagonal difference was 8–17 mm and restoration of the pelvic shape was achieved in all patients. Ten patients had good functional results and 2 had satisfactory results (Level III evidence). Schwarz N, Posch E, Mayr J, Fischmeister FM, Schwarz AF, Ohner T (1998) Long-term results of unstable pelvic ring fractures in children. Injury 29:431–433.Seventeen patients with unstable pelvic fractures Tile’s B or C, treated conservatively, were retrospectively

440

reviewed by Schwarz et al [23]. In one multiply injured patient, an external fixator was applied 10 weeks after the injury to try to correct persistent pelvic displacement. The age at the time of injury was less than 12 years and there were 13 boys and 4 girls. Follow up clinical examination and radiographic assessment were scheduled at least 2 years after the index trauma (10–25 years in 9 patients). Ten patients had limb length discrepancy LLD (9 up to 3 cm differences and one patient had 5 cm LLD) and ten had lumbar scoliosis. Low back pain was correlated to lumbar scoliosis in 9 patients and was more prevalent with leg length difference >1  cm (3 out of 5 patients with LLD >1 cm compared to 2 out of 10 patients with LLD ≤1  cm). They found as well that the pelvic deformity didn’t remodel and correct in any patient (Level IV evidence). Smith W, Shurnas P, Morgan S, Agudelo J, Luszko G, Knox EC, Georgopoulos G (2005) Clinical outcomes of unstable pelvic fractures in skeletally immature patients. The Journal of bone and joint surgery American volume 87:2423–2431. DOI 10.2106/ JBJS.C.01244v. Smith et  al [24] reviewed 20 patients (18 were treated operatively) who had unstable pelvic fractures Tile’s B and C and open triradiate cartilage at the time of injury. Their mean age was 9.5  years (range 2.7– 12.8 years). They were followed up at an average 6.5  years (range 1.6–14.8  years) and were evaluated clinically and radiographically. Comparing the follow up radiographs to those at the time of trauma and healing, pelvic asymmetry did not remodel in any of the twenty patients regardless of their age at the time of the injury. All patients with ≥1.1 cm of pelvic asymmetry (Keshishyan index) had three or more of the following: nonstructural scoliosis, lumbar pain, a Trendelenburg sign, or sacroiliac joint tenderness and pain. Dysfunction scores and the Short Musculoskeletal Function Assessment Questionnaire (SMFA) was better in children with pelvic asymmetry ≤1  cm. Furthermore, there was a positive correlation (r  =  0.6 and p = 0.04) between the displacement and asymmetry at the time of the injury and the SMFA score (Level II evidence).

M. Kenawey

Kruppa CG, Khoriaty JD, Sietsema DL, Dudda M, Schildhauer TA, Jones CB (2016) Pediatric pelvic ring injuries: How benign are they? Injury 47:2228–2234. DOI 10.1016/j. injury.2016.07.002. Kruppa et al [26] compared retrospectively two groups of children who had pelvic ring injuries; group 1 (16 patients had unstable pelvic fractures and were treated operatively) and group 2 (17 patients had radiologically stable pelvic ring injuries and were managed conservatively). Average age was 12.6 years (4–16 years) and average follow up 28.6 months (6–101 months). Group 1 had more pelvic asymmetry (12.3 mm and 6.6 mm) compared to group 2 at the latest follow up X-rays. Children with 5–10  mm posterior sacral displacement had significantly more low back pain and sacroiliac pain than children with   20 were separately considered, the mortality rates were 19.2% and 40.4% for pediatrics and adults respectively. Twenty-four children (3.2%) had open pelvic fractures and only two died as opposed to 8 (29%) deaths in 28 adult patients (5.2%) with open pelvic fractures. There was significant difference in deaths due to fracture related exsanguination with 18 deaths in adult patients and 2  in the pediatric group. The authors concluded that children don’t die of pelvic fracture related hemorrhage as the case with adult patients. Massive blood loss in children is commonly from solid visceral injury as opposed to pelvic vascular disruption (Level III evidence).

441

•

•

•

•

•

Take Home Message

• It is important to understand the peculiarities of the growing pediatric pelvis and the differences in the encountered pathology and patterns of pelvic ring instability in pediatric pelvic fractures compared to adult patients. • Young children with an immature bony pelvis are more susceptible to diastasis of the pubic symphysis and sacroiliac joint, rather than the bony failures sustained by mature adolescents. • Disruptions of pubic symphysis and sacroiliac joints in pediatric patients with open triradiate cartilage takes place

through the growth plates; the iliac and pubic apophyses respectively. There is increasing evidence that post-­ traumatic pelvic deformity and asymmetry would not remodel regardless the age of the child at the time of trauma and that persistent pelvic asymmetry might cause long-term disabilities in the form of chronic low back pain and sacroiliac pain, limb length discrepancy, scoliosis and gait disturbances. Displaced posterior ring and sacroiliac joint disruptions should be reduced and managed aggressively to decrease the likelihood of significant future problems. There is increasing interest of operative fixation of unstable displaced pediatric pelvic ring injuries and nondisplaced fractures are ideally treated conservatively. Anterior supra-acetabular external fixators is an ideal treatment choice for pediatric anterior pelvic ring disruptions even in cases with symphyseal diastasis due to the prevalence of pubic apophysis avulsions which have very good healing. Ilio-sacral screws should be used with caution in the fixation of the pediatric posterior pelvic ring due to the soft iliac wing in children and the risk of piercing the iliac wing with the head of the screw. Be prepared to insert ilio-sacral screws through the holes of a plate.

References 1. Rossi F, Dragoni S.  Acute avulsion fractures of the pelvis in adolescent competitive athletes: prevalence, location and sports distribution of 203 cases collected. Skeletal Radiol. 2001;30:127–31. 2. Lambert MJ, Fligner DJ.  Avulsion of the iliac crest apophysis: a rare fracture in adolescent athletes. Ann Emerg Med. 1993;22:1218–20. 3. McKinney BI, Nelson C, Carrion W.  Apophyseal avulsion fractures of the hip and pelvis. Orthopedics. 2009;32:42.

442 4. Steerman JG, Reeder MT, Udermann BE, Pettitt RW, Murray SR.  Avulsion fracture of the iliac crest apophysis in a collegiate wrestler. Clin J Sport Med. 2008;18:102–3. https://doi.org/10.1097/ JSM.0b013e31815ad14f. 5. Schuett DJ, Bomar JD, Pennock AT. Pelvic apophyseal avulsion fractures: a retrospective review of 228 cases. J Pediatr Orthop. 2015;35:617–23. https://doi. org/10.1097/BPO.0000000000000328. 6. Eberbach H, Hohloch L, Feucht MJ, Konstantinidis L, Sudkamp NP, Zwingmann J. Operative versus conservative treatment of apophyseal avulsion fractures of the pelvis in the adolescents: a systematical review with meta-analysis of clinical outcome and return to sports. BMC Musculoskelet Disord. 2017;18:162. https://doi.org/10.1186/s12891-017-1527-z. 7. Pennal GF, Tile M, Waddell JP, Garside H.  Pelvic disruption: assessment and classification. Clin Orthop Relat Res. 1980;151:12–21. 8. Young JW, Burgess AR, Brumback RJ, Poka A. Pelvic fractures: value of plain radiography in early assessment and management. Radiology. 1986;160:445–51. https://doi.org/10.1148/radiology.160.2.3726125. 9. Tile M. Pelvic ring fractures: should they be fixed? J Bone Joint Surg. 1988;70:1–12. 10. Denis F, Davis S, Comfort T.  Sacral fractures: an important problem. Retrospective analysis of 236 cases. Clin Orthop Relat Res. 1988;227:67–81. 11. Tile M. Acute pelvic fractures: I. Causation and classification. J Am Acad Orthop Surg. 1996;4:143–51. 12. Tile M. Acute pelvic fractures: II. Principles of management. J Am Acad Orthop Surg. 1996;4:152–61. 13. Schlickewei W, Keck T.  Pelvic and acetabular fractures in childhood. Injury. 2005;36(Suppl 1):A57–63. https://doi.org/10.1016/j.injury.2004.12.014. 14. Silber JS, Flynn JM.  Changing patterns of pediatric pelvic fractures with skeletal maturation: implications for classification and management. J Pediatr Orthop. 2002;22:22–6. 15. Kenawey M. Surgical considerations with the operative fixation of unstable paediatric pelvic ring injuries. Int Orthop. 2017; https://doi.org/10.1007/ s00264-017-3475-5. 16. Kenawey M, Krettek C, Addosooki A, Salama W, Liodakis E.  Unstable paediatric pelvic injuries: the patho-anatomical patterns of pelvic ring failure and the role of avulsion of the iliac apophysis. Bone Joint J. 2015;97-B:696–704. https://doi. org/10.1302/0301-620X.97B5.35162. 17. Ogden JA.  Pelvis. In: Skeletal injury in the child. New York Berlin Heidelberg: Springer-Verlag; 1999. p. 790–830. 18. Oransky M, Arduini M, Tortora M, Zoppi AR. Surgical treatment of unstable pelvic fracture in children: long term results. Injury. 2010;41:1140–4. https://doi. org/10.1016/j.injury.2010.08.002. 19. Torode I, Zieg D. Pelvic fractures in children. J Pediatr Orthop. 1985;5:76–84. 20. Gansslen A, Pohlemann T, Paul C, Lobenhoffer P, Tscherne H.  Epidemiology of pelvic ring injuries. Injury. 1996;27(Suppl 1):S-A13–20.

M. Kenawey 21. Habacker TA, Heinrich SD, Dehne R. Fracture of the superior pelvic quadrant in a child. J Pediatr Orthop. 1995;15:69–72. 22. McDonald GA.  Pelvic disruptions in children. Clin Orthop Relat Res. 1980;151:130–4. 23. Schwarz N, Posch E, Mayr J, Fischmeister FM, Schwarz AF, Ohner T. Long-term results of unstable pelvic ring fractures in children. Injury. 1998;29:431–3. 24. Smith W, Shurnas P, Morgan S, Agudelo J, Luszko G, Knox EC, Georgopoulos G.  Clinical outcomes of unstable pelvic fractures in skeletally immature patients. J Bone Joint Surg Am. 2005;87:2423–31. https://doi.org/10.2106/JBJS.C.01244v. 25. Keshishyan RA, Rozinov VM, Malakhov OA, Kuznetsov LE, Strunin EG, Chogovadze GA, Tsukanov VE. Pelvic polyfractures in children. Radiographic diagnosis and treatment. Clin Orthop Relat Res. 1995;320:28–33. 26. Kruppa CG, Khoriaty JD, Sietsema DL, Dudda M, Schildhauer TA, Jones CB.  Pediatric pelvic ring injuries: How benign are they? Injury. 2016;47:2228–34. https://doi.org/10.1016/j.injury. 2016.07.002. 27. Shore BJ, Palmer CS, Bevin C, Johnson MB, Torode IP. Pediatric pelvic fracture: a modification of a preexisting classification. J Pediatr Orthop. 2012;32:162–8. https://doi.org/10.1097/BPO.0b013e3182408be6. 28. Reichard SA, Helikson MA, Shorter N, White RI Jr, Shemeta DW, Haller JA Jr. Pelvic fractures in children–review of 120 patients with a new look at general management. J Pediatr Surg. 1980;15:727–34. 29. Garvin KL, McCarthy RE, Barnes CL, Dodge BM. Pediatric pelvic ring fractures. J Pediatr Orthop. 1990;10:577–82. 30. Silber JS, Flynn JM, Katz MA, Ganley TJ, Koffler KM, Drummond DS. Role of computed tomography in the classification and management of pediatric pelvic fractures. J Pediatr Orthop. 2001;21:148–51. 31. Bond SJ, Gotschall CS, Eichelberger MR. Predictors of abdominal injury in children with pelvic fracture. J Trauma. 1991;31:1169–73. 32. Ismail N, Bellemare JF, Mollitt DL, DiScala C, Koeppel B, Tepas JJ 3rd. Death from pelvic fracture: children are different. J Pediatr Surg. 1996;31:82–5. 33. Furey AJ, Toole RV, Nascone JW, Sciadini MF, Copeland CE, Turen C. Classification of pelvic fractures: analysis of inter-and intraobserver variability using the Young-Burgess and Tile classification systems. Orthopedics. 2009;32:401. 34. Gabbe B, Esser M, Bucknill A, Russ M, Hofstee D-J, Cameron P, Handley C.  The imaging and classification of severe pelvic ring fractures. Bone Joint J. 2013;95:1396–401. 35. American College of Surgeons. ATLS advanced trauma life support (student course manual). Chicago, IL: American College of Surgeons; 2012. 36. Scolaro JA, Chao T, Zamorano DP.  The Morel-­ Lavallee Lesion: diagnosis and management. J Am Acad Orthop Surg. 2016;24:667–72. https://doi. org/10.5435/JAAOS-D-15-00181. 37. Hak DJ, Olson SA, Matta JM. Diagnosis and management of closed internal degloving injuries associated

16  Pediatric Pelvic Injuries with pelvic and acetabular fractures: the Morel-­ Lavallee lesion. J Trauma. 1997;42:1046–51. 38. Song W, Zhou D, Xu W, Zhang G, Wang C, Qiu D, Dong J.  Factors of Pelvic infection and death in patients with open Pelvic fractures and rectal injuries. Surg Infect (Larchmt). 2017;18:711–5. https://doi. org/10.1089/sur.2017.083. 39. Dente CJ, Feliciano DV, Rozycki GS, Wyrzykowski AD, Nicholas JM, Salomone JP, Ingram WL. The outcome of open pelvic fractures in the modern era. Am J Surg. 2005;190:830–5. https://doi.org/10.1016/j. amjsurg.2005.05.050. 40. Reis ND, Keret D. Fracture of the transverse process of the fifth lumbar vertebra. Injury. 1985;16:421–3. 41. Falchi M, Rollandi GA.  CT of pelvic fractures. Eur J Radiol. 2004;50:96–105. https://doi.org/10.1016/j. ejrad.2003.11.019. 42. Kessel B, Sevi R, Jeroukhimov I, Kalganov A, Khashan T, Ashkenazi I, Bartal G, Halevi A, Alfici R.  Is routine portable pelvic X-ray in stable multiple trauma patients always justified in a high technology era? Injury. 2007;38:559–63. https://doi. org/10.1016/j.injury.2006.12.020. 43. Guillamondegui OD, Mahboubi S, Stafford PW, Nance ML.  The utility of the pelvic radiograph in the assessment of pediatric pelvic fractures. J Trauma Acute Care Surg. 2003;55:236–40. 4 4. Agur AMR, Dalley AF. Grant’s Atlas of anatomy. Philadelphia: Lippincott Williams & Wilkins; 2009. 45. Kenawey M, Addosooki A.  U-shaped sacral frac ture with iliac crest apophyseal avulsion in a young child. J Pediatr Orthop. 2014;34:e6–e11. https://doi. org/10.1097/BPO.0000000000000139. 46. Gary JL, Mulligan M, Banagan K, Sciadini MF, Nascone JW, O’Toole RV. Magnetic resonance imaging for the evaluation of ligamentous injury in the pelvis: a prospective case-controlled study. J Orthop Trauma. 2014;28:41–7. https://doi.org/10.1097/ BOT.0b013e318299ce1b. 47. O’Connor PJ, Barron D.  MRI assessment of pelvic trauma. Semin Musculoskelet Radiol. 2006;10:345– 56. https://doi.org/10.1055/s-2007-972003.

443 48. Webb LX, Gristina AG, Wilson JR, Rhyne AL, Meredith JH, Hansen ST Jr. Two-hole plate fixation for traumatic symphysis pubis diastasis. J Trauma. 1988;28:813–7. 49. White G, Kanakaris NK, Faour O, Valverde JA, Martin MA, Giannoudis PV. Quadrilateral plate fractures of the acetabulum: an update. Injury. 2013;44:159–67. https://doi.org/10.1016/j.injury.2012.10.010. 50. Giannoudis PV, Tzioupis CC, Pape HC, Roberts CS.  Percutaneous fixation of the pelvic ring: an update. J Bone Joint Surg. 2007;89:145–54. https:// doi.org/10.1302/0301-620X.89B2.18551. 51. Gansslen A, Pohlemann T, Krettek C. A simple supraacetabular external fixation for pelvic ring fractures. Oper Orthop Traumatol. 2005;17:296–312. https:// doi.org/10.1007/s00064-005-1134-2. 52. Ponsen KJ, Joosse P, Van Dijke GA, Snijders CJ. External fixation of the pelvic ring: an experimental study on the role of pin diameter, pin position, and parasymphyseal fixator pins. Acta Orthop. 2007;78:648– 53. https://doi.org/10.1080/17453670710014347. 53. Matta JM, Saucedo T. Internal fixation of pelvic ring fractures. Clin Orthop Relat Res. 1989;242:83–97. 54. Scolaro JA, Firoozabadi R, Routt ML.  Treatment of pediatric and adolescent pelvic ring injuries with percutaneous screw placement. J Pediatr Orthop. 2016;38(3):133–7. https://doi.org/10.1097/ BPO.0000000000000790. 55. Starr AJ, Walter JC, Harris RW, Reinert CM, Jones AL.  Percutaneous screw fixation of fractures of the iliac wing and fracture-dislocations of the sacro-iliac joint (OTA Types 61-B2.2 and 61-B2.3, or Young-­ Burgess "lateral compression type II" pelvic fractures). J Orthop Trauma. 2002;16:116–23. 56. Beaule PE, Antoniades J, Matta JM.  Trans-sacral fixation for failed posterior fixation of the pelvic ring. Arch Orthop Trauma Surg. 2006;126:49–52. https:// doi.org/10.1007/s00402-005-0069-2. 57. Banerjee R, Brink P, Cimerman M, Pohlemann T, Tomazevic M. Pelvic ring - Reduction & Fixation ORIF - Ilioiliac plate - Sacrum - AO Surgery Reference n.d. https://www2.aofounda,on.org/ (accessed March 10, 2019).

Traumatic Hip Dislocation in Children

17

Hossam Hosny, Wael Salama, Ahmed Abdelaal, and Mohamed Kenawey

Introduction Traumatic hip dislocations in childhood are rare injuries [1]. The current evidence is based mainly on many case reports and a few small case series. Unlike adults, traumatic hip dislocation in children can be caused by minor trauma and therefore hip fracture-dislocations are not as frequent in pediatrics. Further, there are lower rates of associated injuries and complications and thus better long-term results can be expected [1, 2]. The mainstay of treatment is urgent closed reduction to decrease the risk of avascular necrosis (AVN) and the goal is to achieve a congruent and stable hip joint [3]. Any post-reduction hip joint incongruity and/or instability should be investigated thoroughly to identify and treat its cause. It is not acceptable to assume that the cause of post-reduction joint space widening is simply due to the presence of hemarthrosis and joint laxity. In cases of failed trials of gentle closed reduction—or if a concentric reduction cannot be achieved—an open reduction is always indicated to achieve a stable and congruent hip joint [4, 5].

H. Hosny · W. Salama · A. Abdelaal M. Kenawey (*) Orthopaedic Department, Faculty of Medicine, Sohag University, Sohag, Egypt e-mail: [email protected] © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_17

An important consideration in adolescent hip dislocation is the risk of concomitant proximal femoral epiphysiolysis; a widening of the growth plate at risk of full physeal separation, with an extremely high risk of AVN [6]. The treatment recommendation in this age group is to perform a closed reduction under general anaesthesia to allow for muscle relaxation to help facilitate the reduction under fluoroscopic monitoring [6, 7]. The post-operative regimen following successful closed reduction of a pediatric hip dislocation is variable and is instituted according to the discretion of the treating surgeon. The goal is to allow for adequate healing of the torn capsule, acetabular labrum, and other peri-articular soft tissues to decrease the risk of recurrent hip dislocation [2, 3]. “The goal of treatment of traumatic hip dislocation is to achieve a congruent and stable hip joint”.

Pathophysiology The energy required to dislocate a pediatric hip increases with age and skeletal maturity [8]. While in children less than 10  years of age, a minimal or low energy mechanism can be associated with hip dislocation, children older than 10 years of age typically require a much higher energy mechanism (e.g. motor vehicle accident) 445

H. Hosny et al.

446

to cause a dislocation. The direction of most pediatric hip dislocations is posterior; likely due to the laxity of soft tissues around the hip in younger children. Associated acetabular and femoral head fractures are rare [9]. This would explain the lower incidence of complications like AVN and post-traumatic arthritis in pediatric patients compared to adults [1]. Occasionally, labral detachment and/or infolding prevents a congruent reduction, manifesting with a widened joint space radiographically. Associated trauma to the nearby growth plates can lead to an associated proximal femoral epiphysiolysis [2, 8].

 ascular Supply of the Proximal V Femur At birth, the femoral head chondro-epiphysis is supplied by a vascular network form the metaphyseal branches of the medial and lateral circumflex arteries traversing the femoral neck. Between the age of 4  months to 4  years, progressive ossification of the proximal capital epiphysis occurs. Concomitantly, the proximal femoral physis progressively acts as a barrier to the metaphyseal femoral neck intraosseous vessels, gradually diminishing in size until being virtually non-­existent by the age of 4 years. At this stage, the contribution of the lateral circumflex femoral artery diminishes while the lateral epiphyseal vessels (postero-superior and postero-inferior branches from the medial circumflex femoral artery) become the predominant blood supply to the growing epiphysis. These branches penetrate the capsule posteriorly and ascend vertically (covered by the retinacular folds) along the femoral neck to supply the femoral head. By approximately 3–4  years of age, the lateral postero-­superior epiphyseal vessels predominate and supply the entire antero-lateral portion of the capital femoral epiphysis. By the age of 9–10 years, the vessel of the ligamentum teres reaches the depth of the epiphysis and anastomoses with other vessels, while after closure of the proximal femoral physis, this vessel contributes about 20% of blood supply of the femoral head [10–14].

Avascular Necrosis In traumatic hip dislocation, the ligamentum teres is ruptured, and the posterior capsule is usually torn when the dislocation is posterior. This has the potential to compromise both the retinacular vessels and ligamentum teres blood supply to the femoral head, resulting in avascular necrosis. The risk of avascular necrosis is mainly dependent on the ischemia time rather than on the age or injury mechanism [3]. In children less 12 years of age, avascular necrosis will produce Perthes-like changes in the proximal femoral epiphysis. The prognosis in these cases would be related to the extent of the ischemic insult and the age of the patient. Post-traumatic avascular necrosis, however, usually carries poor prognosis [2]. “The risk of avascular necrosis is mainly dependent on the ischemia time rather than on the age or injury mechanism”

Proximal Femoral Epiphysiolysis Proximal femoral epiphysiolysis is another complication with a poor prognosis [6, 7]. This complication is most common following adolescent traumatic hip dislocations and could be considered as an acute traumatic slipped capital femoral epiphysis (SCFE) with dislocation. Five different patterns were described by Kennon et  al. [7] (Table 17.1). In the first two patterns, the proximal femur is seemingly intact and dislocated with either minimally displaced capital femoral epiphysis (CFE) (type A) or apparently intact, non-displaced femoral epiphysis (type B). Unfortunately, most of the cases with the first two patterns are diagnosed following displacement of the CFE during trials of closed reduction of the dislocated hip. In types C and D, there is partial dislocation of either the femoral neck (type C) or the capital epiphysis (type D) out of the acetabulum while the other component of the proximal femur, the capital epiphysis or the femoral neck respectively, remains located inside the acetabulum. In type E, a complete complex dislocation is present with both components dislocated and separated from each other. It seems that

17  Traumatic Hip Dislocation in Children

447

Table 17.1  Classification patterns of adolescent hip dislocation combined with traumatic SCFE [7] Type A Type B Type C Type D Type E

Displaced epiphyseal separation, with dislocation: intact reduction with SCFE noted. Intact hip dislocation: only the femoral neck or capital femoral epiphysis is reduced into the acetabulum. Partial dislocation: the capital femoral epiphysis remains in the acetabulum while the femoral neck dislocates. Partial dislocation: the capital femoral epiphysis is dislocated while the femoral neck remains in the acetabulum. Complete complex dislocation: both the capital femoral epiphysis and femoral neck are dislocated and completely separated from each other.

decreased physeal stability in adolescence is an important contributing factor in the occurrence of this complication together with the more violent trauma required to dislocate the hip in this older age group [6].

Epidemiology Traumatic hip dislocation in childhood is a rare injury with a reported incidence of approximately 7.5% of all hip dislocations; between 2% and 5% of all joints dislocations [1, 15]. Male predominance was reported with a ratio of 2:1 to 3:1 [1, 8, 16, 17]. The age range of reported cases was from 11  months to 16  years [8, 16, 18]. The mechanism of injury is dependent on the patients’ age and skeletal maturity status (see section “Pathophysiology”) [8, 9, 19]. There is an age dependent bimodal distribution for the clinical presentation of traumatic hip dislocation in childhood. Unlike pediatric femoral neck fractures—for which high energy trauma is the typical mechanism—lower energy mechanisms such as running, tripping or simple falls, can cause traumatic hip dislocation in younger children (e.g. less than 10  years old). In older children and adolescents, more violent trauma (e.g. contact sports, motor vehicle accidents) is required to dislocate the hip [4, 8, 9, 19]. The reasons for this are likely due to the shallow, cartilaginous (soft) acetabulum combined with proximal femoral valgus and increased femoral neck anteversion in younger children. Moreover, the hip joint is surrounded by more elastic ligamentous support as compared to adults which would explain the susceptibility of those hips to dislocate with minimal trauma (Fig. 17.1) [2, 20].

Pearson and Mann [19] studied 24 children with traumatic posterior dislocation of the hip and they grouped their patients into three age groups (0–5  years, 6–10  years and 11–16  years). They found that, for the older the age group, the more violent the trauma that is required to dislocate the hip. Offierski [8] reported the same when he reviewed 33 children and graded the trauma mechanisms into three grades: mild injuries caused by running, tripping and falling; moderate injuries as a result of excessive speed like in skiing and cycling or excessive force as in football; and severe injuries as caused by high energy impact or crushing. He found that half of children aged ≤10-years had their hips dislocated by mild trauma while in children >10-years, the mechanism of injury was by a moderate or severe trauma. In another review of 35 skeletally immature patients who suffered from traumatic hip dislocation, Vialle et al. [4] found that dislocations caused by high energy trauma (21 patients) were in significantly older children compared to low energy trauma dislocations (average age 11.8 years and 8.5 years respectively). The majority of pediatric hip dislocations are posterior, representing 71–94% of these cases [8, 16, 19]. In adults, most posterior hip dislocations are associated with fracture while simple dislocations (Type I Thompson-Epstein) represent only about 33% of these injuries [1]. By contrast, femoral and acetabular fractures are rarely associated with hip dislocation in childhood [3, 8, 19]. In the classic study of Epstein [1], 17% of children (aged 2–15 years) with posterior hip dislocations were associated with fracture (Types II to V Thompson-Epstein). A similar low rate of associated fracture—27% of the children with posterior traumatic hip dislocation—was reported by Vialle and colleagues [4].

H. Hosny et al.

448

a

b

Fig. 17.1 (a) Three-year old boy who had dislocation of the left hip due to minor trauma while running that was missed for 5 days. (b) Post-reduction pelvic X-ray shows a congruent joint. Predisposing anatomic factors—includ-

ing femoral neck valgus and a shallow acetabulum—helps explain how such a low energy mechanism could result in a hip dislocation

The age and skeletal maturity status of children with traumatic hip dislocation changes the frequency and severity of concomitant injuries [9]. The higher energy mechanisms of injury required to dislocate a hip with increasing age, brings a proportional increase in these additional injuries; reported to be present in 30% of cases. These include head injuries, pelvic trauma and long bone fractures [8]. In a review of literature for children with traumatic hip dislocation under age 7 years, 8 out of 51 children (15%) had associated injuries; mainly lower extremity and pelvic fractures [9].

Table 17.2  Thompson–Epstein classification of posterior hip dislocation [21]

Classifications Traumatic hip dislocation has been classified broadly according to the direction of dislocation (anterior, posterior or central) and by pathoanatomy. The most popular classification system for posterior hip dislocation is that introduced by Thompson and Epstein in 1951 [21]. This classification differentiates simple dislocation (without associated fractures; Type I) from those with either associated acetabular rim fractures (Types II and III), acetabular floor injury (Type IV), or femoral head fractures (Type V) (Table 17.2). The

Type I Type II Type III

Type IV Type V

With or without minor fracture With single large fracture of the posterior acetabular rim With a comminuted fracture of the posterior rim of the acetabulum with or without a major fragment With fracture of the acetabular rim and floor With fracture of the femoral head

acetabular floor fractures encountered commonly involve the posterior column or are transverse, with or without associated posterior wall fractures. On the other hand, Epstein divided anterior dislocations into either superior (pubic) or inferior (obturator) dislocations, further subdivided according to the presence of associated femoral head or acetabular fractures (Table 17.3) [1]. The Stewart-Milford Classification System [22] includes grades that increase with increasing injury severity. In this system, grades progress from simple dislocations (without associated acetabular or femoral head/neck fractures) to more complex fracture dislocations. They stressed the importance of gross hip instability in cases with associated acetabular fracture and whether the

17  Traumatic Hip Dislocation in Children

remaining intact acetabular rim would ensure stability following reduction or not (Table 17.4). Traumatic epiphysiolysis in association with pediatric and adolescent hip dislocations is a very important and grave complication. Barquet and Vecsei [23] described two types: Type I where the CFE dislocated and there is complete separation of the femoral epiphysis and Type II where the hip is dislocated and there is incomplete separation of the femoral epiphysis. More recently, Kennon et al. [7] described their own classification for this association which they categorized into five types. In the first two types, the dislocated proximal femur seems intact on initial radiographic presentation with the epiphyseal separation being discovered following closed reduction trials, displacing the CFE from the femoral neck. In the next two types, there is partial dislocation of either the CFE or displacement of the femoral neck from the acetabulum while the other component is still located. In the last type, both components are dislocated and fully separated from each other (Table 17.1). For the evaluation of the long-term results of treating simple hip dislocations and fracture dislocations, Thompson and Epstein [21] proposed clinical and radiographic criteria which categorized the results into either excellent, Table 17.3  Epstein classification of anterior dislocation of the hip [1] A  1.  2.  3. B  1.  2.  3.

Pubic (superior) With no fracture (simple) With fracture of the head of the femur With fracture of the acetabulum Obturator (inferior) With no fracture (simple) With fracture of the head of the femur With fracture of the acetabulum

449

good, fair, or poor (Tables 17.5 and 17.6). Another clinical scoring system is that described by Harris (the Harris hip score system) [24]. The Harris hip score depends on clinical criteria which are intimately related to the activities of daily living and it doesn’t include radiographic results. An online tool can be used to calculate the Harris hip score [25].

Clinical Evaluation Traumatic hip dislocation in childhood might be caused by minimal trauma particularly in young children. Typically, the child with hip dislocation would present with a history of trauma (including low energy mechanisms for younger children as discussed above), inability to walk, and an abnormal position of the lower extremity. The position of the lower extremity is indicative of the direction of dislocation. As in adults, about 90% of pediatric traumatic hip dislocations are posterior, while anterior and inferior dislocations are rare. In cases of posterior dislocation, the hip is held in flexion, adduction and internal rotation; while for anterior dislocations, the hip is extended, abducted and externally rotated. For the rare case of inferior dislocation, there is extreme hip abduction and flexion, with the thigh pressed against the abdomen with the knee flexed [2, 26]. Pre-reduction and post-reduction neurovascular examinations of the affected lower extremity are mandatory. With posterior dislocation, the sciatic nerve is most commonly injured while for anterior dislocations, femoral nerve injury is more common [27]. Traumatic hip dislocation in children can be easily missed in multi-trauma patients; particularly those with concomitant ipsilateral femoral

Table 17.4  Stewart and Milford’s classification [22] Grade I Grade II Grade III Grade IV

Simple dislocation without fracture or with a chip from the acetabulum so small as to be of no consequence Dislocation with one or more large rim fragment, but with sufficient socket remaining to ensure stability after reduction Explosive or blast fracture with disintegration of the rim of the acetabulum that produces gross instability Dislocation with a fracture of the head or neck of the femur

H. Hosny et al.

450 Table 17.5  Clinical criteria for evaluation of the results of treatment of hip dislocation [21] Excellent (normal) •  No pain •  Full range of motion of the hip •  No limp • No roentgenographic evidence of progressive changes Fair (moderate changes) •  Any one or more of the following:  –  Pain but not disabling   – Limited motion of the hip, no adduction deformity   –  Moderate limp  –  Moderately severe roentgenographic changes

Good (minimal changes) •  No pain •  Free motion (75%) of the normal hip •  No more than slight limp •  Minimum roentgenographic changes Poor (severe changes) •  Any one or more of the following:  –  Disabling pain   –  Marked limitation of hip motion, or adduction deformity   – Re-dislocation   –  Progressive roentgenographic changes

Table 17.6  Radiographic criteria for evaluation of the results of treatment of hip dislocation [21] Excellent (normal) • Normal relationship between femoral head and acetabulum •  Normal articular cartilaginous space •  Normal density of the femoral head •  No spur formation •  No calcification in the capsule Fair (moderate changes) • Normal relationship between femoral head and acetabulum •  Any one or more of the following:  –  Moderate articular cartilaginous space  – Mottling of the femoral head, areas of sclerosis and decreased density  –  Moderate spur formation  –  Moderate capsular calcification  – Depression of the subchondral cortex of the femoral head

fractures—which masks the typical hip positioning with dislocation—and/or head injuries [2, 8]. A hip dislocation may not be expected for cases with minor trauma; particularly in young children where the chance of this injury occurring is much higher. The presence of persistent pain, inability to weightbear, limb length inequality, restricted range of motion, and abnormal hip positioning should raise the suspicion of a traumatic hip dislocation in these cases (Fig. 17.1) [2]. “Pre-reduction and post-reduction neurovascular examinations of the affected lower extremity are mandatory”

Good (minimal changes) • Normal relationship between femoral head and acetabulum •  Minimum articular cartilaginous space •  Minimum de-ossification of the femoral head •  Minimum spur formation •  Minimum calcification in the capsule Poor (severe changes) • Almost complete obliteration of the articular cartilaginous space •  Relative increase density of the femoral head •  Subchondral cyst formation •  Formation of sequestra •  Gross deformity of the femoral head •  Sever spur formation •  Acetabular sclerosis

Essential Clinical Tests

• Assess the position of the hip to help in diagnosis and determine the direction of dislocation • A neurovascular examination is essential before and after closed reduction of the dislocation • Have a high index of suspicion for hip dislocation in younger children, where low energy trauma can be causative

17  Traumatic Hip Dislocation in Children

451

Imaging

undisplaced physeal fracture during closed reduction [6, 7]. For children identified to have a traumatic hip dislocation on an AP pelvis X-ray, the entire femur should be imaged to exclude concomitant shaft and condylar fractures, and vice versa [10]. Post-reduction X-rays should assess the quality of reduction and the presence of any missed associated injuries (Figs. 17.2 and 17.3). Post-­reduction

The gold standard diagnostic tool for traumatic hip dislocation in both adults and children is the anteroposterior (AP) pelvic radiograph; allowing for side to side comparison and for the identification of concomitant fractures (Fig.  17.2) [1]. Careful assessment of the integrity of the growth plate is essential to avoid displacing an occult,

a

b

Fig. 17.2 (a) Ten-year old boy had traumatic dislocation of the right hip as a result of falling from a height. (b) The post-reduction X-ray showing concentric reduction

a

b

Fig. 17.3 (a) Twelve-year old boy with traumatic left hip dislocation. (b) Post-reduction pelvic radiograph shows an incongruent reduction and widened joint space as compared to the right (normal) side

452

joint asymmetry is defined as more than a 2 mm difference between the affected and normal hip joint spaces; measured from the medial edge of the femoral head ossification center to the lateral aspect of the acetabular tear drop on an AP pelvic radiographs (Fig. 17.3) [3]. Causes of post-reduction joint asymmetry include hemarthrosis, entrapped soft tissues, infolded labrum, or intraarticular osteochondral fragments [3, 8]. Any postreduction joint space widening should be evaluated with CT or MRI scans to determine the cause (Figs. 17.3 and 17.4). Routine post-reduction pelvic CT scans are not recommended in children. Mehlman and colleagues [3] reported the following indications for post-reduction CT scans: (1) fracture dislocations, (2) gross instability following closed reduction, and (3) post-reduction joint space widening 3 mm or greater as compared to the uninjured side. CT is useful in detecting intra-articular entrapped osteochondral fragments and for evala

b

Fig. 17.4 (a) CT scan of the boy in Fig. 17.2 showing an incarcerated osteochondral fragment between the articular surfaces of the femoral head and the acetabulum. (b) An open arthrotomy was performed during which the labrum was found to be infolded and a rim avulsion from the posterior acetabular wall was identified, causing this incongruent reduction of the hip. The labrum was repaired back to its anatomic position and the reduction was concentric

H. Hosny et al.

uating associated acetabular and femoral head fractures post dislocation (Fig. 17.4). In the series of Vialle and colleagues [4], post-reduction joint asymmetry was identified in 8 of 35 pediatric patients with traumatic hip dislocation; with 5 of the 8 demonstrating intra-articular osteochondral fragments and the other 3 cases with associated acetabular or femoral head fractures on CT. MRI would better show soft tissue entrapment, interposed labrum and loose cartilage fragments which might be underestimated by CT scans [2, 26]. MRI is also useful in detecting any pathoanatomic causes of recurrent hip dislocation, such as capsular tears/avulsions, labral detachments, and fractures affecting the cartilaginous portion of the posterior acetabular wall [5, 18, 28]. Novais and colleagues [5] recommended the use of MRI scans for all children with a non-­concentric hip joint reduction and an open triradiate cartilage to fully assess the cartilaginous posterior acetabular wall which doesn’t completely ossify until age 13 years. The identification of soft tissue pathologies associated with recurrent hip dislocation and be inferred through the use of arthrography. The utility of hip arthrogram for cases of recurrent hip dislocation following closed reduction was reported by Wilchinsky and Pappas [28] who successfully diagnosed a capsular defect and labral bucket handle tear in a small case series. Some surgeons recommend obtaining post-­ reduction bone scintigraphy or MRI scans to evaluate the femoral head vascularity and to predict the development of avascular necrosis [3, 4]. This was investigated in a study by Mehlman and colleagues [3] where 14 of 42 patients with traumatic hip dislocation underwent post-reduction evaluation of femoral head perfusion using Tc99m bone scintigraphy. They found that the sensitivity of the bone scan in predicting the development of avascular necrosis was 0% while the specificity was 100%, with a positive predictive value of 0% and a negative predictive value of 86%. Vialle and colleagues [4] reviewed 35 cases with traumatic hip dislocation, 25 who underwent bone scan to assess femoral head vascularity. Four patients demonstrated increased radiotracer uptake which normalized within few

17  Traumatic Hip Dislocation in Children

weeks without consequences. Another 4 patients demonstrated a transiently decreased uptake, half of whom developed avascular necrosis despite later normalization of the bone scan. Both studies concluded that the bone scan is not a reliable technique to predict avascular necrosis following pediatric hip dislocation. MRI was also found to be unreliable in this regard. Vialle and colleagues [4] performed MRI scans post-reduction in their small series of 11 patients with traumatic hip dislocation and found that diminished T1 epiphyseal signal (suggesting ischemia) did not reliably predict the development of avascular necrosis. “no radiological tests can predict future development of AVN consistently”

Essential Imaging

Plain radiographs • Identification of concomitant fractures • Look for joint space asymmetry • Assess physeal integrity (traumatic epiphysiolysis) Cross-sectional imaging • CT or MRI for assessment of interposing structures associated with joint space asymmetry post reduction • Identification of causes of post reduction joint instability (e.g. posterior wall fracture) • MRI for assessment of suspected labral tear or fractures of the cartilaginous part of the acetabulum

Non-operative Management Emergency Department Versus Operating Room Closed Reduction A diagnosis of traumatic hip dislocation mandates urgent closed reduction in all cases. Closed reduction can be performed either under conscious sedation in the emergency department (ED) or under general anaesthesia. In either case,

453

achieving adequate muscle relaxation helps to facilitate the reduction. For children less than 7  years old, Bressan and colleagues [9] found that closed reduction under conscious sedation in the ED was equally as safe and effective as compared to reduction in the operating room. Reducing the hip in the ED allows for a shorter interval between the time of injury and reduction; reported to be an important prognostic factor with respect to the development of AVN [3, 9]. “Traumatic hip dislocation should be treated on urgent basis with closed reduction performed within the first 6 hours post injury to avoid the development of AVN”. With respect to closed reduction under conscious sedation, no more that 2–3 gentle attempts should be performed. Failure to reduce the hip under conscious sedation is an indication to bring the child to the operating room under general anesthesia, where muscle paralysis which can greatly facilitate reduction. As previously discussed above, there is a risk of unrecognized proximal capital epiphysiolysis in older children and adolescents [6, 9, 29]. In such age group, the capital femoral epiphysis is structurally weak and may be further affected by the trauma that caused the dislocation, that’s why there is a risk of inadvertently displacing an unrecognized non-displaced capital femoral physeal injury which is associated with 100% rate of AVN. As such, it is recommended that an adolescent with traumatic hip dislocation and an open physis should have a gentle closed reduction under general anaesthesia to allow for adequate muscle relaxation and a careful assessment of the state of the physis under image intensification. If there is any evidence of proximal femoral physeal separation during reduction maneuvers, the capital epiphysis should first be stabilized with a screw with the hip dislocated prior to reducing the hip [6, 7]. “The risk of inadvertently displacing an unrecognized non-displaced capital femoral physeal injury in traumatic hip dislocation is associated with 100% rate of AVN”

H. Hosny et al.

454

 echniques: Closed Reduction T of Traumatic Hip Dislocation Different techniques for closed reduction of posterior hip dislocations have been described. The Bigelow maneuver is performed with the patient in the supine position [30]. The surgeon grasps the ipsilateral limb at the ankle with one hand, puts the opposite forearm behind the knee, and applies longitudinal traction in the axis of the femur. Internal rotation, adduction, and flexion of 90° or more relaxes the Y capsular ligament and allows the surgeon to bring the femoral head to the level of the acetabulum posteriorly. The femoral head is then gently levered into the acetabulum by abducting, externally rotating, and extending the hip. The assistant surgeon provides downward pressure on the pelvis during this maneuver. The maneuver must be used with great caution to avoid femoral neck or head fracture. Allis maneuver is another method to reduce posteriorly dislocated hips [31]. The patient is placed in the supine position and the pelvis is stabilized by an assistant. The hip and knee are kept in the flexed position with the thigh in slight adduction and internal rotation. The treating surgeon then applies longitudinal traction in the axis of the femur with the forearm behind the knee, lifting the femoral head over the posterior acetabular rim, through the capsular rent, and back into the acetabular socket. The hip and knee are then gradually extended. In Stimson’s maneuver [32], the patient is placed prone with the lower limbs hanging over the edge of a table. The surgeon holds the affected knee and hip flexed 90° and then applies gentle downward pressure in an attempt to bring the posteriorly dislocated femoral head over the ­posterior rim of the acetabulum. This is done while gently rocking the hip with internal and external rotation motions, allowing the hip to slip back into the socket. The assistant surgeon applies downward pressure from above to stabilize the pelvis during this maneuver. For anterior dislocation of the hip, the reverse Bigelow maneuver can be used. For this technique, the hip is held in partial flexion and abduction. Then, one of two reduction methods may be

used. The first is a lifting method in which a firm jerk is applied to the thigh, which may result in reduction. If that fails, the second method applies traction in the line of the thigh and the hip is then sharply internally rotated, adducted, and extended. For all these methods, care must be taken to avoid fracture of the femoral head or neck [14]. After treatment following successful reduction of traumatic hip dislocation is discussed in the following section.

Essential Non-operative Management

• Ensure adequate muscle relaxation to facilitate closed reduction, either using conscious sedation in the ED, or under general anesthesia • For older children and adolescents with open physes, consider performing the closed reduction under general anesthesia/with fluoroscopy to avoid displacing an occult epiphysiolysis • Bigelow, Stimson, and Allis maneuvers have all been described for closed reduction, which can be interchanged as needed to reduce the hip • Anterior hip dislocations can be reducted with a “reverse” Bigelow maneuver • Look for joint space asymmetry post reduction and investigate accordingly

Operative Management Indications for Open Reduction The causes of difficult or unsuccessful closed reduction can be grouped into soft tissue or bony factors. Closed reduction may not be successful due to interposed posterior capsule; as reported by Pearson and Mann [19] for one patient where the femoral head had button-holed through the capsule and could not be reduced through the small capsular rent. Cases with fracture dislocation and large displaced osteochondral fragments might impede closed reduction trials. Closed reduction

17  Traumatic Hip Dislocation in Children

usually fails in neglected dislocation for more than 14 days, due to soft tissue contractures and the formation of adhesions around the femoral head associated with limb shortening [16]. Epstein recommended primary open reduction of all fracture dislocations of the hip joint types II to V according to Thompson-Epstein classification. The objective was to remove loose osteochondral fragments and stabilize any associated acetabular fracture [1]. In pediatric hip dislocation, the most commonly seen indications for open reduction or surgical arthrotomy are: (1) failed closed reduction, (2) non-concentric reduction with joint space asymmetry postreduction and (3) growth instability following closed reduction [2–4]. The approach for open reduction or joint arthrotomy should be chosen according to the direction of the dislocation, i.e. posterior approach is used in posterior dislocations and anterior approach for anterior dislocations [2]. Some authors recommended the use of surgical hip dislocation for cases with non-concentric joint reduction to best assess and treat the intra-­ articular pathology [5]. The rationale for using surgical hip dislocation instead of the traditional Kocher-Langenbeck posterior exposure is not to further jeopardize the femoral head blood supply while redislocating the hip to retrieve intra-­ articular loose fragments.

 rthrotomy in Non-concentric Hip A Joint Reduction Is it always required to do an open arthrotomy for all cases with post-reduction joint asymmetry? Mehlman et al. [3] suggested that open hip joint arthrotomy is required only in cases of incongruent reduction associated with peri-acetabular fractures. They successfully treated four patients with non-concentric reduction without fracture conservatively, as they thought that the joint space widening in these cases was due to hemarthrosis. In a small series five patients with ­ non-­ concentric reduction, Offierski and colleagues [8] described the successful use of traction in one patient while the remaining four

455

patients had an arthrotomy via posterior approach. For the patients requiring arthrotomy, an infolded posterior acetabular labrum was the main factor preventing congruent reduction in all cases. Vialle et  al. [4] recommended joint exploration for all patients with non-concentric reduction. In their small series of eight patients with joint space widening, they found entrapped intra-­ articular osteochondral fragments (five patients), associated acetabular fracture (one patient) and osteochondral avulsions of the femoral head (two patients), as blocks to reduction. Novais and colleagues [5] reported on the use of surgical hip dislocation in six patients with incongruent reduction for the identification and treatment of associated intra-articular pathologies. These treatments included: labral repair (five patients), fixation of posterior acetabular wall fracture (one patient), fixation of large femoral head osteochondral avulsion (one patient), and excision of an osteochondral fragment and interposed labral tissue (one patient). Following successful reduction, a careful neurovascular examination is performed. Nerve exploration should be undertaken in all patients with an abnormal (i.e. changed from pre-­reduction status) examination post reduction [33].

Treatment Following Reduction After treatment is very variable between different authors and surgeons. The choice of after treatment protocol is dependent on many factors including: age, type/direction of dislocation, method of treatment, and post-reduction hip stability [1–3]. Epstein [1] recommended traction for 3–4 weeks for cases with simple hip dislocation, to allow for healing of the capsular tear. For cases with associated acetabular rim or floor fractures, he suggested 6–8 weeks in traction. Other authors have reported the use of either hip spica for 6  weeks or variable periods of traction followed by protected weight bearing post reduction [3, 8]. Vialle et al. [4] recommended only simple protected non-weight bearing for at least 6 weeks, but there is a risk of non-compliance in young

456

children. We think that the protocol devised by Herrera-Soto and Price is the most appropriate: immobilization in spica cast or bed rest with an abduction splint for children younger than 10 years; protected weight bearing for 6–8 weeks for older children and adolescents [2].

Complications Due to the rarity of the condition and the fact that our current knowledge is based on case reports and small case series, the frequencies of associated complications are not well established. In particular, complications requiring long term follow up to detect—as the case for AVN, recurrent hip dislocation, coxa magna, post-traumatic arthritis and triradiate cartilage fusion—are underestimated [3, 16]. A good example of this can be seen in classic article of Epstein [1], where two cases of traumatic hip dislocation were rated as having good result after 4–5  years but this rating was changed to poor 12–13 years later due to the development of post-traumatic arthritis.

 ciatic Nerve Injury S Sciatic nerve injury is a well-known complication of posterior hip dislocation in adults, however it is rarely encountered in children [1]. In the case series of Vialle and colleagues [4], only 1 patient out of 35 (3%) had a sciatic nerve injury; a 15-year old boy with a posterior hip dislocation due to a motor vehicle accident. The review of literature conducted by Bressan and colleagues [9] identified 76 cases of traumatic hip dislocation in children younger than 7  years. In their review, they found no patients with sciatic nerve injury. Their explanation was that, in this younger age group, a low energy mechanism of injury is usually causative; with less impact on surrounding tissues including neurovascular structures. Incongruent Reduction Soft tissue interposition resulting in an incongruent joint post reduction is an important complication following closed reduction of pediatric traumatic hip dislocation [3, 8]. Up to 35% of the

H. Hosny et al.

cases in one study had non-concentric reduction and joint asymmetry [5]. Causes of incongruent reduction include: entrapped intra-articular osteochondral fragments, capsular and labral interposition, torn ligamentum teres, hemarthrosis, or post-traumatic joint laxity [2, 5, 26]. An incongruent reduction would be evident on anteroposterior pelvic radiographs as joint space widening greater than 2 mm as compared to the other side. In the case series by Mehlman and colleagues [3], 9 patients out of 42 had joint asymmetry post-reduction; associated with peri-articular fractures (4 patients), or hemarthrosis by exclusion (i.e. no fracture or joint instability; 4 patients). In the remaining patient with joint asymmetry, a CT scan was obtained which failed to detect any entrapped soft tissue or osteochondral fragments. In the series of Offierski [8], 5 out of 33 patients (15%) had soft tissue interposition with widened medial hip joint space post closed reduction. In 4 of these patients, an infolded posterior acetabular labrum was identified intraoperatively which prevented a congruent reduction. In the study of Novais et al. [5], 6 out of 23 children with posterior hip dislocation had post-­ reduction joint asymmetry. Intraoperative findings during surgical hip dislocation for those 6 children—in addition to 2 other patients with recurrent hip dislocation—were posterior labral avulsions for all, a fractured cartilaginous posterior acetabular wall (3 patients) and femoral head lesions (6 patients).

Avascular Necrosis The first description for the development of AVN following traumatic hip dislocation in children was in the classic report by Choyce, who reviewed the literature and found only two patients with this complication [16]. It was previously thought that avulsion of the ligamentum teres with its blood supply to the femoral head was the main cause of AVN in hip dislocations [1, 16]. The work of Trueta and Ogden [10, 12] formed our current understanding for the vascular supply of the proximal femur—with primary contributions from the medial femoral circumflex artery—and its changing patterns with growth.

17  Traumatic Hip Dislocation in Children

The incidence of AVN was reported to be between 4% and 12% for pediatric hip dislocations. The most important factors correlated with AVN include: a substantial delay in obtaining reduction, and the severity of trauma [3, 4, 19, 34, 35]. Unfortunately, in most of the studies, the follow up period was short and thus inadequate to detect an evolving AVN; which may develop up to 3 years post injury [3, 19]. As such, the reported AVN rate is likely underestimated. Many authors recommend urgent closed reduction within the first 6 h after trauma to help reduce the incidence of AVN [1, 3]. Mehlman and colleagues [3] reported 5 cases of AVN (12%) out of 42 traumatic hip dislocations in children. Of those cases with AVN, 4 had closed reduction greater than 6 h; increasing the risk of AVN by 20-fold. Hence, the timing of closed reduction was significantly associated with the development of AVN.  With respect to other potential factors, neither age, injury mechanism, or the presence of associated injuries were contributory. In the study of Vialle et al. [4], 5 out 35 pediatric patients had their hip dislocations reduced greater than 6  h post injury with 2 of these patients developing AVN.  None of the remaining 30 patients who underwent urgent closed reduction less than 6  h, however, developed AVN.

Coxa Magna Coxa magna is caused by a mild growth disturbance, defined as a greater than 2  mm difference in femoral head diameter compared to the non-­affected side. It is a well reported sequelae of traumatic hip dislocation of childhood but is rarely symptomatic. The incidence has been reported as being between 17% and 53% [3, 8]. Vialle and colleagues [4] identified coxa magna in 7 patients out of 35 (20%) children following hip dislocation, all less than 10-years old at the time of injury (47% of 15 children less than 10 years old). In a literature review by Bressan and colleagues [9], for children aged 7 years or younger, the incidence of coxa magna was 10% following acute dislocations and was more than double (26%) for neglected dislocations.

457

Mehlman and colleagues [3] had 7 cases (17%) of coxa magna out 42 traumatic hip dislocations. The femoral head diameter difference in their series ranged from 2 to 10 mm. The cause of coxa magna was thought to be due to post-­ traumatic hyperemia and synovitis [3, 36].

 ecurrent Traumatic Hip Dislocation R Although recurrent traumatic hip dislocation is a very rare complication, it has been reported following traumatic hip dislocation in both children and adults [18, 37]. Predisposing causes are (1) age under 8  years, (2) the presence of acetabular fracture with deficient posterior bony support, (3) the severity of trauma (4) dysplastic hip, (5) infection, (6) pre-existing paralytic conditions, (7) delayed reduction, and (8) inadequate immobilization of the hip with consequent incomplete healing of the posterior capsule [18, 37]. In his literature review of traumatic hip dislocation in childhood [16], Choyce identified 5 patients out of 59 (age range, 3–9 years) who had recurrent hip dislocation. All had a marked delay before the initial closed reduction was performed (between 4 and 10  weeks post injury). The data regarding the mechanism of injury, the details of the recurrent dislocation episodes, and the treatment outcomes were not reported. Barquet [18] reviewed the literature for reported cases of recurrent traumatic hip dislocation in childhood, excluding cases with insufficient data, acetabular dysplasia, or a predisposing general condition like hypothyroidism or Down’s syndrome. He analyzed different factors that might be correlated to the risk of recurrence and found age under 8 years to be the most important predictor. This increased risk of recurrence seen in younger children was discussed above (section “Pathophysiology”). A failure of adequate post-reduction immobilization is a potential cause of recurrent dislocation. Pearson and Mann [19], however, found that none of the 24 children in their series had recurrent dislocation despite having different methods and duration of immobilization post reduction. Mehlman and colleagues [3] reported 1 case out of 42 with recurrent hip dislocation. This was

458

an 8-year old boy who had posterior dislocation during a football game. He underwent an uneventful closed reduction, with no gross instability or joint space asymmetry post reduction. He was mobilized on crutches. Six-months later, he had a second dislocation after a simple fall which was treated successfully again by closed reduction. Post-reduction MRI showed no interposed soft tissue and no capsulo-labral detachment. Post-­ reduction hip spica for 6 weeks was used at that time for immobilization. Gaul [38] reported another case of recurrent dislocation of the hip. This was a boy who had his first dislocation at the age of 6 years which was reduced under general anaesthesia. Despite immobilization in a hip spica for 6 weeks, he sustained two additional episodes of recurrent dislocation (at 1 and 3 years post injury, respectively). Following the third dislocation, surgical exploration via a posterior approach identified a detached posterior capsule from a rounded acetabular rim, with an intact labrum. The capsule was repaired to the acetabular labrum and the patient reported no difficulty related to the hip at 5-years postoperative follow up. Wilchinsky and Pappas [28] reported two cases of recurrent dislocation (a 4-year old girl and 7-year old boy) who were both immobilized in bilateral Buck’s traction for 8  weeks following their initial reduction. There was no evidence of bony acetabular rim injury. They had their second dislocation 2  years and 19  months, respectively. This second dislocation was successfully reduced under general anaesthesia and a hip spica cast for 8 weeks was used for immobilization. Following removal of the hip spica cast, bilateral hip arthrograms were obtained in both cases. For the 6 year old, there was an evidence of posterior capsular defect and a bucket handle labral tear who subsequently underwent an open posterior labral repair with plication of the posterior capsule. No specific pathology was documented in the arthrogram of the second patient and no further treatment was required. Both were followed up for 9 and 13 years following the second reduction, respectively. Both were active in sports

H. Hosny et al.

without specific complaints regarding the hip joints. Novais and colleagues [5] reported 2 cases out of 23 with recurrent hip dislocation, both of whom had surgical hip dislocation. Intraoperatively, a cartilaginous posterior wall fracture, associated with a labral avulsion and disruption of the posterior capsule, was identified in both patients. Both had subsequent fixation of the labrum using suture anchors and screw fixation for the posterior acetabular wall. The authors recommended the use of MRI in patients with non-concentric reduction and in case of recurrent hip dislocation to evaluate for cartilaginous posterior acetabular wall fractures and/or labral avulsions not easily identified by CT scan.

Proximal Traumatic Epiphysiolysis Proximal femoral traumatic epiphysiolysis in association with traumatic hip dislocation is a grave complication with 100% AVN rate [6, 23]. It was classically described in adolescents in whom the proximal femoral physeal plate is structurally weak and may be further weakened by the trauma that caused the dislocation. Barquet and Vecsei [23] reported 2 cases of traumatic epiphyseal separation with hip dislocation in 2 boys, both 15-years old, with a further 26 patients identified after literature review. They found the age range for this complication to be between 2 and 16 years, with only 56% of these children being adolescents older than 11  years. As such, this complication is not exclusive to adolescents. Herrera-Soto and colleagues [6] reported five adolescents with age range between 13 and 15  years with proximal femoral traumatic epiphysiolysis following hip dislocation. There was no radiologic evidence of injury to the proximal femoral physis in any case. All had trials of closed reduction in the emergency department under conscious sedation and the diagnosis of traumatic epiphysiolysis was made in the post-­ reduction radiographs. All then had open reduction and internal fixation of the physeal fracture followed by reduction of the hip joint. All patients developed AVN at a mean of 9 months post injury (range, 3–13 months).

17  Traumatic Hip Dislocation in Children

To the best of our knowledge, the study conducted by Kennon and colleagues [7] is the largest case series reporting on traumatic hip dislocation combined with proximal femoral epiphysiolysis. They reported 12 patients with age range between 11 and 15  years. They devised a classification system for patterns of traumatic hip dislocation combined with SCFE (Table 17.1). Four of their patients had types A and B injuries, 3 of which were discovered following closed reduction trials with the epiphysis fully displaced. None had pre-reduction radiographic evidence of physeal trauma. In 1 patient, a minimally displaced proximal physeal separation was diagnosed before reduction, managed by pinning in situ with the hip dislocated followed by a gentle closed reduction. Three patients had type C dislocations with the CFE remaining located in the acetabulum with anterior displacement of the femoral neck. Four patients had type D dislocations with the CFE dislocated posteriorly while the femoral neck was located inside the acetabulum. A type E dislocation was found in 1 patient. Ten patients had open reduction (through posterior or anterior approaches) and fixation with screws. One patient had a percutaneous pinning in situ for a Type A injury followed by closed reduction, and the last had only closed reduction and orthosis due to the concomitant severe head injury. The only patient with type A injury that did not develop AVN had percutaneous pinning of the CFE before reduction. One patient with type B injury had partial AVN, 9 had total AVN and 1 patient had ischemia of the proximal femoral neck and nonunion at the physeal separation.

Complications Post Reduction of Hip Dislocation

• • • • • • • •

Incongruent reduction Avascular necrosis Post-traumatic arthritis Sciatic nerve injury Coxa magna Heterotopic ossification Recurrent dislocation Proximal femoral physeal separation

459

Essential Surgical Techniques

• Perform an open reduction for all cases with joint space asymmetry secondary to interposed bony or soft tissues • Perform an open reduction for all cases with gross instability following closed reduction to identify and treat the cause • Choose your surgical approach concordant with the direction of the dislocation • The surgical hip dislocation approach can be useful to identify and treat intra-­ articular pathology associated with hip dislocation • Post-reduction immobilization methods following both closed and open reduction are varied and not well studied

Classic Papers Choyce CC: Traumatic dislocation of the hip in childhood, and relation of trauma to pseudocoxalgia: Analysis of 59 cases published up to January 1924. British Journal of Surgery 1924;12:52–59. Choyce reported a case report of 6-year-old boy who sustained a posterior dislocation of the hip and he reviewed the literature for another 58 reported cases of traumatic dislocation of the hip in childhood up to January 1924. In his review, he found a male predominance, and a posterior dislocation 78% of cases. Choyce found that manipulation under anaesthesia was mostly successful in reducing those hips up to 14 days post injury. Open reduction without femoral head resection was performed in 15 patients for neglected dislocation from 14 days and up to 5 months post injury. He also reported on associated complications. This review was one of the first in this subject area. Epstein HC. Traumatic dislocations of the hip. Clinical orthopaedics and related research 1973;92:116–142. Epstein reviewed 559 traumatic hip dislocations, of them 42 (7.5%) dislocations in children 2–15 years of age, treated in Los Angeles County University of Southern

460

California Medical Center in the period between 1922 and 1970. The most common mechanisms of injury were dashboard trauma causing posterior dislocation or ejection from a motor vehicle resulting in anterior or posterior dislocation. Of the 42 children in this series, 6 (14%) patients had anterior dislocation and 36 (86%) with posterior dislocation. Most of the children with posterior dislocation had a simple dislocation (83%) while 6 (17%) had fracture dislocations. The results of treatment of children with anterior and simple posterior dislocation were either good or excellent in all patients; all except 1 had a closed reduction within the first 24 h. Three patients had type III dislocation, one was rated as good result following primary open reduction, one fair result after delayed open reduction and one poor result due to AVN following closed reduction. Two patients with type IV dislocation had poor results due to neglected dislocation in association with ipsilateral femoral shaft fracture in one, and AVN in the other. There was one type V dislocation with a fair result following closed reduction due to post-traumatic arthritis. Based on the results of this study, Epstein recommended ordering a plain pelvic X-ray in all patients with major injuries. All hip dislocations should be treated urgently with the first 24 h. Multiple trials of closed reduction are contraindicated. Type I dislocation is typically treated with closed reduction while all fracture dislocations types II, III, IV and V should have primary open reduction. The objective is to remove all loose bodies within the joint space and to fix the associated acetabular or femoral neck or head fractures.

Key Evidence Mehlman CT, Hubbard GW, Crawford AH, Roy DR, Wall EJ. Traumatic Hip Dislocation in Children: Long-Term Follow up of 42 Patients. Clinical orthopaedics and related research 2000;376:68–79. Mehlman and colleagues retrospectively reviewed 42 children with open proximal femoral growth plates who had traumatic hip dislocation, with an average clinical follow up of 10  years. The average age at the time of injury

H. Hosny et al.

was 9 years. The mechanism of injury was a low energy trauma in 27 (64%) of 42 patients. Thirtyfive out of 42 (83%) had urgent closed reduction within the first 6 h. On the latest follow up, 31 patients (74%) had no pain, 2 had mild ache or pain and 2 had moderate pain. Thirty-seven patients had no noticeable limp while 3 had mild limp and 2 with moderate or severe limp. The authors stressed the importance of urgently reducing dislocated hips to decrease the incidence of AVN and found that hips that are dislocated for greater than 6 h had a 20-fold increase in the risk of developing this complication. Offierski C.  Traumatic dislocation of the hip in children. Bone & Joint Journal 1981;63:194–197. Offierski reviewed the charts of 33 children who sustained traumatic hip dislocation and were treated at the Hospital for Sick Children, Toronto between 1960 and 1977. The age range was 1.5–14 years with male to female ratio of 2:1. A low energy mechanism caused half of the dislocations in children below 10 years of age, while a more violent trauma was found in older children above 10 years. There were 9 children (27%) with complications in this series, including AVN and recurrent dislocation. After an average of 10 years after the index traum, hips were graded as normal in 16 of 19 children available for followup. Pearson DE, Mann RJ: Traumatic hip dislocation in children. Clin Orthop Relat Res 1973:189–194. Pearson and Mann reported 24 traumatic dislocations of the hip in children from University of Miami teaching hospitals. There were 20 posterior dislocations (83%) and 18 males (75%). Closed reduction was successful in 22 cases, one had open reduction for associated acetabular fracture and the other patient had open reduction following failed closed reduction. After treatment was variable. Nineteen patients had skin traction followed by either non-weight bearing crutch walking or were placed in hip spica, 3 patients had initial spica cast, 1 had skeletal traction and one patient had only bed rest. The follow up ranged from 2  months to 11  years. They grouped their patients into three groups according to the age. The first group included children 0–5 years old (4 patients), all of whom had a low

17  Traumatic Hip Dislocation in Children

energy mechanism and no late sequelae were noted. The second group, 6–10 years old (6 patients), had higher energy trauma and there was one case of AVN identified. The third group included patients 11–16 years of age (14 patients) with motor vehicle accidents being the predominant mechanism of injury. Four anterior dislocations were found in this group. Two patients underwent open reduction and late AVN was not recorded in any case but the followup was short. Vialle R, Odent T, Pannier S, Pauthier F, Laumonier F, Glorion C: Traumatic hip dislocation in childhood. Journal of Pediatric Orthopaedics 2005;25:138–144. Vialle and colleagues reviewed 35 cases of traumatic hip dislocation in skeletally immature patients with minimum follow up of 18 months. The average age was 10  years (range 2–15  years). Open arthrotomy was required in 8 patients due to post-­ reduction joint asymmetry, with CT evidence of intra-articular osteochondral fragments in 5 patients. For the remaining cases, there were associated acetabular or femoral head fractures requiring fixation. After treatment was protected non-weight bearing for at least 6  weeks. AVN developed in 2 patients. Two patients had asymptomatic heterotopic ossification and 7 had coxa magna, all less than 10 years old at the time of injury. According to Harris hip score, 30 patients had normal hips, 4 had scores 75–100 due to stiffness and pain and 1 had poor Harris score 53 due to total AVN, at final followup.

Take Home Messages

• Traumatic hip dislocation should be treated on urgent basis with closed reduction performed within the first 6  h post injury to avoid the development of AVN. • Emergency room closed reduction under conscious sedation is equally effective and safe in young children with traumatic hip dislocation with the advantage of a shorter interval to reduction. • The aim of closed reduction is to achieve a congruent and stable hip joint.

461

• All adolescents with traumatic hip dislocation should have adequate pelvic X-rays to identify subtle trauma to the proximal femoral physis with the risk of proximal traumatic epiphysiolysis in mind. • Closed reduction under general anaesthesia and adequate muscle relaxation in the operating room is essential for all adolescents with traumatic hip dislocation with the reduction being performed under fluoroscopic guidance. If there are any signs of an unstable epiphysis, it should be pinned in situ with the hip dislocated followed by a gentle reduction of the joint. • Only two to three trials of gentle closed reduction are allowed and, in case of failed closed reduction, open reduction—concordant with the direction of dislocation—is indicated. • In all patients with a non-concentric reduction, CT or MRI scans should be ordered to identify the cause of joint space widening. There should be very low threshold to perform an arthrotomy due to the high prevalence of associated offending structures including an infolded avulsed labrum together and posterior cartilaginous acetabular wall fracture. • All hips that have post closed reduction gross instability should have an CT or MRI scan to identify the source of the instability which should always be surgically corrected. • Postoperative immobilization using spica casting is preferable for children younger than 10 years of age as they are usually non-compliant with the weight bearing precautions. • Patients with recurrent hip dislocation should have an arthrogram or MRI scan to detect capsular tears and labral avulsion which should be repaired operatively.

462

References 1. Epstein HC.  Traumatic dislocations of the hip. Clin Orthop Relat Res. 1973;92:116–42. 2. Herrera-Soto JA, Price CT. Traumatic hip dislocations in children and adolescents: pitfalls and complications. J Am Acad Orthop Surg. 2009;17(1):15–21. 3. Mehlman CT, Hubbard GW, Crawford AH, Roy DR, Wall EJ. Traumatic hip dislocation in children: long-­ term followup of 42 patients. Clin Orthop Relat Res. 2000;376:68–79. 4. Vialle R, Odent T, Pannier S, Pauthier F, Laumonier F, Glorion C. Traumatic hip dislocation in childhood. J Pediatr Orthop. 2005;25(2):138–44. 5. Novais EN, Heare TC, Hill MK, Mayer SW. Surgical hip dislocation for the treatment of intra-articular injuries and hip instability following traumatic posterior dislocation in children and adolescents. J Pediatr Orthop. 2016;36(7):673–9. 6. Herrera-Soto JA, Price CT, Reuss BL, Riley P, Kasser JR, Beaty JH. Proximal femoral epiphysiolysis during reduction of hip dislocation in adolescents. J Pediatr Orthop. 2006;26(3):371–4. 7. Kennon JC, Bohsali KI, Ogden JA, Ogden J III, Ganey TM. Adolescent hip dislocation combined with proximal femoral physeal fractures and epiphysiolysis. J Pediatr Orthop. 2016;36(3):253–61. 8. Offierski C. Traumatic dislocation of the hip in children. Bone Joint J. 1981;63(2):194–7. 9. Bressan S, Steiner IP, Shavit I. Emergency department diagnosis and treatment of traumatic hip dislocations in children under the age of 7 years: a 10-year review. Emerg Med J. 2014;31(5):425–31. 10. Ogden JA. Changing patterns of proximal femoral vascularity. J Bone Joint Surg Am. 1974;56(5):941–50. 11. Chung S. The arterial supply of the developing proximal end of the human femur in childhood. A report of six cases. Ann Surg. 1928;88:902–7. 12. Trueta J.  The normal vascular anatomy of the human femoral head during growth. Bone Joint J. 1957;39(2):358–94. 13. Trueta J, Amato VP.  The vascular contribution to osteogenesis. Bone Joint J. 1960;42(3):571–87. 14. Blasier RD, Hughes LO. Fractures and traumatic dislocations of the hip in children. In: Flynn JM, Skaggs DL, Waters PM, editors. Rockwood and Wilkins’ fractures in children. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 913–39. 15. Rieger H, Pennig D, Klein W, Grünert J.  Traumatic dislocation of the hip in young children. Arch Orthop Trauma Surg. 1991;110(2):114–7. 16. Choyce CC. Traumatic dislocation of the hip in childhood, and relation of trauma to pseudocoxalgia: analysis of 59 cases published up to January, 1924. Br J Surg. 1924;12(45):52–9.

H. Hosny et al. 17. Hung NN.  Traumatic hip dislocation in children. J Pediatr Orthop B. 2012;21(6):542–51. 18. Barquet A. Recurrent traumatic dislocation of the hip in childhood. J Trauma. 1980;20(11):1003–6. 19. Pearson DE, Mann RJ.  Traumatic hip dislocation in children. Clin Orthop Relat Res. 1973;92:189–94. 20. Ralis Z, McKibbin B.  Changes in shape of the human hip joint during its development and their relation to its stability. J Bone Joint Surg Br. 1973;55(4):780–5. 21. Thompson VP, Epstein HC. Traumatic dislocation of the hip; a survey of two hundred and four cases covering a period of twenty-one years. J Bone Joint Surg Am. 1951;33-A(3):746–778; passim. 22. Stewart MJ, Milford LW.  Fracture-dislocation of the hip; an end-result study. J Bone Joint Surg Am. 1954;36-A(2):315–42. 23. Barquet A, Vecsei V. Traumatic dislocation of the hip with separation of the proximal femoral epiphysis. Report of two cases and review of the literature. Arch Orthop Trauma Surg. 1984;103(3):219–23. 24. Harris WH.  Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J Bone Joint Surg Am. 1969;51(4):737–55. 25. Kurer M, Gooding C. Orthopaedic scores. Harris Hip Score. http://www.orthopaedicscore.com/scorepages/ harris_hip_score.html. Accessed 11 Aug 2017. 26. Elder G, Harvey EJ.  Surgical images: musculoskeletal. Imaging in musculoskeletal trauma: the value of magnetic resonance imaging for traumatic pediatric hip dislocations. Can J Surg. 2004;47(4):290–1. 27. Ogden JA. Hip. In: Ogden JA, editor. Skeletal injury in the child. 3rd ed. New  York: Springer; 1999. p. 831–56. 28. Wilchinsky ME, Pappas AM.  Unusual complica tions in traumatic dislocation of the hip in children. J Pediatr Orthop. 1985;5(5):534–9. 29. Kutty S, Thornes B, Curtin W, Gilmore M. Traumatic posterior dislocation of hip in children. Pediatr Emerg Care. 2001;17(1):32–5. 30. Bigelow H.  On dislocation of the hip. Lancet. 1878;111(2859):860–2. 31. Allis OH.  An inquiry into the difficulties encountered in the reduction of dislocations of the hip. Philadelphia: Dornan; 1896. 32. Stimson L. An easy method of reducing dislocations of the shoulder and hip. Med Record. 1900;57:356–7. 33. McCarthy JJ, Noonan KJ, Gandhi SD.  Pediatric hip fractures and dislocations. In: Abzug JM, Herman MJ, editors. Pediatric orthopedic surgical emergencies. New York: Springer; 2012. p. 127–40. 34. Hougaard K, Thomsen PB.  Traumatic hip dislocation in children follow up of 13 cases. Orthopedics. 1989;12(3):375–8.

17  Traumatic Hip Dislocation in Children 35. Banskota AK, Spiegel DA, Shrestha S, Shrestha OP, Rajbhandary T.  Open reduction for neglected traumatic hip dislocation in children and adolescents. J Pediatr Orthop. 2007;27(2):187–91. 36. Powers JA, Bach PJ.  Coxa magna. South Med J. 1977;70(11):1297–9.

463 37. Liebenberg F, Dommisse GF.  Recurrent post-­ traumatic dislocation of the hip. J Bone Joint Surg Br. 1969;51(4):632–7. 38. Gaul RW. Recurrent traumatic dislocation of the hip in children. Clin Orthop Relat Res. 1973;(90):107–9.

Part VII Neuromuscular Conditions

The Hip in Cerebral Palsy

18

Jason J. Howard, Abhay Khot, and H. Kerr Graham

Introduction Although the term cerebral palsy was first coined by Osler in 1889 [1] it was Dr. William Little, who first made the connection between musculoskeletal (MSK) deformities and injury to the brain during labour [2]. Hence the term “Little’s disease” became popular for the typical fixed deformities associated with CP.  Bax developed the first commonly used definition in the modern era, defining CP as, “an umbrella term covering a group of non-progressive, but often changing, motor impairment syndromes secondary to lesions or anomalies of the brain arising in the early stages of its development” [3]. A more recent and holistic definition was established by

J. J. Howard (*) Weill Cornell Medicine, Chief of Orthopaedic Surgery, Sidra Medicine, Doha, Qatar e-mail: [email protected] A. Khot Royal Children’s Hospital, Melbourne, Melbourne, VIC, Australia Monash Medical Centre, Melbourne, VIC, Australia e-mail: [email protected] H. K. Graham Royal Children’s Hospital, Melbourne, Melbourne, VIC, Australia National Health and Medical Research Council of Australia (NHMRC) Centre of Research Excellence in Cerebral Palsy (CRE-CP), Canberra, Australia e-mail: [email protected] © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_18

an international working group in 2004 and published in 2005. Cerebral palsy is defined as: “A group of permanent disorders of the development of movement and posture, causing activity limitation, that are attributed to non-progressive disturbances that occurred in the developing fetal or infant brain. The motor disorders of CP are often accompanied by disturbances of sensation, perception, cognition, communication and behavior, by epilepsy, and by secondary musculoskeletal problems [4]”. This definition reflects the recent shift in focus by medical professionals from considering only measures of body structure and function (primarily musculoskeletal concerns) when treating patients with CP, to consider other domains, which have an impact on quality of life [5, 6]. Hip displacement is second only to equinus in terms of frequency of musculoskeletal (MSK) problems, in children with CP. The displaced hip is often clinically silent in younger children but becomes symptomatic with age, coincident with the development of fixed deformities, damage to articular cartilage, sensitization by mediators of inflammation and eventually, degenerative arthritis [7, 8]. The onset and progression of arthritis is thought to be secondary to the abnormal forces imparted to the CP hip and the resulting mismatch of a femoral head and acetabulum that have become progressively dysplastic over time [9]. Even before the onset of painful arthritis, typically commencing in the teenage years, fixed 467

468

deformities associated with muscle contractures can lead to difficulties in standing, walking, comfortable seating, perineal hygiene, and decubitus ulceration—substantial determinants of quality of life in this patient population [10]. Preventative surgical management may be indicated to ­preserve the CP hip, given that salvage options are unpredictable with respect to their ability to relieve pain and have a high risk of post-operative complications and revision surgery [11]. With a prevalence of approximately 2 per 1000 live births, CP is the most common cause of physical disability in children [12]. The prevalence of hip displacement is 35% for the CP population as a whole and increases with functional severity as defined by Gross Motor Function Classification System (GMFCS; see section “Epidemiology”) [13]. As such, the burden of disease secondary to the sequelae of hip displacement is correspondingly large, particularly for children with severe functional impairment. Children at GMFCS Level V have an incidence of hip displacement approaching 90% [13]. The majority of these children will require surgical intervention, ranging from adductor lengthening to reconstructive osteotomies of the proximal femur and acetabulum. Reconstructing the CP hip prior to the onset of painful degenerative arthritis is the major goal, with improved surgical outcomes linked to the initiation of treatment before the underlying dysplasia becomes too severe [14–16]. The goal of this chapter is to provide a comprehensive review of the most up-to-date evidence on the diagnosis and management of hip displacement in CP, while at the same time incorporating the ‘Melbourne philosophy’ into the overall approach presented.

Pathophysiology  he ‘Cerebral Palsies’: Heterogeneity T According to Motor Type CP is a heterogeneous disorder caused by an injury or lesion in the developing brain. The clinical phenotype is related to the size and severity of the brain lesion. Periventricular leucomalacia (PVL) is a brain lesion linked to prematurity in which

J. J. Howard et al.

the white matter injury is associated with loss of inhibitory pathways regulating muscle tone. This lack of inhibition causes muscles to exhibit high tone (hypertonia) and to be increasingly sensitive to velocity-dependent stretch (spasticity). Children exhibiting this pattern can be said to have a ‘spastic’ motor type, the most common motor type in children with CP.  This motor type is the archetype for CP, with prolonged spasticity eventually leading to fixed muscle contractures and other deformities. For lesions involving other areas of the brain, however, different motor types, with their own respective clinical manifestations, may predominate [12, 17]. For example, lesions in the basal ganglia typically manifest with a ‘dyskinetic’ motor type, where involuntary, intermittent muscle contractions cause twisting or repetitive movements. Children with some forms of dyskinesia (e.g. athetosis) may not develop contractures and the outcome of orthopedic surgery in such children is less predictable than in spastic motor types. Similarly, children with hypotonia may not develop contractures but still develop deformities of the proximal femur (e.g. coxa valga) and acetabulum. These deformities are thought to be secondary to an absence of normal joint forces rather than as the sequelae of spastic contractures. Although CP results from a static encephalopathy, associated musculoskeletal aspects, such as hip displacement, are known to be progressive with growth [18]. In spite of the heterogeneity of CP and its inherent phenotypical differences, hip displacement has been found to be ubiquitous across all motor types, except ataxia. This leads to a search for a more universal etiology/pathophysiology, which is unrelated to motor type [13]. Cerebral palsy is the most common cause of the upper motor neuron syndrome in children and has both ‘positive’ features (spasticity, hyperflexia, and co-contraction) and ‘negative’ features (weakness, poor selective motor control, poor balance, and sensory deficits). The negative features are a stronger determinant of a child’s ambulatory potential than the more obvious, positive factors [19, 20]. With this in mind, it may be appropriate to consider weakness and decreased weight-bearing as primary factors responsible for the development of hip displacement, rather than only spasticity and muscle contracture [13, 21].

18  The Hip in Cerebral Palsy

 rivers of Hip Displacement: D Challenging the ‘Traditional’ View The spastic motor type is the most common in CP, manifesting with a velocity dependent (dynamic) increase in muscle stiffness that precedes the development of fixed (static) muscle contracture [17]. These contractures, in turn, have previously been thought to lead to the secondary bony deformities commonly associated with hip, specifically: increased femoral anteversion, coxa valga, and acetabular dysplasia. In this traditional view, spastic, contracted adductor and flexor muscles pull the proximal femur into adduction, flexion, and internal rotation, effectively driving the femoral head out of the joint. It was the development of these contractures and their apparent skeletal effects that had prompted others to label CP as a “short muscle disease”, leaving no doubt as to the primary drivers of the bony deformities encountered [8] (Fig. 18.1). Although this “spastic muscle model” may offer a convenient and intuitive explanation, it does not

469

account for the fact that hip displacement, with its associated deformities of the proximal femur and acetabulum, has been found to occur irrespective of motor type, with hypotonic children displaying the same risk as children with hypertonia, irrespective of the presence or absence of muscle contractures [13]. In a large, population based study of children with CP, functional impairment according to GMFCS level was the main factor associated with hip displacement [13]. As such, factors other than those primarily related to muscle contracture must be considered. Subsequent research has further substantiated the challenge to the traditional view, as explained in this section [12, 22].

Pathoanatomy of Hip Displacement  roximal Femoral Dysplasia P The most important anatomical factors to consider when assessing proximal femoral geometry in CP are (1) the persistence of fetal anteversion, (2) horizontal or lateral inclination of the proximal

©JSchoenecker2018

Fig. 18.1  ‘Traditional’ model for pathophysiology of hip displacement. In this traditional view, spastic, contracted adductor and flexor muscles pull the proximal femur into adduction, flexion, and internal rotation, effectively driv-

ing the femoral head out of the joint leading to acetabular dysplasia. (Used with permission, Copyright Jonathan Schoenecker)

470

femoral physis and (3) the development of coxa valga. These factors have been found to be associated with hip subluxation and dislocation in CP and their correction during reconstructive procedures is necessary for surgical success [14, 15]. Femoral anteversion has been defined as an anteriorly-directed torsion of the femoral neck relative to the alignment of the distal femoral condylar axis and can be located at any point along the femur [22]. In typically developing children, anteversion at birth is approximately 35–45°, a value that decreases with growth to a value of 10–15° at skeletal maturity [23]. It has been suggested that the birth value may result from the influence of forces imparted secondary to intrauterine packaging, with the subsequent decrease moderated by the application of normal joint forces experienced during standing and walking [14, 22]. In children with CP, the decrease in fetal anteversion is either absent or diminished according to the severity of neurologic involvement and is seen concomitantly, though not directly related, to hip migration from the acetabulum [22, 24–26] (Fig. 18.2). This persistence of fetal anteversion was substantiated recently in a large, population based study of children with CP which correlated changes in proximal femoral geometry to disease severity according to GMFCS [22]. In this study of 292 children, the mean femoral anteversion was found to be 36.5°, with a range from 11° to 67.5°. When stratified by GMFCS level, mean femoral anteversion was found to be approxiFig. 18.2  Persistence of fetal anteversion in CP at skeletal maturity (left: typically developing, anteversion approximately 10°; right: CP, anteversion approximately 35°) (Used with permission, Copyright © Kerr Graham)

J. J. Howard et al.

mately 30° for level I, 35° for level II, and 40° for levels III to V.  For non-ambulant children (GMFCS IV and V) there was little or no decrease in anteversion from birth values, reflecting the absence of normal joint mechanics that accompanies standing and ambulation. For GMFCS levels I to III, a significant though far from normal decrease in anteversion was identified, consistent with a delay in the onset, rather than the absence, of the application of joint forces associated with standing and walking. At GMFCS level I, children with near normal ambulation potential, the persistence of femoral anteversion was still quite evident (Fig.  18.3). This may suggest that the determinants of femoral anteversion at skeletal maturity are rooted in the early years of development, when musculoskeletal growth is at its maximum velocity. In high functioning children with spastic diplegia (GMFCS I and II) a delay in reaching independent walking to ages typically ranging from 24 to 36 months, may be sufficient to miss out on a substantial reduction in remodeling of femoral anteversion. The mechanisms behind the persistence of femoral anteversion have not been fully explained. In one study utilizing finite element analysis to simulate proximal femoral growth in the CP hip, a change in applied stress to the proximal femur different from that seen in typically developing children was identified [27]. Specifically, an imbalance between octahedral shear stress (which increases bone growth and ossification) and hydrostatic stress (which decreases bone growth

18  The Hip in Cerebral Palsy

471

GMFCS I

GMFCS II

GMFCS III

FNA = 30°

FNA = 36°

FNA = 40°

GMFCS IV

GMFCS V

FNA

MP = 8%

NSA + MP

MP = 13%

NSA = 136°

MP = 25%

NSA = 141°

FNA = 40° MP = 37%

NSA = 149°

FNA = 40° MP = 46%

NSA = 155°

NSA = 163°

Fig. 18.3  Relationship of NSA and FNA to GMFCS level. NSA neck-shaft angle, FNA femoral neck anteversion. (Reproduced and translated with permission and

copyright © of the British Editorial Society of Bone and Joint Surgery [22])

and ossification) was found. In typically developing children, these two different types of stresses are applied to the proximal femur medially and laterally, respectively, resulting in a normal proximal femoral morphology. When forces consistent with the CP hip were applied using their model, the magnitude and direction of these forces changed such that there was decreased octahedral shear stress medially and an absence of hydrostatic force laterally; the result being a persistence of fetal anteversion as is seen clinically. It was further found that the development and morphology of proximal femoral deformity was proportional to the amount of epiphyseal cartilage deformation at the femoral head. The direction of growth, and thus change in femoral anteversion, was determined by where and by how much the femoral cartilage was deformed. In this model, the application of normal forces caused a decrease in anteversion by 2° over a 6  month period predicted while the application of CP-like forces resulted in only a 1° decrease over the same period. As stated above, although hip displacement has been known to be associated with CP for more than 60 years, the pathophysiology remains controversial. Most studies and most surgeons support the traditional view, that spastic muscle imbalance is the primary driver of hip displacement in children with CP [14]. Despite convergence towards this intuitive explanation, other theories have been

suggested, particularly with respect to the etiology of proximal femoral dysplasia. These theories, though neither based in experimental work nor large study cohorts, anticipate more recent thinking, particularly on the development of a horizontal alignment of the proximal femoral physis, leading to coxa valga and hip displacement. Coxa valga can be defined as a deformity of the proximal femur for which the angle between the femoral neck axis and the femoral shaft is greater than the normal range. In typically developing children, the femoral neck-shaft angle (NSA) decreases from 135° to 140° at birth to approximately 125° at skeletal maturity [24]. Valgus of the proximal femur can be noted in different locations including the epiphysis, the physis, and the metaphysis. Lateral epiphyseal tilt is the probable source with remodeling and growth moving the deformity distally through the proximal femur, adversely affecting the neck shaft angle. In his classic monograph on congenital hip dysplasia, Tönnis emphasized the role of lateral physeal tilt in the development of coxa valga, noting that the physis tended to realign itself at right angles to the compressive forces across the hip joint [28]. In cerebral palsy, the changes in these forces, secondary to a lack of weight bearing and abductor insufficiency, are the likely determinants of this abnormal lateral physeal realignment and progressive coxa valga.

472

“A lack of weightbearing and abductor insufficiency are the likely causes of persistent anteversion and coxa valga” Realization of the contribution of lateral physeal tilt to the development of coxa valga, and concomitantly, hip displacement, has led several authors to apply guided growth techniques to the proximal femur, with some early successes (see section Operative Management) [29–31]. Increased coxa valga and lateral tilt of the proximal physis may contribute to the development of acetabular dysplasia in the CP hip, through progressive lateral displacement of the femoral head and a shift in forces to the lateral aspect of the socket [32]. As the proximal femoral physis reorients to a more lateral inclination, increased pressure from a more lateralized femoral head results in a progressively dysplastic acetabulum, effectively decreasing support for the head by the acetabular ‘roof’, allowing it to dislocate. This pathophysiologic theory behind the development of coxa valga and subsequent hip displacement been supported by several previous authors [22, 33, 34]. In one of the earliest papers focused on hip displacement in CP, Phelps himself observed that these hips were normal at birth “…with subsequent development of frank dislocation… [and] it would appear that CP is in some way the cause of this dislocation”. According to Wolff’s Law, he deduced that, due to delayed weightbearing and a lack of normal joint forces, “… some abnormalities in the formation of the femoral neck and acetabulum might be anticipated” [33]. Of the “abnormalities” on the femoral side, coxa valga was suggested to impart a shorter lever arm, leading to relative hip abductor muscle inefficiency. This resulting abductor insufficiency and muscle imbalance was thus purported to be the cause of hip displacement rather than the actions of spastic hip adductor and flexor muscles on their own. In fact, Phelps stated that hip dislocation due to adductor contracture was a different mechanism altogether; one that could occur even in the absence of significant bony deformities. This insight is consistent with our current thinking. Delayed or absent ambulation

J. J. Howard et al.

and persistently reduced activity in hip abductor function, may be the primary drivers of abnormal proximal femoral geometry and hip displacement [12]. Tachdjian supported the theory that “dynamic imbalance” between agonist-antagonist muscle groups—adductor/abductor and internal rotator/ external rotator—could lead to bony deformity and subsequent subluxation [34]. In typically developing children, the resultant joint reaction force, from the gluteus medius and minimus, is perpendicular to the proximal femoral physis, an orientation that is adjusted during growth to maintain this relationship. In the CP hip, Tachdjian suggested that abductor weakness caused a lurch, which shifted the line of pull more vertically, the change in direction of force prompting a physeal reorientation into a valgus position, which put the hip at risk of dislocation. The extent of proximal femoral deformity was stated to be proportional to the loss of abductor muscle power, with more severely involved children displaying an increased risk of coxa valga and subsequent hip displacement. The observations made in the preceding historical studies were remarkably consistent with a recent population-based study of proximal femoral geometry in CP [22]. In this study, the severity of coxa valga (as in the case of femoral anteversion) was associated with increasing neurologic involvement and GMFCS level. The mean NSA in this cohort was 147.5° with progressive increases by GMFCS from levels I to V, respectively: 135.9°, 141.0°, 148.8°, 155°, 163° (Fig.  18.3). The increase in coxa valga with decreasing potential for independent ambulation, irrespective of motor type, lends support to the hypothesis that proximal femoral deformity is acquired secondary to an absence of applied normal joint forces even without the presence of spastic muscle contractures. In the study by Robin and colleagues, the hip Migration Percentage (MP) increased proportionally with severity of proximal femoral deformity. They suggested that increases in both FNA and NSA are significant contributors to the development of hip subluxation. This correlation and suggested causation has also been supported by other authors [24, 25, 32].

18  The Hip in Cerebral Palsy

In addition to the local deformities of the femoral neck as described above, dysplasia of the femoral head is also commonly associated with hip displacement, especially for older children with long standing or severe subluxation or dislocation [35]. Most commonly, the femoral head undergoes a concave, lateral epiphyseal flattening forming a sharp peak at its apex (Fig. 18.4). This characteristic femoral head deformity has been attributed to pressure from the hip capsule, ligamentum teres, periarticular musculature, and/ or acetabulum [35, 36]. Abel and colleagues theorized that lateral pressure from the hip capsule and gluteal muscles was responsible for the most commonly seen ‘caput valgum’ deformity, with wedge shaped defects in the anterolateral head caused by impingement on the posterior acetabular rim. The Bern group investigated this further in an in vivo study, finding consistently that constant pressure from the gluteus minimus and the associated underlying hip joint capsule created a groove in the femoral head, the location of which was dependent on the extent and direction of hip displacement [36]. Early identification of this epiphyseal deformity is paramount. The presence of severe femoral head deformity, especially when associated with

Fig. 18.4  Fifteen year-old female with spastic quadriplegia (GMFCS V) immediately prior to femoral head resection and valgus proximal femoral osteotomy for intractable pain. The characteristic femoral head dysplasia is apparent, displaying concave, lateral epiphyseal flattening that forms a sharp peak at its apex. Note the severe cartilage degeneration in this area. (Courtesy Jason Howard)

473

loss of hyaline articular cartilage, may preclude the ability to reconstruct the hip by bony surgery. Salvage surgery may then be the only option (e.g. McHale femoral head resection plus proximal femoral valgus osteotomy). These procedures have inferior clinical outcomes to reconstructive procedures [37]. Radiograph-based hip surveillance programs can help to identify the presence of hip displacement before these changes occur (see section “Epidemiology”) [38, 39].

Acetabular Dysplasia Classically, the location of acetabular dysplasia in CP hips has been described as being posterosuperior, consistent with the most common direction of femoral head migration [15, 40]. The dysplasia tends to manifest as a trough or a dislocation channel—a path by which the femoral head exits the acetabulum—established by joint forces up to six times that seen in typically developing children [41, 42] (Fig.  18.5). Radiographically, CP hip dysplasia manifests itself as an increase in acetabular inclination (as measured by the acetabular index (AI)). The acetabular dysplasia may

Fig. 18.5  3D CT scan reconstruction showing posterosuperior acetabular dysplasia comprised of a trough-like channel in the direction of femoral head migration. (Courtesy Jason Howard)

474

be overestimated because the lateral acetabular cartilage is not visualized unless an arthrogram or MRI is performed. The cartilage anlage exhibits delayed maturation and may remain un-ossified, as a result of the eccentric forces imparted by a subluxating femoral head [43] (Fig.  18.6). If these forces are normalized at an early age, prior to the development of severe acetabular deformity, there is potential for the acetabulum to assume a more normal development [41]. The location of acetabular dysplasia may be difficult to ascertain on two-dimensional imaging. Several studies utilizing 3D CT found that deficiencies are also seen anteriorly, superiorly, and globally depending on both the direction of femoral head migration and whether or not the hip is dislocated [24, 41, 44–46]. In a large case control study investigating the development and location of acetabular deficiency according to GMFCS level, the authors found that CT based indices representing anterior, posterior, and globally located dysplasia, respectively, were increased for all GMFCS levels as compared to controls [45]. In this cohort, with a mean age of 9 years 5 months, all indices increased with successive GMFCS levels and displayed a strong linear relationship with the extent of femoral

Fig. 18.6  Arthrogram post varus derotational osteotomy for hip displacement in a 5  year-old child with spastic quadriplegia (GMFCS level V). Note the extent of acetabular coverage by the non-ossified lateral cartilaginous anlage outlined by the radiopaque dye. (Courtesy Jason Howard)

J. J. Howard et al.

head subluxation. For hips that progressed to dislocation, the acetabular dysplasia changed from being primarily posterior to more global in nature (i.e. shallow); likely a consequence of decreased molding pressure from the femoral head. “The location of acetabular deficiency is most often posterior, increasing in severity proportional to GMFCS level”. Given the variability in the location of acetabular deficiency in this study, the authors suggested the use of a pre-operative 3D CT scan to more accurately plan surgical reconstruction. Gose and colleagues studied CT-based acetabular morphology in a younger group of 75 children with CP (mean age 5.4 years) and found the location of deficiency to be posterior in 57% of children’s hips with anterior and global deficiencies in 35% and 8% of hips, respectively [24]. As in the preceding study, they determined that the extent of dysplasia was more severe in GMFCS levels IV and V and with increasing migration percentage (MP). A subject of some debate has been whether or not acetabular volume is decreased in children with CP [35, 40]. This has implications for the type of acetabular procedure chosen when reconstructing the CP hip, given that a socket with more normal volume would be better indicated for a periacetabular osteotomy that was more re-­directional in nature (e.g. triple pelvic osteotomy) rather than one that hinged on the triradiate cartilage (e.g. San Diego or Dega-style acetabuloplasty). In a case control study from Korea, the authors found that a decrease in acetabular volume, directly correlated to the extent of femoral head migration [44]. Specifically, a steady decrease in acetabular volume occurred as the hip progressed from no displacement, to subluxation, and finally, to dislocation at 14.2  mL, 5.2 mL, and 3.4 mL, respectively. As a result, the authors suggested the use of acetabular osteotomies that increase the volume of the socket rather than redirect it when treating these children. “Contrary to popular belief, acetabuloplasty procedures designed to treat the capacious acetabulum are not ‘volume reducing’ but quite the opposite”.

18  The Hip in Cerebral Palsy

In a prospective study for children with CP and mean age 8.1 years, this same research group further investigated the influence of a modified Dega osteotomy (i.e. one that hinges on the triradiate cartilage) on the change in acetabular volume post-operatively [47]. Although they found that all measures of dysplasia improved and volume was increased by 68% overall, the final volume achieved was still only 39% of that seen in controls. Given that a ‘Dega-style’ acetabuloplasty only affects a change in volume superior to the triradiate cartilage, there is likely a significant contribution from the inferior half of the socket that is not addressed by current surgical techniques. Unlike the case of the proximal femur, the subject of acetabular version has not been well studied. Like volume, its identification has some importance in determining the best surgical strategy for reconstruction since reshaping procedures like the commonly used Dega acetabuloplasty, for example, have little impact on acetabular version but only on volume and inclination [47, 48]. In one case controlled study, acetabular version was found to be unchanged in children with CP as compared to typically developing children [41]. In both patient groups, mean age of 9 years, the acetabulae were slightly anteverted regardless of the severity of associated acetabular deficiency. This result has been corroborated by other authors [49].

Muscle Morphology Despite the recent shift in thinking away from spastic muscle contracture as the primary cause of hip displacement, there is little doubt that the presence of hip adduction and flexion contractures have a role to play in children with hypertonia. Whether this role is due to the presence of hip adductor spasticity alone or to associated weakness in hip abductors is a matter of debate. Unfortunately, despite an abundance of medical research regarding the clinical and epidemiological aspects of CP, a limited amount of scientific research exists regarding the intrinsic properties of spastic muscle and even less that focuses on the muscles associated with hip displacement. Spastic skeletal muscle is stiffer than normal muscle but there is little agreement on the mechanisms behind increased stiffness. Recent

475

work with diseased cardiac muscle has shown that changes in passive stiffness of myocytes are accompanied by changes in titin isoforms [50]. The molecular spring titin maintains the structural arrangement of filaments in the sarcomere and increases passive stiffness to varying degrees depending on isoform expression. Another study involving biopsies of hamstring muscles in children with cerebral palsy has shown that increased stiffness from within the muscle fiber was associated with an increase in collagen content in the surrounding extra-cellular matrix. This alteration was hypothesized to be in response to the increased stress imparted to spastic muscle in CP. This same study also confirmed an increase in sarcomere lengths as compared to normal children, also thought to be adaptive [51]. In one of the few studies specifically investigating the intrinsic properties of periarticular muscles around the hip, Kaiser-Larkin and colleagues performed gracilis muscle biopsies from children with spastic quadriplegia (GMFCS levels III to V) undergoing soft tissue releases for hip displacement [52]. In this study, as in previous work investigating other muscles in CP, the authors found that sarcomere lengths were significantly longer than typically developing controls and were decreased in number. Furthermore, sarcomere length was found to be significantly correlated to the severity of hip displacement as measured by migration percentage. At the increased lengths identified, force generation by the over-lengthened sarcomeres would be substantially decreased, leading to an overall weakness of the muscle [53]. This weakness, in turn, may contribute to the muscle imbalance and walking impairment that has been purported to lead to the development of proximal femoral deformities rather than from a spasticity-­generated force that pulls the hip out of joint. Analyzing biopsies of adductor longus and gracilis muscles, recent work has found that, at matched sarcomere lengths, myofibrils were more elastic (i.e. decreased stiffness) in children with CP. It was suggested that a more compliant isoform of titin may be responsible for this increased elasticity [54]. Synthesis of a smaller, more compliant, isoform of titin could represent an adaptive

J. J. Howard et al.

476

mechanism, which prevents muscle damage associated with sarcomeres that are stretched to near their limit. Despite this elasticity, CP myofibrils were found to be under significantly increased tension at in vivo sarcomere lengths as compared to typically developmental children. A further study by the same research group confirmed that titin isoforms from muscles in CP were significantly correlated to the severity of hip displacement [52]. Despite promising results from this early research, much work remains to be done to elucidate the pathophysiology of muscle in CP, in particular with respect to its role in the development of hip displacement.

Osteoarthritis and Pain The primary reason to treat the subluxated or dislocated hip is to prevent the onset of painful osteoarthritis, which is found in up to 90% of children with hip displacement [55, 56]. Children with painful hips have an inflammatory reaction, degeneration of articular cartilage and a cascade of inflammatory mediators that increase the sensitivity of the hip joint to mechanical stimuli [57]. The question remains as to the cause of this cartilage degeneration. The most commonly offered explanation is the increase in joint reaction force imparted by spastic muscles with the resulting pressure leading to cartilage damage [42, 58]. Given the previous discussion regarding the finding that the prevalence of hip displacement is not related to motor type, the search for a unifying theory of causation, is important. In a recent study utilizing finite element analysis techniques, the growth and preservation of articular cartilage in general was found to be indirectly regulated by the mechanical stresses imparted to it which, in turn, generate a hydrostatic pressure within the cartilage itself. This hydrostatic pressure promotes cartilage growth by increasing extracellular matrix production while at the same time inhibiting cell hypertrophy; the result being thicker articular cartilage in areas of high stress [9]. In other words, the hydrostatic pressure induced through normal joint loading is chondroprotective. The converse, however, has also been found to be true, with decreased or abnormal joint forces leading to a

decrease in chondral hydrostatic pressure. This decrease stimulates an increase in subchondral endochondral ossification and a relative decrease in cartilage thickness, making it more susceptible to mechanical damage [59]. Given the abnormal joint forces and lack of cyclic loading associated with children with CP, this process is accentuated, leading to severe articular cartilage degeneration and subsequent painful inflammation. Thus, it is the unloaded, rather than the excessively loaded, hip which is most at risk for degenerative arthritis in CP, though mechanical abutment of the superolateral femoral head on the lateral acetabular rim has also been implicated [9, 60]. Given the absence of nociceptors and afferent nerve fibers within articular cartilage, other periarticular tissues must be responsible for the high frequency of pain associated with hip displacement in CP. In children with CP, undergoing open reduction for the treatment of hip dislocation, biopsies of the ligamentum teres and hip joint capsule were analyzed for the presence of substance P and S-100 protein—two neuropeptides that enhance the sensitivity of nociceptors to pain. The investigators found a significantly increased density of nerve fibers containing these substances in children who reported pain and in those who had cartilage degeneration affecting more than 25% of the surface of the femoral head. The authors concluded that the presence of substance P and S100 caused nerve fibers that are normally “mechano-insensitive” to become responsive to mechanical stimulation of the hip joint, the consequence of which was increased pain sensitization more centrally as well as peripherally. The authors further confirmed the role of these pain mediators, finding increased levels in CP as compared to patients with hip subluxation secondary to DDH [7].

Natural History  atural History of Untreated Hip N Displacement The natural history of untreated hip displacement in children with CP is critical to clinical decision-­ making [61]. Some studies have not found a sig-

18  The Hip in Cerebral Palsy

nificant correlation between hip displacement and pain, prompting a “reactive” approach to management, whereby surgery was offered only after the hip became painful or caused functional limitations [62, 63]. The majority of existing natural history studies are short-term and are compromised by selection bias. There is substantial evidence to suggest that the “reactive” approach has led to inferior outcomes than the “prophylactic” approach, in which children have routine surveillance and early preventative and reconstructive procedures, when the hips are asymptomatic [56]. It is now widely accepted that the development and progression of hip displacement puts the child with CP at risk for the development of pain, seating imbalance, difficulties with perineal hygiene, decubitus ulceration, gait abnormalities, and windswept hip deformity [64, 65]. These factors have a substantial impact on quality of life, which can be improved by surgical reconstruction [66, 67].

Are CP Hips Painful? The most important reason to treat CP hip displacement is to prevent the onset of debilitating pain. Pain is a common manifestation in children with CP with hip displacement being the most frequent etiology [68]. As discussed in the preceding section, increased nociceptive inputs have been identified in the CP hip. The extent of cartilage degeneration has been shown to be related to the severity of subluxation [9, 57]. Hip pain may impair activities and participation as well as health related quality of life (HRQoL) including comfortable seating and ease of perineal hygiene [61, 68]. Given the unsatisfactory outcomes associated with salvage procedures undertaken after the hip becomes painful, the identification and treatment of hip displacement before the onset of pain would seem warranted and is supported by recent evidence [37, 68]. Given the long history of surgical management for these children, it is surprising that the link between pain and hip displacement in CP has only become established within the last few years. Recent studies have confirmed that the

477

incidence of hip displacement in CP is correlated with the severity of neurologic impairment and that children experience pain with an intensity and frequency related to GMFCS level, the extent of hip subluxation/dislocation, and worsening hip morphology [13, 39, 56, 69]. “Painful hips in CP are common and proportional to the extent of hip displacement and GMFCS level”. In one population-based study from Norway, the only independent risk factor associated with the development of hip pain was a MP greater than or equal to 50% [55]. These findings were corroborated in another population-based study that followed a birth cohort of children with CP until skeletal maturity (mean age at final follow­up, 18  years). This study found that the prevalence and frequency of pain was directly related to increasing GMFCS level and to worsening hip morphology according to the Melbourne Cerebral Palsy Hip Classification Scale (MCPHCS) (see section “Imaging”; Fig.  18.14). Specifically, severe pain was demonstrated for MCPHCS levels 5–7 (i.e. severe subluxation, dislocation, salvage hip, respectively) with less pain in the lower levels. Thus, as with others, the authors supported the institution of radiographic surveillance programs to identify and treat hip displacement before the morphology deteriorates substantially and surgical outcomes become less predictable [55, 56, 68].

 ealth-Related Quality of Life H (HRQOL) Although pain is one of the most significant factors to consider, a more comprehensive measure of overall health and function is needed to more completely evaluate the impact of hip displacement in a child with CP and in the family/care givers [70]. In essence, this tool would measure health-related quality of life (HRQOL), defined as the “health related aspects of life that contribute to the perception of well-being, and the ability to fulfil and take part in certain life roles” [67]. For children with CP, improved HRQOL has been purported to be linked to the presence

J. J. Howard et al.

478

of a ‘healthy hip’; one defined as being “free from pain, mobile, and with normal or near normal morphology”; conferring ease of sitting and transfers, personal hygiene and dressing, increased activity participation and community engagement, in addition to a reduction in pain [56]. For children most commonly affected by hip displacement, non ambulant children, GMFCS IV and V, improvement in these factors should lead to an improvement in quality of life, the overall goal in the treatment of these patients. Until recently, there were no validated outcome tools available to measure HRQOL in children with severe CP. Given the inherent surgical risks associated in this vulnerable patient population, the use of valid outcome measures are critical to determine the impact of hip surgery as it relates to HRQOL [66]. The development of the Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD™) questionnaire by Narayanan and colleagues has provided a valid, disease-specific outcome measure to apply to patients with CP [5]. This tool covers several domains related to HRQOL including personal care and activities of daily living, positioning, transferring and mobility, comfort and emotions, communication and social interaction, health, and overall quality of life. To answer the question of whether hip displacement affects HRQOL, a study from Germany used the CPCHILD™ in a cohort of children with CP (GMFCS levels II to V). The study reported a significant decrease in overall HRQOL with increasing MP [67]. Decreased scores were reported in communication and social interaction and overall health status domains. These findings may be related to increased levels of hospitalization, analgesic intake, increased need for caregiving, and impairments in mobility. Overall QOL domain was not affected by hip displacement, however, reflecting the many non-musculoskeletal factors, in more involved children. A subsequent study, again using the CPCHILD™ questionnaire, corroborated the relationship of hip displacement to HRQOL but only in the Comfort and Emotions domain. GMFCS level V status was an independent risk factor across all domains [70].

Given the confirmation of the impact of hip displacement on HRQOL, the question remained as to whether surgical intervention resulted in improvements in CPCHILD™ scores. The key questions are: (1) does surgery have an impact on HRQOL, and (2) is the CPCHILD™ responsive to change? In a prospective study by Difazio and colleagues, significant improvements in CPCHILD™ scores were reported, from increases in the personal care/activities of daily living and positioning, transfers, and mobility domains [66]. A second study by the same group, reported that children with MP less than 50% had better postoperative CPCHILD™ scores than those with MP greater than 50% [10, 71]. These differences may be related to both hip morphology and disease severity [56]. “Hip surgery improves quality of life in CP, particularly for patients with less severe displacement”. With the results of the preceding studies in mind, it can be hypothesized that orthopedic surgery may have a focal impact on physical domains rather than on the more global developmental aspects that contribute to overall HRQOL. This would be consistent with the pre-­ operative effects of hip displacement on HRQOL [66, 70].

Epidemiology I ncidence of Hip Displacement According to Disease Severity Identifying ‘hips-at-risk’ has been advocated to permit early intervention and mitigate the poor natural history associated with hip displacement [38]. More aggressive surgical strategies—with a goal to prevent or reconstruct the associated proximal femoral and acetabular dysplasia— are now commonly implemented but need to be applied before the femoral and acetabular dysplasia becomes too severe [15]. As such, an accurate picture of the epidemiology of hip displacement is paramount, using risk factors determined through population-based studies

18  The Hip in Cerebral Palsy

to set up surveillance programs to identify children at an early stage. Previous studies have suggested disease severity as the most important risk factor associated with the hip displacement [21, 72]. Specifically, children with quadriplegia have been shown to have the highest risk of developing hip displacement while those with less severe neurologic impairment have a lower risk. Given the results of a recent population-based cohort study, the spastic motor type was most commonly seen (86%) with one-third having quadriplegia. With a birth prevalence rate of 2 per 1000, the number of patients at risk for hip displacement can be expected to be correspondingly high [17, 73]. Case series are not suitable to determine the prevalence of hip displacement in children with CP because the population denominator is not known. In addition, measures of disease severity have recently become more sophisticated. Utilizing reliable, valid classifications of gross motor function which stratify levels of neurologic severity, rather than motor type and topography (e.g. spastic quadriplegia), have helped to make the picture that much clearer. In an attempt to address the problem of classifying motor function and degree of neurologic impairment in CP, the Gross Motor Function Classification System (GMFCS) was established by Palisano and colleagues [20]. It is a five-level ordinal grading system based on the assessment of self-initiated movement with emphasis on function during sitting and walking. Distinctions between levels are based on functional limitations, the need for walking aids or wheeled mobility, and quality of movement. This classification is illustrated and further described in Fig. 18.7. The impairment in gross motor function increases with each successive GMFCS level. A child at GMFCS level I has near normal gross motor function, few limitations and usually has good general health. At GMFCS level V children have major limitations in gross motor function, require wheelchairs for mobility, have poor head and trunk control and a high frequency of medical co-morbidities such as epilepsy, respiratory disease and nutritional impairments. The GMFCS is a valid, reliable, stable, and clinically

479

relevant method for the classification and prediction of motor function in children with cerebral palsy and has become the standard by which these patients should be classified [74]. In the first population-based study in this area, Soo and colleagues confirmed the link between GMFCS level and hip displacement. They reported a near linear increase in hip displacement and GMFCS level, such that children at GMFCS level I had a 0% risk of hip displacement (MP > 30%) while those in level V had a 90% risk [13] (Fig.  18.8). In this same study, the risk of hip displacement also increased with limb involvement, with a 1% rate in children with spastic hemiplegia (mostly GMFCS level I) and an 82% risk for children with spastic quadriplegia (mostly GMFCS IV and V). In addition, with the exception of ataxia, the incidence of hip displacement was unrelated to motor type, with children with hypertonia (i.e. spastic, dystonic, mixed) having the same risk as children with hypotonia. These findings were corroborated by subsequent population-based studies and are useful as a basis for comprehensive hip surveillance programs [75, 76].

The Case for Hip Surveillance Given that early identification permits early treatment of hip displacement with improved outcomes, hip surveillance programs are increasingly utilized to realize these goals. Previous studies have confirmed that a well-run hip surveillance program can reduce the incidence of late presenting dislocation by identifying the need for early surgery [77]. These programs have subsequently seen a reduction in the number of salvage procedures performed in favour of preventative and reconstructive procedures [38]. The cornerstone of a hip surveillance program centers on radiographic assessment with the MP being the most commonly accepted primary measure. The MP threshold to identify “hips-at-risk” varies from 30% to 33% [78]; see section “Imaging”). Indications for referral to hip surveillance program are based on GMFCS level, the extent of topographic involvement and

J. J. Howard et al.

480 GMFCS E & R between 6th and 12th birthday: Descriptors and illustrations

GMFCS Level I Children walk at home, school, outdoors and in the community. They can climb stairs withot the use of a railing. Children perform gross motor skills such as running and jumping, but speed, balance and coordination are limited.

GMFCS Level II Children walk in most settings and climb stairs holding onto a railing. They may experience difficulty walking long distances and balancing on uneven terrain, inclines, in crowded areas or confined spaces. Children may walk with physical assistance, a handheld mobility device or used wheeled mobility over long distances. Children have only minimal ability to perform gross motor skills such as running and jumping.

GMFCS Level III Children walk using a hand-held mobility device in most indoor settings. They may climb stairs holding onto a railing with supervision or assistance. Children and may self-propel for shorter distances.

GMFCS Level IV Children use methods of mobility that require physical assistance or powered mobility in most settings. They may walk for short distances at home with physical assistance or use powered mobility or a body support walker when positioned. At school, outdoors and in the community children are transported in a manual wheelchair or use powered mobility.

GMFCS Level V Children are transported in a manual wheelchair in all settings. Children are limited in their ability to maintain antigravity head and trunk postures and to control leg and arm movements.

GMFCS descriptors: Palisano et al. (1997) Dev Med Child Neurol 39:214-33 CanChild: www.canchild.ca

Fig. 18.7 The Gross Motor Function Classification System (GMFCS) for children with cerebral palsy between 6 and 12  years old (Expanded and Revised).

Illustrations Version 2 © Bill Reid, Kate Willoughby, Adrienne Harvey and Kerr Graham, The Royal Children’s Hospital Melbourne ERC151050

(Used with permission, Copyright © Kerr Graham, Bill Reid, and Adrienne Harvey, The Royal Childrens’ Hospital Melbourne)

INCIDENCE (%)

18  The Hip in Cerebral Palsy 100 90 80 70 60 50 40 30 20 10 0

I

II

481

III

IV

V

GMFCS Level

Fig. 18.8  Incidence of hip displacement (MP  >  30%) according to the GMFCS (MP migration percentage, GMFCS gross motor function classification system). (Adapted from Soo et al. [13])

ambulatory status [16, 79]. Reduced hip abduction on clinical examination can also suggest a “hip at-­risk” but clinical examination alone has not been shown to be reliable. The Australian Hip Surveillance guidelines have been reported and have been used as a template for the development of guidelines in other jurisdictions and by a range of professional bodies, including the American Academy of Cerebral Palsy and Developmental Medicine (AACPDM) and the Pediatric Orthopaedic Society of North America (POSNA) [80] (Fig. 18.9). “The risk of hip displacement increases linearly with successive GMFCS levels and should be detected with a standardized radiographic surveillance program”.

Clinical Presentation The clinical evaluation of a child with suspected hip involvement in the context of CP is done as part of a regular hip surveillance program, where such a program exists. Alternatively, it may be part of a clinician’s assessment of such a child during routine consultation. In either scenario, it should be comprehensive enough to identify all aspects of the manifestations of hip displacement in CP. Through history and physical examination, the most important pieces of information to acquire relate to the presentation of hip displacement and

its impact on the child’s function and quality of life. These include the child’s age, GMFCS level, topographical limb involvement, motor type and underlying muscle tone, presence of absence of static muscle contractures about the hip and issues relating to functional seating (including the presence of pelvic obliquity and scoliosis). Combined with radiographic measures and, for operative candidates, an examination under anesthesia, these aspects represent the ‘decision matrix’ for hip disease for children with CP (Fig. 18.10). The history allows clarification of the developmental status, i.e. the GMFCS level, in relation to age, as disease severity is one of the most important determinants of hip displacement (see section “Epidemiology”). Issues relating to pain, difficulties with hygiene, functional limitations, and changes to sitting, standing and walking ability must be identified, as they are often related to hip disease and progression of hip pathology. The topographical pattern of limb involvement (e.g. quadriplegia), and the motor type (e.g. spastic, dystonic or mixed etc.) is determined, as part of the clinical picture. Most children with CP will display at spastic motor type, typified by a ­velocity dependent increase in muscle stiffness. Information regarding co-morbidities such as epilepsy, respiratory status, GI problems, previous interventions and surgery, can be obtained from the patient, family, therapists, other care givers and specialists involved in the child’s care. This is crucial to determine the child’s fitness for surgery and to predict the risk of peri-operative complications, should surgery be indicated. The clinical examination follows a set format. For ambulatory patients, the gait should be observed and the need for assistive devices ­determined. Other determinants of gait deviation, including leg length discrepancy (LLD), ‘scissoring’ posture, abductor lurch, pelvic tilt and spinal asymmetry, should also be elucidated. For children able to stand, an assessment of functional LLD should be performed using blocks; often difficult in the presence of static lower limb contractures. For the non ambulatory patient, assessment of LLD by Galeazzi sign may be discrepant, suggesting a unilateral hip dislocation. Coronal plane assessment in the erect position may reveal excessive anterior pelvic tilt;

J. J. Howard et al.

482 Fig. 18.9 Australian consensus guidelines for hip surveillance in CP. (Reprinted with permission from Shore et al. [79])

GMFCS I • • • GMFCS I

GMFCS II

GMFCS II • Initial AP Pelvis at 12-24 months (or at age of identification) • Repeat clinical & radiographic review via yearly surveillance until MP stability If MP unstable/abnormal - continue yearly surveillance If MP stability established - review at 4-5 years • Repeat clinical & radiographic review at 4-5 years† If GMFCS level has changed - alter surveillance accordingly If MP stable - review at 8-10 years If MP unstable - continue yearly surveillance until stability • Repeat clinical & radiographic review at 8-10 years† If MP stable - discharge GMFCS III • • •

GMFCS III

Initial AP Pelvis at 12-14 months (or at age of identification) Repeat clinical and radiographic review at 3 years‡ If GMFCS level has changed - alter surveillance accordingly Repeat clinical & radiographic review at 5 years† If GMFCS level has changed - alter surveillance accordingly If remains GMFCS level 1 - discharge from hip surveillance

•

Initial AP Pelvis at 12-24 months (or at age of identification) Repeat clinical & radiographic review 6 months later† If MP unstable/abnormal - continue yearly surveillance If MP stability established - review at 4-5 years Repeat clinical & radiographic review at 4-5 years† If GMFCS level has changed - alter surveillance accordingly If MP stable - review at 8-10 years If MP unstable - continue yearly surveillance until stability Repeat clinical & radiographic review at 8-10 years† If MP stable - discharge If MP unstable - continue yearly surveillance until stability established

GMFCS IV • •

GMFCS IV

• •

Initial AP Pelvis at 12-24 months (or at age of identification) Repeat clinical & radiographic review every 6 months† If GMFCS level has changed - alter surveillance accordingly If MP unstable/abnormal - continue 6 month surveillance until stability If MP stability established - continue yearly surveillance Repeat clinical & radiographic review every 7 years† If MP stable ( 40% and significant bony deformities have been established, the success rate of soft tissue surgeries alone dramatically decreases and concomitant bony procedures are indicated (Fig. 18.15) [14, 15].

Hip Surveillance According to Clinical Guidelines MP Progression 100 90 80 70 60 50 40 30 20 10 0

MP Stability

MP Progression

MP Stability

MP Progression

Continue to Skeletal Maturity

(b) (a) (c)

1.3

1.8 2.3 3.0 3.5 Muscle Lengthening

4.3

4.8 5.2 5.5 6.0 Bony Reconstruction

6.6

7.7

8.1 9.2 10.1 12.3 13.3 14.3 Spine Surveillance

Management Pathway

Fig. 18.15  Surveillance and management of hip displacement in cerebral palsy (GMFCS IV and V). Serial MPs are much more useful than single measurements, especially if they are graphed as in this example of a child, GMFCS V, from age 1 year to age 14 years. The red line is left hip and the blue line the right hip. Note the changes

with adductor lengthening marked (a) and VDROs marked (b). Our target is to get the MP for both hips under 30% and keep them in that zone. (c) spine surveillance should continue till skeletal maturity. (MP migration percentage, GMFCS gross motor function classification level, VDRO varus derotational osteotomy)

492

“Once the MP > 40% and significant bony deformities are present, the success rate of adductor releases alone dramatically decreases”. Despite this, the borderline between the indications for preventive, reconstructive and salvage surgery remain ill-defined. Some groups have described good outcomes from reconstructive surgery in older children and teenagers despite the presence of significant femoral head deformity and loss of articular cartilage [106] However, in the literature, it is generally agreed that the likelihood of achieving good outcomes (i.e. pain free hips with good motion) decreases as the migration percentage increases and as the degree of femoral head and acetabular deformity increases [10, 56, 71]. Given the unpredictability and high failure rate of salvage surgery, every effort should be made to manage CP hip displacement in by a combination of preventive and reconstructive options [37]. “Given the high failure rate of salvage surgery, hip displacement should be treated early by a combination of preventive and reconstructive options”. Given the widespread adoption of hip surveillance guidelines, combined with the high failure rate of early adductor surgery in children at GMFCS IV and V [16, 79], the authors suggest that the term ‘preventive surgery’ should not be restricted to soft tissue surgery alone as previously suggested by Dr. Mercer Rang but should also include bilateral femoral varus derotation osteotomies (VDROs), previously classified as ‘reconstructive’ procedures. It has to be recognised that when Dr. Rang first promoted hip surveillance at the Hospital for Sick Children in Toronto in the 1970s, it was envisaged that a single, one time operative intervention (i.e. release of the hip adductors) would be sufficient to prevent hip dislocation in the majority of children. Long term follow-up studies have confirmed that this is not the case and for this reason we suggest that both adductor releases and bilateral VDROs should be considered as preventive strategies for non-ambulant children with severe CP [16, 79].

J. J. Howard et al.

This also recognises that there is a clear overlap between bony surgery to prevent hip dislocation and bony surgery for reconstruction after hip dislocation. The former requires only bilateral VDROs. The latter requires bilateral VDROs combined with pelvic osteotomies. “For severe CP, VDROs alone may be required to prevent hip dislocation and should (usually) be combined with pelvic osteotomies for reconstruction after hip dislocation”.

 reventative Procedures: Hip P Adductor Releases Contracture of the hip adductors is ubiquitous in children with CP seen in all levels of disease severity, spanning GMFCS levels I to V.  For ambulant children (GMFCS II and III), soft tissue surgery is designed to prevent hip displacement, reduce scissoring and improve gait function. In these cases, adductor releases are very successful [16, 79]. Performing surgery to reduce and reconstruct a dislocated hip in these children is a rare occurrence, unlike those on the more severe end of the disease spectrum. The exception to this general rule are a small group of children with Type IV Hemiplegia, according to the classification proposed by Winter et al. [107]. Despite functioning at GMFCS II, the affected hip in Type IV hemiplegia frequently requires soft tissue surgery and bony surgery. Fortunately, a proximal femoral osteotomy usually improves gait function as well as hip stability [108]. By contrast, contractures of the hip adductors in non-ambulant children are invariably accompanied by marked abnormalities in the proximal femoral geometry with a combination of coxa valga and increased femoral head anteversion [14, 22]. In such children, it is not logical to rely on adductor releases as the only measure to prevent hip displacement, correction of the abnormal proximal femoral geometry will also be required. In rare cases, adductor surgery alone can be curative, but this is the exception rather than the rule (Fig. 18.16).

18  The Hip in Cerebral Palsy

493

Fig. 18.16  Hip development and intervention in a child with severe CP, GMFCS V ​and shunted hydrocephalus. The age at each time-point is at the bottom left of each image and the radiographic sequence is numbered top right. Note that hips were enlocated age 11  months (1). Bilateral dislocation by age 2 years and 5 months (2, 3). The patient was too frail for bony surgery and the hips were successfully reduced by bilateral adductor releases (4). However, this was followed by rapid re-dislocation (5,

6). Repeat adductor releases were performed combined with bilateral VDROs (7). At 5-year follow up, the hips are in joint after metal removal (8) with only mild residual dysplasia (9). There is no pain, excellent motion, no restrictions in sitting. Early hip dislocation in children at GMFCS V usually require VDROs, for long term stability. (Used with permission, Copyright Kerr Graham and Abhay Khot)

The indication for soft tissue surgery alone is for the younger child with a confirmed diagnosis of hip displacement, spasticity and contracture of the hip adductors and flexors, a migration percentage of 30–50% and minimal femoral or acetabular deformity. Respiratory health, nutrition and seizure control should be optimized before surgery.

 urgical Technique: Hip Adductor S Lengthening (Fig. 18.17, 1, 2) The procedure is performed under a general anaesthesia, combined with a caudal block. If the child has severe hypertonia, epidural analgesia can be considered to avoid a stormy post-­ operative course. The child is placed supine on the operating table with the pelvis elevated under a rolled up sheet or towel. The perineum is excluded with an adhesive plastic strip and the groin area is prepared. The lower limbs are draped free to permit periodic checking of the range of hip abduction

“The indication for soft tissue surgery alone is for the younger child with contractures of the adductors and flexors, an MP between 30%–50%, and minimal bony deformity”.

J. J. Howard et al.

494

1

Skin incision Copyright ©, BR & KG, RCH Melbourne

2

Adductor brevis

Anterior branch of obturator nerve

Adductor longus

Gracilis Copyright ©, BR & KG, RCH Melbourne

Fig. 18.17  Surgical technique: hip adductor releases and hip flexor lengthenings. (1) The skin incision is 2–3  cm long and is made 1 cm distal and parallel to the groin crease, centred over the palpable adductor longus. The operator stands opposite to the surgical side for ease of visualization. (2) With the hip in flexion and abduction, the adductor longus is divided close to its insertion on the pubis. Following this, with the hip in extension and abduction, the gracilis is released close to its origin. Occasionally, a partial myotomy of the adductor brevis is required to achieve symmetric abduction of more than 60° bilaterally. (3) For non-ambulators, the iliopsoas is released completely from the lesser trochanter. The interval between the pectineus anteriorly, and the adductor brevis posteriorly is used to approach the

lesser trochanter. (4) Hohman retractors are inserted anterior and posterior to the femur to allow visualisation of the iliopsoas inserting on the lesser trochanter. A right-angle forceps is passed around the tendon which is subsequently divided completely, using electrocautery. (5A) For ambulators, a psoas contracture should be addressed by a fractional lengthening over the brim of the pelvis rather than complete tenotomy at the lesser trochanter. It is performed through a separate 3–4 cm anterior skin crease incision. It should be noted that the femoral nerve lies on the anterior surface of the iliopsoas and the tendon is posterior (inset). (5B) The tendinous portion of the psoas is subsequently transected, with preservation of the muscle fibres of iliacus. (Used with permission, Copyright © Kerr Graham and Bill Reid)

18  The Hip in Cerebral Palsy

3

495 Adductor longus Pectineus Femoral nerve

Gracilis

IIiopsoas

Adductor brevis Adductor magnus Obturator externus

Sciatic nerve

Copyright ©, BR & KG, RCH Melbourne

4 Iliopsoas

Lesser trochanter Femur

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.17 (continued)

J. J. Howard et al.

496

5A

5B Pos

Iliopsoas tendon Psoas

Psoas

Femoral nerve

Ant

IIiopsoas tendon Tendon division

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.17 (continued)

during surgery. With the surgeon and assistant on opposite sides of the table, the range of hip abduction is checked in both flexion (to check for tightness in adductor longus and brevis) and in extension (to check for contractures of the gracilis and proximal hamstrings). In addition, a Thomas Test (see section “Clinical Presentation”) is performed to check for the presence of a hip flexion contracture. We find it easier to visualise the adductor region by operating on the contralateral hip (i.e. to operate on the right hip adductors, the operating surgeon stands on the left hand side of the operating table). The skin incision needs to be only 2–3  cm long and is made 1 cm distal and parallel to the groin crease, centred over the palpable adductor longus. The deep fascia is incised at 90° to the skin incision; that is, along the line of the adductor longus. If phenol (1  mL, 6% aqueous solution) is considered for application to the anterior branch of the obturator nerve (ABON) for denervation, it is best to identify the nerve in the interval between the adductor longus and brevis, prior

to any muscle division [109]. In ambulant children, we lengthen only the adductor longus and gracilis and rarely phenolize the ABON. In non-ambulant children, we release the adductor longus, the gracilis, the iliopsoas at the lesser trochanter (see below, section “Surgical Technique: Lengthening of the Iliopsoas in Non-­ ambulators”), and occasionally add a partial myotomy of the adductor brevis. The tendons should be released close to their insertion, to minimise muscle division, bleeding, and a dead space where a hematoma may collect and predispose the deep wound infection. The sequence is as follows. With the hip in flexion and abduction, the adductor longus is divided first close to its insertion on the pubis. The hip abduction range is re-checked and in the majority of children, the adductor brevis can be stretched slowly and gently to allow full abduction to approximately 70–80°. Where needed to achieve the necessary abduction, the anterior 50% of the muscle fibres of the adductor brevis may be released to facilitate this stretching. When performing this

18  The Hip in Cerebral Palsy

myotomy, the main branch of the obturator nerve must be protected. If it is inadvertently divided, the risk of a post-operative abduction contracture or a windswept deformity is increased. Next, with the knee extended to facilitate dissection, the origin of the gracilis is mobilised and divided close to its origin. The range of hip abduction in extension is checked and the gracilis origin is palpated to ensure complete transaction since it is frequently observed that some fibres remain undivided in the depths of the incision. The minimum goals of surgery are to obtain symmetric abduction of more than 60°. Following haemostasis, the incision is irrigated with normal saline containing a cephalosporin antibiotic. The deep fascia is closed to reduce dead space. If haemostasis is satisfactory, no drain is required. The skin is closed with an absorbable subcuticular suture and the amount of suture material in the deeper layers should be kept to a minimum to reduce the risk of infection and suture sinuses. We then cover the incision with steristrips and carefully dry the surrounding area. We apply a translucent adhesive dressing, which remains in place for 2–3  weeks after surgery. An adhesive, translucent dressing allows easy identification of bleeding, haematoma, wound inflammation or discharge as well as preventing contamination from urine or faeces. Post-operative Management After adductor lengthening, children often experience significant pain and spasm in the adductor muscles. Their natural reaction is to adduct and flex the hips, bringing the lengthened tendons’ ends closer together and thus setting the stage for the development of scar tissue and secondary contracture. For this reason, we routinely use an A-frame cast with each leg abducted to 30°—to a total of 60° for both hips—for a period of 3  weeks. These casts require meticulous ­application for the child’s comfort, ease of nursing and for the prevention of skin sores. The top and bottom of the casts should be copiously padded. We do not include the ankle in the cast as it seems to increase the risk of hypertonia throughout the limbs and the risk of sores over the heels. We apply copious padding around the malleoli and Achilles’ tendon, stopping the cast material

497

short of the padded area to prevent cast sores. It is not necessary and it is inadvisable to apply the casts in excessive abduction or for periods of longer than 3 weeks. Excessive abduction for prolonged periods may result in abduction contractures of the hip and may predispose children to windswept deformities. “Excessive hip abduction for prolonged periods post adductor release can cause abduction contractures and may predispose children to windswept deformities”. Abduction contractures are much more difficult to deal with than adduction contractures. During the time in the A-frame cast, we encourage three different positioning techniques, in an effort to prevent contractures and to promote good muscle length. These are: 1. Long sitting (i.e. with the hips flexed to 90°, knees extended within the casts), with support if necessary. This provides good stretching of the hamstring muscles. 2. Prone lying for 1 h, twice per day. This promotes stretching of the hip flexors and is more important if the psoas is released. 3. Standing. Supporting the child in the standing position encourages plantar grade feet and helps to prevent the development of ankle equinus contractures during the time in cast (when it is not feasible to use ankle foot orthoses (AFOs)). Following cast removal, a hip abduction brace may be used at night for the first 6–12  months to reduce the risk of recurrent contracture. There is no evidence that this is beneficial and if it is burdensome to the child and family, or disturbing sleep, it should be abandoned. Physical therapy focused on maintaining a good range of motion and symmetrical postures around the hip is much more important. Post-operative Complications With careful technique, post-operative complications should be minimal. The most ­ common problem is prolonged pain and spasm post-­operatively and this needs to be addressed

498

by aggressive use of appropriate analgesics and anti-­spasmodics. A combination of caudal block, morphine (or ketamine) infusion, and regular diazepam is effective for most children. Excessive lengthening of the hip adductors must be avoided, especially in ambulators, as the adductors are important to stance phase hip stability in gait. In non-ambulators, excessive or asymmetric releases should be avoided to reduce the risk of abduction contractures and windswept deformity. Traditionally, abduction casts were often used for 6–8  weeks. By reducing the time in cast to 2–3  weeks, these problems can be reduced and the burden of care decreased. Haematoma and superficial or deep wound infection are uncommon but troublesome to deal with. A significant haematoma requires evacuation, and it may be possible to perform a secondary closure of the incision over a suction drain. Superficial infection related to the ends of sutures is more common and is easily rectified by trimming suture ends and meticulous wound care. Deep infection is almost always secondary to the presence of a haematoma and requires aggressive management by wound debridement, antibiotics, dressings and when appropriate, secondary closure. Pearls and Pitfalls • Avoid excessive adductor lengthening, especially in ambulators, and avoid both excessive abduction in the A-frame cast and casting for more than 3 weeks. • In ambulators, perform an intramuscular lengthening of the psoas, to preserve hip flexion power. In non-ambulators, release the entire iliopsoas from the lesser trochanter. • The role of obturator neurectomy is controversial but it can be useful in some children with severe hypertonia, which would be refractory to any other form of management. However, because of the potential for obturator neurectomy to result in excessive weakening which increases the risk of abduction contracture, we generally prefer phenol neurolysis at the time of surgery. Under direct vision, the anterior branch of the obturator nerve is bathed in

J. J. Howard et al.

1–2  mL of 6% phenol. This results in a 12–24 month period of reduced tone in the hip adductors. We think that this reduces post-­ operative adductor spasms and permits healing with improved abduction range. • Provide aggressive post-operative analgesia and anti-spasmotics to relieve pain and improve the experience of the child and parents for what is often the first surgical intervention in the life of the child with severe CP.

 reventative Procedures: Hip Flexor P Lengthening Hip flexor lengthening requires a different approach in ambulatory children compared to non-ambulators. In ambulators, a contracture of the psoas should be addressed by a fractional lengthening over the brim of the pelvis rather than tenotomy of complete release at the lesser trochanter. “In ambulators, psoas contracture should be addressed by a fractional lengthening at the pelvic brim rather than a complete iliopsoas tenotomy at the lesser trochanter”. The idea behind this difference in approach is rooted in the need to preserve hip flexor strength in ambulators by performing a muscle recession over a complete release, although this concept is disputed by some authors [110]. For non-ambulators, the iliopsoas is released completely from the lesser trochanter; the reduction of the displaced hip taking precedence over the preservation of hip flexor strength, which is not as functionally important for children in GMFCS levels IV and V. Skaggs and colleagues described three tests that should be performed before cutting the tendon in the intramuscular area, to avoid injury to the femoral nerve [111]. The three tests are as follows: 1. Identify that the structure to be divided has muscle fibres leading into the “tendon”. 2. Confirm that the musculotendinous unit tightens with internal rotation of the hip.

18  The Hip in Cerebral Palsy

3. Check that there is no muscle contraction or “jump” with a brief stimulation of the “tendon” using the electrocautery.

Surgical Technique: Lengthening of the Iliopsoas in Non-ambulators (Fig. 18.17, 3, 4) During the adductor lengthening surgery described above, the interval between the pectineus anteriorly, and the adductor brevis posteriorly, is palpated and dilated by the index finger of the operating surgeon. In the younger, thinner child, it is relatively easy to palpate the proximal femur and, by rotation of flexed abducted hip, the “bump” of the lesser trochanter can be easily palpated. Following identification of the lesser trochanter by palpation, Hohman retractors are inserted anterior and posterior to the femur to allow visualisation of the lesser trochanter and tendon of the iliopsoas inserting on the lesser trochanter. A right angle forceps is passed around the lesser trochanter and the tendon is divided using the electrocautery, with care to avoid inadvertent injury to the adjacent neurovascular structures. Following division of the tendon, the lesser trochanter should be palpated again to confirm that it is clear, and free from any remaining attachments of the iliopsoas. Surgical Technique: Lengthening of the Psoas in Ambulators (Fig. 18.17, 5a, 5b) In ambulators, lengthening of the psoas at the brim of the pelvis is performed through a separate 3–4  cm skin crease incision, as part of the anterior approach to the hip. The tensor-sartorius interval is identified as a palpable depression 1–2 cm distal to the anterior superior iliac spine (ASIS). Following separation of the subcutaneous fat, tenotomy scissors are used to open the deep fascia and the inter-muscular interval. A layer of fat is routinely noted between the tensor fascia lata and sartorius. The lateral femoral cutaneous nerve should be identified in this interval and protected. The rectus femoris is deep in this interval, just anterior to the hip capsule. The dissection is deepened medial to the rectus tendon and the pelvic brim is then palpated. The hip is

499

flexed to relax the psoas and a right-angle retractor is used to retract the psoas and to visualize the tendon on the deep surface of the iliacus muscle. The femoral nerve lies on the anterior surface of the iliopsoas and the tendon is posterior. They are close to one another and can look very similar through a small incision. The three tests described by Skaggs are performed prior to transecting the tendinous portion of the psoas with preservation of the muscle fibres of iliacus.

Post-operative Management Lengthening of the iliopsoas is usually performed in conjunction with adductor lengthening or other procedures. Maintenance of hip extension is performed, during the time the child is in A-frame cast, by prone positioning for 1–2 h twice a day. Complications to Avoid Injury to the neurovascular structures is a rare but devastating complication to be avoided as outlined above. This is a procedure which should be taught by experienced surgeons to their residents and fellows to emphasize caution and safety. Pearls and Pitfalls • Preserve hip flexor strength by fractional lengthening in the musculotendinous portion of the tendon in children who are ambulators. • Maximise hip flexor lengthening by release at the lesser trochanter in non-ambulators. • Avoid injury to the neurovascular structures by never dividing a “tendon” until it has been confirmed that it is in fact a tendon and not a nerve.

Reconstructive Procedures: Varus Derotation Osteotomy of the Proximal Femur (VDRO) As discussed previously, with the adoption of hip surveillance in children with cerebral palsy, and the small treatment effect of adductor ­lengthening, varus derotation osteotomy of the femur (VDRO) can be considered to be both a preventive and reconstructive measure in the

J. J. Howard et al.

500

management of hip displacement in children with cerebral palsy. Proximal femoral deformity, consisting of combinations of increased femoral neck anteversion and coxa valga, are ubiquitous in children with cerebral palsy and are directly related to the child’s GMFCS Level [22]. Varus derotation osteotomy (VDRO) can be used in children aged 3–8 years as a preventive measure and in older children (Fig.  18.18), in combination with a pelvic osteotomy, as a reconstructive measure (Fig. 18.19). As such, the indications for VDRO in cerebral palsy include: the persistence of a high (>50%) or increasing MP after adductor lengthening, or the presence of a migration percentage in excess of 50% at initial presentation.

a

b

“The indications for VDRO include a persistently high (>50%) or increasing MP after adductor lengthening, or an MP >50% at initial presentation”. The primary indication for VDRO is a reducible subluxation of the femoral head in abduction and internal rotation. It must be clearly understood that VDRO does not reduce the femoral head; reduction needs to be accomplished by adductor lengthening and occasionally an open reduction. The VDRO is utilised to stabilise a hip that has been made reducible by either adductor lengthening or open reduction (i.e. opening the capsule). In this regard, spastic hip disease is completely different to DDH.  In children with DDH, opening up the capsule in the older child is routine because of the interposition of blocks to reduction. In contrast, in the majority of children with cerebral palsy and early hip displacement, lengthening of the hip adductors will permit the femoral head to sit deeply within the acetabulum in a position of abduction and internal rotation. Unless the hip has been neglected, it is our experience that opening the capsule is rarely necessary. The contraindications to VDRO in children with cerebral palsy include: severe femoral head deformity and the inability to reduce the femoral head inside the acetabulum by open or closed reduction.

c

Fig. 18.18  Eight year-old with quadriplegic cerebral palsy (GMFCS IV) who presented with pain, scissoring, and gait deterioration. (a) Bilateral hip displacement with MP > 50%, increased NSA, and mild to moderate acetabular dysplasia. (b) Three weeks post-operative bilateral hip adductor releases and VDROs. The hips are now covered and the NSA is over-corrected. (c) At 5  year post-­ operative follow up, the patient is pain free and has no fixed deformities. The lateral acetabular epiphyses have ossified and she has femoral head overcoverage. There has been mild growth disturbance and a “sagging rope sign” noted on the left hip. The decision to add a pelvic osteotomy is difficult but can sometimes be resolved by identifying the presence of a covering acetabular anlage on arthrogram. (NSA neck shaft angle, MP migration percentage, VDRO varus derotational osteotomy). (Used with permission, Copyright © Kerr Graham)

18  The Hip in Cerebral Palsy

a

b

c

Fig. 18.19 (a) Ten year-old (GMFCS IV) with left hip subluxation and progressive hip dysplasia identified on hip surveillance. (b) One week post-operative bilateral hip adductor releases, bilateral VDROs and left San Diego Acetabuloplasty. Note the use of the ‘keystone’ graft from the femoral varus and shortening osteotomy used to support the acetabular osteotomy. (c) At 4 year follow up, the patient has mobile, pain free hips. Radiographically, the hips are stable, well covered and the pelvis is level. (VDRO varus derotational osteotomy). (Used with permission, Copyright © Kerr Graham)

Pre-operative Planning For the majority of children, plain radiographs, taken in the neutral position, combined with additional views in abduction and internal rotation, will confirm if the femoral head is reducible. In older children with femoral head deformity and

501

acetabular dysplasia, it is sometimes helpful to obtain axial imaging, especially in the presence of a dislocation channel. However, a CT scan is not routinely required and should be used sparingly and selectively, only when it is going to be an adjunct to pre-operative planning. In our practice, we found it straightforward to plan the amount of varus and derotation for each hip, by pre-operative fluoroscopy when the child is in the operating room and ready for surgery (i.e. the abduction angle required to “reduce” the hip under fluoroscopy equals the amount of varus correction). Assessment of associated deformities including: scoliosis, pelvic obliquity, and lower limb contractures are important especially in the older child. As outlined in the previous section, assessment of the child’s medical state and aggressive preparation by correction of medical co-morbidities is essential. These include: optimising bone health, nutrition, respiratory status, correction of anaemia, stabilisation of seizure disorders etc. In the majority of children, bilateral surgery is required and this poses a substantial physiological stress to children who are frail with medical co-morbidities. The prevalence of medical complications (e.g. pneumonia, urinary tract infection) after VDROs is at least 50% and the prevalence of surgical complications is approximately 10% [66]. In approximately 10–20% of children, a blood transfusion will be required and this should be discussed with parents preoperatively. In medically frail children, there is a risk of mortality secondary to progression of pre-existing respiratory disease or an acute respiratory event, precipitated by a depression in respiratory function related to prolonged surgery, post-operative analgesia and other medications. The mortality rates will be determined by the pre-operative preparation and the degree of risk assumed by the surgeon and parents in medically frail children. A mortality rate of less than 1% is achievable and to be expected. In our practice, the majority of children will have had a prior lengthening of the hip adductors. If the adductor lengthening has been more than 2 years prior to the VDROs, a revision open adductor lengthening is usually required. Revision

502

adductor lengthening through scar tissue should be performed by a careful open procedure, prior to the performance of the VDRO. Every degree of varus in the proximal femur takes up between 1 and 2° of the available range of hip abduction post-operatively. Therefore, if the hip abduction is restricted by recurrent adductor contracture, a VDRO may result in an acute loss of hip abduction, the adoption of an adducted posture, and profound difficulties with sitting as well as failure to obtain a stable hip. Pre-operative planning should include a determination of the likely choice of implant, the full range of which must be available to meet any eventualities. Two-part devices such as: the Coventry lag screw and various forms of sliding hip screws are inherently unstable, do not permit medialisation of the distal fragment, usually require the addition of a hip spica and should be avoided in children with cerebral palsy. Of the currently available implants, blade plates offer distinct advantages and we believe they are much superior to proximal femoral locking plates. There is much more flexibility in the choice of pre-operative planning with a fixed angle blade plate than with the commonly available locking plates and they have been reported to result in shorter operative times and time to healing [112, 113].

Surgical Technique: Varus Derotation Osteotomy of the Proximal Femur (VDRO) (Fig. 18.20, 1–5) The patient is positioned supine on the operating table, with the pelvis level but raised on a folded sheet. If the patient is under 6 years and small in height, the C-arm may be positioned at the end of the operating table, to permit visualisation of both hips without having to change the location of the C-arm and the operating team to gain access to the right and left hip sequentially. For the reasons discussed previously, we routinely practise bilateral varus derotational osteotomies. Our only indication for unilateral femoral osteotomy is in patients with unilateral involvement (i.e. hemiplegia). This approach is supported by previous authors [114, 115]. Surgery is performed under general anaesthesia with insertion of the epidural catheter to pro-

J. J. Howard et al.

vide post-operative analgesia for 4–5  days. The urinary bladder is catheterised in a sterile fashion and the perineum is sealed with a plastic adhesive strip. If adductor lengthening surgery is required, we prepare a separate set of surgical instruments for the VDROs as it is more difficult to achieve the sterility in the perineal region required for the insertion of an implant in the proximal femur. The lower limbs are prepared and draped free so that the hips can be moved intraoperatively to conduct periodic fluoroscopic examinations with minimal movement of the C-arm. After the adductor release, and prior to the skin incision for the VDRO, fluoroscopy is used to confirm that the femoral head is reducible deep within the acetabulum and to establish the degree of varus and external rotation required for hip stability. In general, we choose a neck shaft angle at pre-operative planning related to the patient’s GMFCS Level and the degree of hip displacement. At GMFCS V we will aim for a neck shaft angle (NSA) of 90–100°, especially in younger children, to allow for “rebound coxa valga”. In children who are GMFCS III, we aim for a neck shaft angle of 110°. The elevation of the great trochanter will be of lesser importance if the child is going to be a long term user of a walking frame or crutches. In independent ambulators (GMFCS I and II) we aim for a neck shaft angle of 120–130°. With regards to femoral torsion, we aim for a residual femoral neck anteversion of between 5° and 15°, taking into consideration both the anteversion as well as the range of internal and external hip rotation. At the end of the procedure, for non-­ambulators, we prefer the hips to lie passively in a position of mild abduction and external rotation. In ambulators, the planning of the amount of derotation will be based on many factors including: motion analysis and pre-operative axial imaging. In older children with established acetabular dysplasia, the pre-operative fluoroscopic examination may also be a deciding factor in the decision for or against a pelvic osteotomy. We prefer to conduct the operation with the hip abducted and internally rotated so that the ­femoral neck is parallel to the operating table and to the floor as this permits the steps to

18  The Hip in Cerebral Palsy

503

proceed in an orderly fashion with easy insertion of the guide wire and seating chisel. In our hands, both the standard, solid, fixed angle blade plate and the more recently available cannulated locking blade plate are capable of delivering consistently stable fixation and good results [113]. A common mistake is to use implants that are too small for the size of the femur and the weight

1a

of the child. The size of the blade plate should be approximately 3–4  mm narrower than the narrowest part of the femoral neck, to allow an easy passage without risk of cortical penetration or causing an intra-operative fracture. Conversely, it is wise to use the largest and longest implant that will safely fit within the femoral neck on the lateral fluoroscopic view as well as extending to within 1–2 mm of the proximal femoral growth

1b

1c

30°–40°

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.20 Surgical technique for varus derotational osteotomy (VDRO). (1a) Following a lateral ‘sub-vastus’ approach, a guide wire is advanced into the proximal femoral metaphysis, at an angle depending on the amount of varus desired (usually 30–40°) and the angle of blade plate used (typically either 90° or 100° blade). (1b) The guide wire should be placed in the centre of the femoral neck on the lateral projection, facilitated by internally rotating the hip by 30–40°. (1c) The appropriately sized cannulated seating chisel is then advanced over the wire, stopping short of the physis. (2a) The first bone cut is made parallel to the seating chisel at a distance of 8–10 mm distal according to the chosen implant. The cut should be in the inter-trochanteric region, above the lesser trochanter but below the femoral neck. (2b) A vendorspecific spacer can be used to ensure the correct distance between the seating chisel and the osteotomy is maintained. (2c) Following complete division of the proximal femur, the proximal fragment is tipped into varus and a second cut is made to remove an irregular trapezoid “wedge” of bone which often includes the lesser trochanter (for non-ambulators). (2d) The distal fragment is then derotated to a position of approximately 5–15° of residual anteversion (i.e. external rotation of the distal fragment by approximately 20–30°). (3) The trapezoidal “keystone”

graft is removed, allowing enough femoral shortening to achieve appropriate varus, apposition of the osteotomy surfaces, and relaxation of the soft tissues. This graft is set aside for later use during the often required concomitant pelvic osteotomy. (4a) The third cut is in the proximal fragment of the proximal femur to allow for better medialisation of the blade plate without impairing proximal fixation. To achieve this, we cut a small “trap door” fragment distal to the seating chisel and no more than half of the width of the intact cortical bridge as shown. (4b) Lateral view of the “trap door” fragment (dotted line). (5a) After removal of the seating chisel, the selected cannulated blade plate (usually 90° or 100° blade) is inserted over the guide wire by gentle tapping with a mallet. The osteotomy is now reduced by control of the proximal fragment using the plate insertor and manipulation of the distal fragment by an assistant usually requiring a position of flexion and external rotation. This position is secured with a reduction clamp and preliminary fixation with two non-­ locking screws is performed. (5b) The reduction clamp is then removed and the position of the hip is check fluoroscopically. If all is satisfactory, the final screws are inserted. (5c) The final configuration should result in a neck shaft angle of approximately 90–100°. (Used with permission, Copyright © Kerr Graham and Bill Reid)

J. J. Howard et al.

504

2a

2b Spacer

First osteotomy

2c

2d

Second osteotomy Copyright ©, BR & KG, RCH Melbourne

3

4a

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.20 (continued)

4b

Copyright ©, BR & KG, RCH Melbourne

18  The Hip in Cerebral Palsy

505

5b

5a

5c

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.20 (continued)

plate without penetrating the calcar femorale. Undersized implants are a risk to osteotomy stability, healing, resulting in delayed rehabilitation and may necessitate the use of a hip spica. Hip spicas are to be avoided as they are an independent risk factor for major morbidity and possibly mortality [116]. Osteopenia, fragility fractures, skin ulceration and chest infections are associated with hip spica casts and are of major concern. The incision for VDRO begins at the tip of the greater trochanter, with the hip placed in internal rotation of approximately 30°–40° and 20°–30° of abduction. The incision extends along the proximal thigh, approximately the length of the selected blade plate or 1–2  cm more when significant shortening is to be performed. Following division of the skin and subcutaneous tissue, a large self-retaining retractor is placed in the incision and this aids haemostasis. An “L-shaped cut” is performed where the vastus lateralis meets the hip abductors and extended along the posterolateral femur close to the linear aspera, with

identification and cautery of perforating vessels posteriorly. The vastus lateralis is then retracted anteriorly and held in the retracted position by Hohman retractors placed anteriorly and posteriorly. The periosteum is incised and elevated with a Cobb elevator with particular attention to mobilise the periosteum adherent along the line of the linear aspera, beyond the length of the skin incision as the well as the intertrochanteric area. It is wise not to extend the periosteal elevation too proximally as this may inadvertently damage important circulation to the femoral metaphysis and even to the femoral capital epiphysis. The operative steps now depend on whether a cannulated or a solid blade plate is to be used. Given our preference for the cannulated device, this is the technique which will now be described. An appropriately sized guide wire with a cutting tip is advanced into the proximal femoral metaphysis and the femoral neck using fluoroscopic guidance in the AP and frog lateral position, to achieve the planned correction in terms of varus. The angle at which the guide wire

506

is advanced depends on the amount of varus desired (usually 30–40°) and the angle of blade plate used (typically either 90° or 100° blade). Particular attention should be given to placing the guide wire in the centre of the femoral neck in the lateral projection, which is obtained by ranging the hip into flexion, abduction, and external rotation. At this point, the size of the blade plate to be used can be confirmed by placing the seating chisel over the guide wire and comparing this on fluoroscopic viewing, to the narrow section of the femoral neck. Only 1–2 mm of clearance is required on each side of the blade plate and the goal is to “fit and fill” the proximal femur with the largest implant consistent with avoiding injury to the femoral neck and obtaining stable fixation. In the OrthoPediatrics™ cannulated locking blade plate system, the infant plate can be used for children up to approximately age 3–4 years and 12–15 kg in weight. The 3.5 mm child implants are used for children aged approximately 4–12  years and the 4.5  mm system for older children and adolescents. For a VDRO in CP we use a three cut technique [113]. Using the C or I guide, the bone is scored distal to the insertion of the guide wire to mark the position of the first bone cut. A horizontal score is also made in line with the lateral femur to check on the amount of derotation achieved intraoperatively. The first bone cut is made parallel to the seating chisel at a distance of 8–10 mm distal according to the chosen implant. The cut should be in the inter-trochanteric region, above the lesser trochanter but below the femoral neck. It is important not to stray into the femoral neck and the associated vasculature. Following complete division of the proximal femur, the proximal fragment is tipped into varus and the distal fragment is derotated to a position of approximately 5–15° of residual anteversion, which usually means external rotation of the distal fragment by 20–30°. At this point, it will be noted that the two cut surfaces of bone are mismatched. A second cut is made on the distal fragment to remove an irregular trapezoid “wedge” of bone, which often includes the lesser trochanter (for non-ambulators), and enough shortening to achieve appropriate varus, apposition of the osteotomy surfaces

J. J. Howard et al.

and relaxation of the soft tissues. Note that, for ambulators, only a transverse cut just proximal to the lesser trochanter is used—without resection of a bone wedge—to preserve hip flexor function. The third cut is optional but we believe it is a useful addition and perform it routinely. The majority of available implants do not offer enough medialisation of the distal fragment given the degree of varus correction required in the majority of children with cerebral palsy and coxa valga. The third cut is in the proximal fragment of the proximal femur to allow for better medialisation of the blade plate without impairing proximal fixation. To achieve this, we cut a small trap door distal to the seating chisel and no more than half of the width of the intact cortical bridge as shown in the accompanying diagrams (Fig. 18.20, 4a). The selected implant is mounted on the introducer and the seating chisel is removed. The implant is inserted over the guide wire. This should be easily achieved by a gentle tapping with a mallet. Significant resistance indicates dislodgment of the guide wire, a bent guide wire or incorrect positioning of the implant. The implant is advanced and seated appropriately in the proximal fragment. The osteotomy is now reduced by control of the proximal fragment using the plate insertor and manipulation of the distal fragment by an assistant usually requiring a position of flexion and external rotation. The distal fragment is approximated against the proximal fragment and if the fit is not satisfactory, further trimming of the distal fragment may be required. The position is secured with a reduction clamp and preliminary fixation in screw holes 2 and 4 (non-locking screws) is performed. The reduction clamp is then removed and the position of the hip is check fluoroscopically. The range of motion is checked and the stability and positioning of the osteotomy confirmed. If all is satisfactory, fixation is completed by the insertion of two locking screws into the first and third holes in the plate respectively. The proximal locking screw provides a second point of fixation in the proximal fragment and enhances the stability of the construct. Following completion of fixation, under fluoroscopy, the hip is ranged to assess joint stability and the reduc-

18  The Hip in Cerebral Palsy

tion of the femoral head is confirmed. The decision to proceed with a pelvic osteotomy is then made, dependent on joint stability and the adequacy of femoral head cover. The incision is then irrigated copiously with normal saline containing a cephalosporin antibiotic. The incision is closed in layers. The first and most important stitch is to close the vastus lateralis over the prominent shoulder of the plate by suturing the corner of the ‘L’ to the posterior aspect of the periosteum of the proximal femur. This provides padding and cover over the implant so that in the event of a superficial wound infection, it is much less likely to communicate with implant. Ensuring cover of the vastus ­lateralis over the shoulder of the plate, also reduces the risk of wound breakdown and provides additional comfort in malnourished children with poor subcutaneous fat. The incision is then closed in layers and an identical procedure is performed on the contralateral hip so that hips are matched in terms of: • • • • •

Neck shaft angle Leg length Femoral anteversion Hip range of motion Internal and external rotation

Post-operative Management Children should be monitored intensively after bilateral VDROs in terms of vital signs, blood loss, respiratory function, urinary output. We maintain the epidural infusion for a period of 3–5 days according to the child’s degree of hypertonia and frailty. If the child is malnourished, an air mattress is excellent protection against decubitus ulceration. We allow the child to rest with the hips and knees flexed approximately 30° over a pillow with a second pillow placed between the knees to maintain a modest range of abduction. Fixation should be stable enough to permit sitting out of bed in a wheelchair on post-operative Day 2. Children should be able to return to their pre-operative level of function including the use of standing frames without restriction within a week of surgery. We obtain an AP radiograph (no laterals are required) prior to discharge and at 3 and 6 weeks after surgery. In children with

507

cerebral palsy bony healing is generally rapid and complete within 3–6 weeks after surgery. We advise routine removal of implants at approximately 12 months after surgery and before excessive bone has formed over the blade plate making removal more difficult and more hazardous. Routine implant removal is controversial and may not be necessary in older children and some adolescents. However, in younger children with CP, with hip displacement detected early by hip surveillance, there is so much growth remaining that migration of the blade plate through the neck of the femur is a real risk. Early implant removal in younger children is strongly advised and in undernourished children, the majority of blade plates are both prominent and symptomatic. Tailoring VDROs to the Child’s Needs In children with relatively symmetric hip displacement, we perform bilateral and symmetric surgery. However, in children with asymmetric hip displacement, asymmetric hip abductor releases and, sometimes, asymmetries of the bony surgery need to be considered. In children with windswept deformities, the above rules no longer apply and customisation of the soft tissue and bony surgery is required. In children with windswept deformities, careful examination under anaesthetic should be performed to assess the passive range of motion of both hips and to confirm what degree of varus and rotation is required to achieve hip stability. In severe cases, on the abducted side a hip abductor release should be considered as well as a contralateral hip adductor release for the adducted hip. For most children with wind-swept deformities, the degree of structural anteversion and the degree of internal and external rotation are asymmetric. We aim to achieve a level pelvis by correction of infrapelvic deformity, and asymmetric range of hip rotations which generally means no external rotation on the abducted/externally rotated side and much more external rotation on the internally rotated/adducted side. Occasionally, a suprapelvic cause of pelvic obliquity is present which may necessitate surgical correction of a concomitant thoracolumbar scoliosis usually 4–6 months following the hip surgery.

508

“Occasionally, a suprapelvic cause of pelvic obliquity (i.e. scoliosis) is present which may necessitate surgical correction 4–6 months following the hip surgery”. When both hip displacement and scoliosis are present, it is our preference to perform the hip reconstructions first to achieve a level platform on which to correct the spinal deformity. Unfortunately, there is currently no good evidence to support this or the converse approach for cases where both hip displacement and scoliosis are coincident.

Operative Pearls and Pitfalls: VDRO

• Optimise all medical co-morbidities before VDROs. • Have a detailed pre-operative plan and have a full range of implants available. • Ensure that the hip is reducible and has full motion by adductor releases prior to VDRO. • Carefully consider the timing and sequencing of hip and spine surgery when there is hip displacement, pelvic obliquity, and scoliosis.

 uided Growth of the Proximal G Femur Given the frequency of hip displacement in children with cerebral palsy and the medical frailty of some non-ambulant children, less invasive surgery such as guided growth would appear to be an attractive option [66, 116]. In light of the inherent risks, some authors have used screw epiphysiodesis of the proximal femur, primarily a percutaneous procedure, to reduce the neck shaft angle and help stabilize the CP hip following soft tissue releases. Guided growth using a transphyseal screw has been applied previously to manage coxa valga and hip subluxation associated with a Type-2 growth disturbance (secondary to growth

J. J. Howard et al.

disturbance) following treatment of DDH, with some success [117, 118]. However, even in DDH, guided growth using a threaded screw in the inferomedial portion of the physis is at best a “holding procedure” to prevent valgus tilt of the proximal femur and reduce secondary acetabular dysplasia. Despite the apparent success of guided growth in DDH, as with all hip related issues, there is limited correlation between DDH and spastic hip disease. The proximal femur in children with cerebral palsy is typically associated with more substantial increases in both anteversion and coxa valga—with lateral tilt of the proximal femoral growth plate—than children with DDH [22]. In addition, for children with cerebral palsy, the proximal femur is small, the bone is osteopenic, and their growth velocity is unpredictable. Despite guided growth being minimally invasive procedure, children with CP are often poor candidates for the repeated general anaesthetics required to reposition a screw when the proximal femoral physis has “grown off” its threads. Ultimately, given the severe changes in proximal femoral morphology encountered, it is unrealistic to expect full correction of anteversion measuring 40° and coxa valga of 160°, by guided growth. In the literature to date, the changes in epiphyseal tilt, neck shaft angle, and migration percentage are modest at best [29, 31]. Additional studies, with long-term follow-up until skeletal maturity, are required which include the full reporting of adverse events associated with the multiple screw adjustments and replacements required. As such, we see little or no role for guided growth as an isolated procedure, although it may occasionally be of value to “fine tune” epiphyseal tilt in an ambulant patient who has shown some response to adductor surgery (Fig.  18.21). The second indication is in the prevention of rebound coxa valga, following previous successful hip reconstructive surgery. In children who have early VDRO, the biomechanical environment which resulted in coxa valga in the first instance, persists after bony reconstructive surgery. The rate of rebound coxa valga is highest in children who have surgery at an early age and who are

18  The Hip in Cerebral Palsy

a

b

Fig. 18.21  Six year-old with quadriplegic cerebral palsy (GMFCS IV) post bilateral adductor lengthening with residual lateral physeal tilt, increased NSA and MP of 34% and 44% for the right and left hips, respectively. (a) Immediate post-operative AP X-ray demonstrating transphyseal 6.5  mm screw placement in the inferomedial quadrant of the femoral epiphysis bilaterally for guided growth. (b) At 12  year post-operative follow up, the patient has mobile, pain free hips and can stand for transfers despite the presence of mild residual acetabular dysplasia. (NSA neck shaft angle, MP migration percentage). (Used with permission, Dr. Paulo Selber, RCH Melbourne)

non-ambulant. In such children, guided growth may have a limited role in prevention of rebound coxa valga and may reduce the need for repeat proximal femoral osteotomy. To be indicated for this procedure, however, the proximal femur must be of adequate size, and the patient must be medically fit enough for not only the index procedure, but for the inevitable screw replacement(s) and reposition(s) that will likely be required. For a small number of patients close to skeletal maturity, the alternative to temporary screw

509

epiphysiodesis is the induction of a small physeal bar in the inferomedial aspect of the proximal femur. This can be accomplished by the insertion of a guide wire and a 4  mm tap. The development of a small bony bar can be seen on followup X-rays, and this will at worst prevent recurrent lateral tilt of the epiphysis, increase in neck shaft angle (NSA) and secondary acetabular dysplasia. In some children, the induction of a physeal bar will result in progressive inward tilting of the epiphysis, improving containment and femoral head cover. However, since the growth mechanics are not clearly understood, the induction of a physeal bar should be reserved for children who are close to skeletal maturity; with small amounts of coxa valga and small degrees of hip displacement.

 urgical Technique: Guided Growth S of the Proximal Femur With the patient supine on a radiolucent table, a terminally threaded guide wire is inserted percutaneously and advanced into the inferomedial quadrant of the femoral epiphysis. The tip of the guide wire should be advanced to the subchondral bone without penetrating the joint. On the lateral fluoroscopic view, the guide wire should be in the center of the epiphysis. The skin around the guide wire is incised for passage of a 7.3 mm fully-threaded cannulated screw and the screw size is measured. The bone is drilled and the screw inserted, ensuring at least 2–3 threads are above the physis. A ‘near-far’ technique (as is used for in situ screw fixation for slipped capital femoral epiphysis) is used to ensure no screw penetration of the joint has occurred. Post-operative Management Full weightbearing is permitted immediately postoperatively. If concomitant soft tissue releases are performed, the post-operative management is the same as previously discussed. Otherwise, full range of motion is encouraged. Although there will likely be an immediate decrease in MP due to the soft tissue release, further decreases in MP may progress over time depending on the age of the patient at the time of screw insertion.

510

 econstructive Procedures: San R Diego Acetabuloplasty

J. J. Howard et al.

 urgical Technique: San Diego S Acetabuloplasty (Fig. 18.22, 1–4) We prefer the straight “bikini” incision described Given recent studies which indicate significant by Dr. Robert Salter for innominate osteotomy as acetabular remodelling following adductor length- this results in the most cosmetic healing and gives ening and VDRO in children with cerebral palsy, excellent access and visualisation of the acetabuthe precise indications for pelvic osteotomy are lum [121]. The lateral femoral cutaneous nerve not fully established [98]. In general, the older the is identified under the fascia lata and is freed and child, the greater the degree of hip uncovering, and retracted medially. The interval between tensor the greater the degree of acetabular dysplasia, the fascia lata and sartorius is identified and opened more consideration needs to be given to a concom- with blunt dissection down to the conjoint tendon itant pelvic osteotomy. Following adductor release of the rectus femoris. and VDRO, if the hip is demonstratively unstable The iliac apophysis is split centrally by careful in axial loading during fluoroscopic examination palpation from anterior to posterior allowing for (i.e. “pistoning”), the hip is uncovered by more subperiosteal exposure of the medial and lateral than 30%, or the acetabular index is in excess of walls of the ilium immediately above the acetab30°, pelvic osteotomy should be performed. ulum. Typically, the lateral wall is exposed subperiosteally to the level of the sciatic notch. The “Following adductor release and VDRO, medial wall is also exposed to permit complete if the hip is unstable in axial loading, is division of the most anterior and most posterior uncovered by more than 30%, or has an segment of the supra-acetabular ilium. A Cobb acetabular index > 30°, a pelvic osteotomy elevator may be used to mobilize any adherent should be performed”. hip capsule from the supra-acetabular region to For many patients, this will be clear from pre-­ just above the acetabular margin. Bleeding from operative planning; in others, a final decision can the ilial nutrient foramina should be controlled only be made after the adductor lengthening and by coagulation or bone wax. In children with VDRO.  To allow for this eventuality, the pre-­ cerebral palsy, we believe that an open reduction operative skin preparation and draping must have is rarely indicated and as such, the rectus femoris extended to the umbilicus and sufficiently proxi- tendon does not need to be mobilised. The level and direction of the acetabular ostemal and posterior to allow complete access to the otomy is identified using AP fluoroscopy, faciliiliac crest. A San Diego acetabuloplasty is the procedure tated by the insertion of a 2–3 mm Steinmann pin. of choice for the younger child with neuromuscu- The pin should be inserted 10–15 mm proximal lar hip displacement, cerebral palsy and an open to the acetabular margin and directed medially triradiate cartilage [119, 120]. The keystone graft just above the triradiate cartilage. The width of for the acetabuloplasty will have been obtained the segment to be preserved between the bony cut during the VDRO by removing the femoral and the acetabulum needs to be wide enough to fragment in a single piece including the lesser resist the propagation of a fracture into the joint but narrow enough to allow the superacetabular trochanter. bone to be levered downwards to affect a reduction in acetabular index.  re-operative Planning P We perform the first cut with a curved half The decision to perform an acetabuloplasty may have been clear from pre-operative assessment inch osteotome, under fluoroscopic guidance, at or may be concluded after adductor lengthening the apex of the osteotomy; usually midway from and VDRO. Elevation of the ipsilateral hip on a anterior to posterior with the osteotome flat in radiolucent ‘bump’ under the iliosacral region is profile on fluoroscopic images. The osteotome helpful to both surgery and fluoroscopic imaging. is advanced from lateral to medial and stops just

18  The Hip in Cerebral Palsy

511

1

Triradiate cartilage

Triradiate cartilage

Lat

Med

Ant

Copyright ©, BR & KG, RCH Melbourne

2

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.22  Surgical Technique for San Diego acetabuloplasty. Using a straight “bikini” incision, an anterior “Salter” approach to the acetabulum is performed. (1) The line of the osteotomy starts 10–15  mm proximal to the acetabular margin and directed medially to just above the triradiate cartilage. The cuts extend through the medial cortices both anteriorly and posteriorly but must remain incomplete at the level just above the triradiate cartilage. (2) Curved osteotomes of variable sizes are ideal to perform these cuts, with careful retraction of the soft tissues and protection of the contents of the greater sciatic notch by blunt-tipped retractors. (3) When the osteotomy becomes mobile, the supraacetabular fragment is levered inferiorly, initially with a curved osteotome, and then held

open by a laminar spreader. The acetabular correction is checked on fluoroscopy and When satisfied with the correction, the trapezoidal “keystone” femoral graft (Fig. 18.20-3) is shaped appropriately, and impacted into the apex of the osteotomy until its outer cortex is flush with the ilium. (4) Additional wedge autografts from the iliac crest can be used to supplement the fixation anterior and posterior to this keystone graft. With the bone just superior to the triradiated cartilage as the fulcrum, AP fluoroscopic images should confirm the reduction in acetabular index (arrow) and stability under stress views. No internal fixation is typically required. (Used with permission, Copyright © Kerr Graham and Bill Reid)

J. J. Howard et al.

512 Keystone femoral autograft

3

Copyright ©, BR & KG, RCH Melbourne

4

Iliac crest autografts

Copyright ©, BR & KG, RCH Melbourne

Fig. 18.22 (continued)

short of the medial wall of the ilium and slightly superior to the triradiate cartilage. Additional contiguous cuts are then performed first anteriorly and then posteriorly to this apical cut resulting in a semicircular, incomplete osteotomy 10–15 mm above the acetabulum, directed towards the triradiate cartilage. The cuts extend through the medial cortices both anteriorly and posteriorly but must remain incomplete at the level just above the triradiate cartilage. Curved osteotomes

of variable sizes are ideal to perform these cuts, with careful retraction of the soft tissues and protection of the contents of the greater sciatic notch by Rang/Toronto interlocking retractors. The most posterior part of the cut (1–1.5 cm) can be performed with a Gigli saw as performed for the innominate osteotomy, described by Dr. Salter. As the osteotomy becomes more mobile, it becomes possible to lever the supraacetabular fragment inferiorly, using a combination of

18  The Hip in Cerebral Palsy

curved osteotomes, supplemented by a laminar spreader. An additional/alternative instrument that can be used at this point is the distraction forceps, commonly utilised during os calcis lengthening, to open up the osteotomy site. Once the supra-acetabular fragment is sufficiently mobile, the correction achieved should be visualised on fluoroscopy to assess if it is adequate. The goal for correction is to bring the ­acetabular roof into a horizontal orientation. When satisfied with the correction, the trapezoidal femoral bone segment can be shaped and used as a keystone, providing a structural graft and impacted at the apex of the osteotomy. The first apical graft should be large enough and strong enough to be stable when wedged into the open osteotome so as not to either dislodge or collapse. Additional bone graft from the iliac crest can be cut to supplement the fixation anterior and posterior to this keystone graft. These are usually tricortical wedges cut with an oscillating saw from the iliac crest. Once the grafts have been inserted, the stability of the correction needs to be ascertained by ranging the hip and inspecting the grafts as directly as well as using fluoroscopy. The stability should preclude the need for any internal fixation. Following haematosis and irritation with normal saline containing cephalosporin, the iliac apophysis is carefully repaired using multiple towel clips to approximate the cut surfaces together. The repair is performed using interrupted Vicryl sutures starting at the apex and then moving both anteriorly and posteriorly until a complete, watertight repair has been performed. The skin incision is closed in layers and a drain is not usually required. Attention is then directed towards the contralateral hip. Although we perform bilateral adductor releases and bilateral VDROs for the majority of children, we sometimes find that a pelvic osteotomy is required on one side only. The decision as to whether to perform unilateral or bilateral pelvic osteotomies should be customised to the needs of the individual patient. So as long as the bone grafts have been inserted in the correct manner and neither left too proud nor excessively impacted, the osteotomy should be stable.

513

Post-operative Management As the number of procedures increases, so does the need for blood transfusion. In our care, fewer than 10% of patients with bilateral VDROs need a blood transfusion. Once a pelvic osteotomy is added, however, the percentage of children requiring peri-operative blood transfusion increases significantly and an intraoperative tranexamic acid infusion may be helpful in this regard. Despite the addition of the pelvic osteotomy, we place no additional restrictions on mobilisation, allowing the child to use a wheelchair on post-operative day 2 and standing frame between 1 and 2 weeks after surgery. Immediate range of motion is encouraged. We use a hip spica cast only in patients with extreme osteopenia. Complications Incomplete osteotomy, or inadequate width of the acetabular fragment, may result in fracturing or bony deformation during attempted opening. This can be avoided by patience and performing sufficient cuts both anteriorly and posteriorly until the osteotomy is fully mobile and by ensuring an appropriate width of the acetabular fragment. Graft collapse is more common if a strong structural femoral graft is not available. With the intertrochanteric femoral graft, the graft material is both strong enough and big enough to resist compression. If no femoral graft is available, and the child is very osteopenic, allograft supplementation can be considered. Insufficient reduction of the hip joint is a possibility, and this is more likely if the varus in the proximal femur is inadequate. As the number of procedures performed around the hip increases (adductor release, VDRO, pelvic osteotomy) so does the risk of more serious degrees of avascular necrosis (AVN) [122]. In our hands, serious AVN with femoral head collapse is fortunately uncommon. In the reported series, it is difficult to establish which factors are etiologic, apart from the number of concomitant operative procedures [123].

514

“As the number of procedures performed around the hip increases, so does the risk of avascular necrosis”. Customising Acetabular Re-direction The San Diego osteotomy can vary the degree of anterior lateral and posterior cover by adjusting the apical position and completeness of the osteotomy cuts. If mostly anterior and lateral cover is required, the sciatic notch should be left intact as well as the posterior one third of the medial table. As discussed in the Pathophysiology section, the location of acetabular dysplasia in the CP hip is often either posterior or global [45]. As such, consideration should be given to completing the osteotomy through the sciatic notch leaving the mid-portion of the medial ilium intact to act as the hinge. For a complete division of the ilium at the level of the greater sciatic notch, the contents of the greater sciatic notch (sciatic nerve and superior gluteal artery) should be protected by interlocking Rang/Toronto retractors and a Gigli saw can be used to perform the most posterior part of the osteotomy.

Operative Pearls and Pitfalls: San Diego

• Femoral autograft is ideal for the keystone graft. • If an isolated San Diego acetabuloplasty is required, it is wise to have allograft in reserved in the absence of femoral autograft. Bone graft may also be harvested from the anterior iliac crest. In either case, the bone graft must be stable enough to resist the compressive forces of the osteotomy. • The outer cortex of the femoral autograft should line up with the cortex of the lateral ilium and supraacetabular bone. Maintain an adequate medial hinge to ensure stability and compression and adjust the position of the hinge to adjust the cover to more anteriorly or more posteriorly as required. The use of

J. J. Howard et al.

laminar spreaders or cannulated distraction forceps provide a slow, controlled distraction of the osteotomy and allow insertion of the apical graft with less obstruction than simply using osteotomes to hold the osteotomy open.

 ther Pelvic Osteotomies in Cerebral O Palsy The original Dega procedure, the acetabuloplasty described by Tönnis, and the acetabuloplasty described by Miller and colleagues, have all been used in CP hip dysplasia, with good reported results [28, 42, 124]. The differences between these procedures and the San Diego acetabuloplasty are relatively small and the outcomes are likely to be similar; depending more on indications, technique, and after care than the choice of index procedure. Given the high prevalence of posterior acetabular deficiency in the CP hip, the innominate osteotomy of Salter is contraindicated because it provides increased anterior and lateral cover, at the expense of uncovering the femoral head posteriorly. “The innominate osteotomy of Salter is contraindicated in CP because it provides increased anterior and lateral cover, at the expense of uncovering of the femoral head posteriorly”. The Pemberton osteotomy also was also designed to provide more anterior and lateral cover but there are some studies reporting satisfactory outcomes in the CP hip [106]. “Salvage” procedures such as slotted acetabular augmentation and Chiari osteotomy have been reported but the indications for such procedures are unclear [125, 126]. High grade acetabular dysplasia in the ambulant (and occasionally in the non-ambulant) adolescent with CP, may benefit from a greater correction than can be delivered by acetabulo-

18  The Hip in Cerebral Palsy

plasty, particularly for patients with a closed tri-­ radiate cartilage [127]. There is increasing interest in the use of triple innominate osteotomy (TIO) and periacetabular osteotomy (PAO) in the CP hip with several series reporting good short term results [128, 129]. As described above, adductor lengthening and correction of abnormal proximal femoral geometry are essential pre-­requisites to successful TIO or PAO in CP (Fig. 18.23). In the non-ambulant patient, TIO or PAO will rarely be needed if screening, early diagnosis, and early reconstruction are practiced.

 alvage Surgery for Painful S Dislocated Hip in Cerebral Palsy Salvage surgery for painful hip dislocation, not amenable to reconstruction, has a 40  year plus history of trial and failure [11, 37, 130]. There are many different options for salvage, each with their own set of problems (Fig. 18.24). The Girdlestone excision of the proximal femur along the intertrochanteric line [131], used in the past for osteoarthritis with some suca

Fig. 18.23  Twelve year-old with Winters, Gage and Hicks Type IV right hemiplegia (GMFCS II). The patient had a previous right-sided multilevel surgery for gait correction including: adductor lengthening, femoral derotational osteotomy (no varus), hamstring lengthening, gastroc-soleus lengthening, tibialis posterior lengthening and split transfer of tibialis anterior (SPLATT). (a) Four years post-operatively, persistent high grade acetabular dysplasia (with reduced center-edge angle, and mild but

515

cess, has also been used for palliation in CP, but there are few reports and most authorities allude to failure due to persisting contact between the proximal femur and pelvis [132–135]. Contact between the resected surface of the femur and pelvis is a much greater problem in patients with cerebral palsy than in typically developing individuals, because of the sensitisation of the hip capsule to pain and the hypertonia which drives the two surfaces together [7]. In an effort to prevent persisting hypertonic contact between the proximal femur and the pelvis, Castle and Schneider, described a more radical resection of the proximal femur, 2–3 cm distal to the lesser trochanter, combined with an interposition of soft tissue across the acetabulum [136]. The interposition consisted of a repaired capsular flap, and the proximal quadriceps muscle. Since then, the Castle technique has been used relatively frequently, but modifications continue to be made to the soft tissue interposition, firstly by McCarthy [137] and more recently by Eastwood and colleagues [138]. In addition, other authors have inserted various materials in the stump of the proximal end of the femur in b

persistent upward pelvic obliquity) was present on the right side. (b) Peri-acetabular osteotomy was performed with a good clinical and radiographic outcome at 3 year follow up. PAO can be very useful in ambulant adolescents with persisting acetabular dysplasia but muscle lengthening and correction of substantial femoral deformity are important pre-requisites. (Used with permission, Copyright © Kerr Graham)

J. J. Howard et al.

516 Fig. 18.24  Overview of salvage procedures used for palliation for the unreconstructable hip in cerebral palsy. (Used with permission, Copyright © Kerr Graham and Bill Reid)

Salvage Surgery

Replacement arthroplasty

Valgus osteotomy alone

Arthrodesis

Shoulder prosthesis

Excision plus valgus osteotomy (McHale)

effort to limit proximal migration, heterotopic ossification and recurrent pain. These have included “capping” the end of the femur with a cartilage cap, with bone cement or the insertion of a shoulder prosthesis, with no effort to constrain the prosthesis within the acetabulum [139, 140] (Fig. 18.25). Alternatives to proximal femoral resection include the Schanz Pelvic Support Osteotomy [141] which has also been combined with proximal femoral resection, described by McHale and colleagues [142]. Other alternatives to proximal femoral resection include arthrodesis of the hip

Subtrochanteric excision (Castle)

and total hip arthroplasty [135, 143]. However, these latter procedures are suitable only for a small sub-group of patients with end stage spastic hip disease, not the typically fragile teenager with poor bone stock, bilateral hip disease, pelvic obliquity, scoliosis and multiple medical co-­morbidities. The literature clearly suggests that total hip arthroplasty is the most effective solution in terms of pain relief but in general, it has been reserved for ambulant patients, GMFCS patients II and III; with a minority of patients at GMFCS IV and no convincing reports of successful arthroplasty for patients at GMFCS V [11]. Similarly, arthrodesis has a high

18  The Hip in Cerebral Palsy

517

a

b

c

d

Fig. 18.25 Salvage surgery requiring revision in a 14  year-old male with severe spasticity/dystonia and windswept resulting in pain and sitting intolerance after spinal fusion. The right hip (high side) had a long standing anterior dislocation with severe deformity and little remaining articular cartilage. (a) A right Castle procedure (capped with bone cement) was performed, and the left hip (low side) had a VDRO, resulting in transient improvements in pain relief and sitting tolerance. (b) Within 4 weeks post-operatively, proximal migration of the right

femoral stump occurred with painful impingement. (c) Subsequent revision to a shoulder prosthesis (with femoral shortening) and soft tissue interposition was performed again with short term relief of symptoms. (d) Six weeks post-operatively, proximal migration of the prosthesis occurred with a subsequent return of pain and sitting intolerance. This case highlights that the outcome of salvage surgery is at best uncertain and it is usually impossible to “salvage the salvage”. (Used with permission, Copyright © Kerr Graham)

complication rate. Once fusion is achieved, however, pain relief is excellent; but the procedure is only suitable for unilateral hip disease, in patients with no significant scoliosis, and who are by definition ambulant or partially ambulant.

tions of the procedures [11, 144, 145]. In general, the studies recorded suffer from poor methodical quality. Few studies include a valid pre- and post-operative measure of pain in individuals who have limited communication abilities, despite pain being an indicator for surgery. Even fewer studies report a valid measure of health-related quality of life. Reports of both surgical and medical adverse events are limited and few studies have been independently reviewed and fewer still have a valid and reliable method for reporting adverse events.

 iterature on Salvage Surgery L In recent years, there have been a number of systematic reviews of salvage surgery for end stage spastic hip disease which are helpful in understanding the limitations of the literature and the limita-

518

Authors’ Experience As in most areas of Orthopaedic practice, when there are numerous surgical options for the treatment of a single condition, it is usually testament to the fact that none of them are satisfactory. It is the authors’ experience, that the poorest outcomes of the reconstructive surgery are generally superior to the best outcomes of salvage surgery. “The poorest outcomes of the reconstructive surgery are generally superior to the best outcomes of salvage surgery”. We therefore make every effort to avoid the need for salvage surgery by hip surveillance, and the offer of early preventive and reconstructive surgery. When reconstructive surgery fails, early detection can usually result in a secondary reconstructive option rather than resorting to salvage.

 edical and Surgical Adverse Events M Associated with Salvage Surgery Given that the majority of patients presenting with end stage spastic hip disease are non-­ ambulant adolescents in GMFCS levels IV and V, the peri-operative risks due to associated co-­ morbidities are substantial. Many adolescents presenting with end stage hip disease may have had limited access to nutritional support, poor seizure control, and respiratory compromise secondary to recurrent bouts of aspiration-related pneumonia; increasing these risks even further. Some present in a near terminal condition and are best treated with compassionate palliative care rather than heroic surgical adventures. A thorough assessment by a multidisciplinary team is advised prior to any operative procedures. If focal or global measures for tone management are options, this is the time to consider them. Injections of Botulinum Toxin A (BoNT-A) have been reported to give short-term relief [104]. However, injections of BoNT-A, can be dangerous in this population and are unable to deliver sustained relief of pain and spasm in those who have a reasonable chance of survival [146]. For patients who are medically stable, global tone management by an insertion of an intrathecal baclofen pump (ITB) can occasionally

J. J. Howard et al.

be sufficient to alleviate the need for salvage surgery [147]. Even when an ITB pump is not sufficient to relieve all pain, it provides a much more favourable environment for salvage surgery, should that be necessary. The authors have limited experience with salvage surgery and can offer only limited impressions from a small number of patients.

 urgical Adverse Events S The principal mode of failure of salvage surgery is persistent or recurrent pain. For procedures where the proximal femur is resected, the pain is thought to be due to proximal migration of the proximal femur resulting in contact between the femur and acetabulum. “The principle mode of failure of salvage surgery is persistent or recurrent pain likely due to abutment of the proximal femur against the acetabulum”. This is despite the quality of the soft-tissue interposition or the use of post-operative interventions to reduce migration, such as skin traction, skeletal traction, hip spica cast mobilisation or the use of a hinged distractor. It is unrealistic to expect a soft-tissue interposition to prevent abutment of the femoral stump against the pelvis given the strong forces generated by chronically hypertonic muscles crossing the resection site. This is the principal reason for the high revision rate reported for all forms of proximal femoral resection with medium to long-term follow up as discussed above. The second major complication of proximal femoral resection is heterotopic ossification (HO) which, despite being well-recognised, is hard to prevent in medically frail individuals. Heterotopic ossification, is commonly associated with chronic recurrent pain, likely secondary to impingement between the HO mass and the pelvis. Although the literature regarding the prevention of HO following salvage surgery is currently insufficient, it is our experience that neither perioperative non-­steroidal anti-inflammatory drugs nor pre-­ operative radiation can reliably eliminate this most troublesome problem [145, 148].

18  The Hip in Cerebral Palsy

The third disconcerting problem with proximal femoral excision is the instability of the remaining femur, which on occasions can penetrate through the interposed muscle and exit the lateral thigh skin, with predictable consequences. Instability from proximal femoral excision precludes standing transfers and is contraindicated if the patient has the ability to assist with transfers. More conservative forms of proximal femoral resection including the McHale procedure (resection plus valgus proximal femoral osteotomy), or procedures such as the Schanz osteotomy (in which resection is avoided), are associated with a decreased risk of HO and less instability of the limb secondary to the “pelvic support” provided by the valgus osteotomy [11]. These procedures, however, can still be marred by painful abutment of the proximal femur against the pelvis as well as stiffness post-operatively; usually in the form of an abduction contracture. In addition, the McHale procedure has a significant association with delayed union of the valgus osteotomy and problems associated with hardware requiring revision surgery [145]. The only situation in which we found reasonably consistent relief of pain, and improvements in sitting tolerance, were for those individuals with a severe windswept deformity; with a high side unilateral dislocation associated with a low side enlocated hip exhibiting a severe abduction contracture. In such patients, the combination of a McHale procedure on the dislocated side and a major varus derotation osteotomy with shortening on the enlocated side, improves pelvic obliquity, abduction range, seating and comfort. However, we have used this combination in a small number of patients, with a relatively short-­ term follow-up and no valid measures of health-­ related quality of life. In conclusion, it is unwise to neglect hip displacement in ambulant individuals with cerebral palsy who are usually able to report pain and who will most certainly suffer severe functional impairment in the future from neglected hip displacement. Despite associated medical frailties and the difficulties of reconstructive surgery in non-ambulant patients, it is also our view that displaced hips should be identified early and treated. Salvage surgery is unpredictable and

519

when it fails there are almost no procedures to “salvage the salvage”. “Salvage surgery is unpredictable and when it fails there are almost no procedures to “salvage the salvage”.”

 alvage Surgery: Excision S of the Proximal Femur Combined with Valgus Osteotomy, the McHale Procedure The McHale procedure is our preferred choice for salvage hip surgery in adolescents with severe CP, in whom neither hip reconstruction nor total hip arthroplasty are feasible. Like other salvage procedures, it is not predictable in terms of pain relief and has a significant rate of revision surgery. However, we think it is a better option than the Castle procedure because there is more stability, less HO and less proximal migration. Indication for Surgery and Pre-operative Planning The indication for the McHale procedure, and for salvage procedures in general is severe hip pain, with a deformed femoral head and acetabulum, in an adolescent or young adult with severe CP (GMFCS IV and V). The McHale procedure is contraindicated in ambulant persons and in persons who are able to perform standing transfers. The pain should be confirmed as originating in the hip, preferably by a response to injection of local anaesthetic and corticosteroid. Neo-­ adjuvant tone management is an essential pre-­requisite to surgery, using combinations of oral medications (Baclofen and Gabapentin), injectable neurolytics (e.g. phenol) and, when indicated, ITB. “Neo-adjuvant tone management is an essential pre-requisite to salvage surgery, using combinations of oral medications, injectable neurolytics, and, when indicated, intrathecal baclofen” In some patients, the pain relief following ITB pump insertion obviates the need for salvage surgery. At least, for the majority of patients, ITB reduces post-operative pain and proximal migration following the McHale procedure.

J. J. Howard et al.

520

All medical co morbidities should be investigated and optimized before surgery, by an experienced multidisciplinary team. It is important to evaluate both hips, pelvic orientation, and the spine before surgery. If there is a substantial scoliosis present, the surgical plan needs to consider the entire spine/pelvis/hips axis. Patients must be carefully assessed clinically and radiographically for supra-pelvic and infra-pelvic contributions to fixed deformities. For patients with windswept hips, it is usually the hip on the high side of the pelvis which dislocates in a position of adduction and internal rotation. However, abduction/external rotation contracture of the contralateral hip may prevent leveling of the pelvis. Even when the hip on the low side is enlocated, fixed deformity must be addressed, most often by VDRO for relief of abduction contracture. It is usually our practice to level any pelvic obliquity with hip surgery first; providing a stable foundation upon which the scoliosis can be subsequently corrected surgically. Surgical Technique: McHale Procedure The patient is placed supine on the operating table and the affected hip raised on a radiolucent bump. Both lower limbs are prepared and draped free. We have most frequently used a Watson Jones anterolateral approach. Gaining adequate access to dislocate the hip yet preserve blood supply to the greater trochanter can be problematic. Surgical hip dislocation and osteotomy of the greater trochanter may offer some advantages but there are no published reports. The presence of previous scars may dictate modifications to the incision but usually extends proximal to the greater trochanter by approximately one third of the total incision (Fig. 18.26a). If the dislocation is high, it can be helpful to mark the location of the femoral head and neck using fluoroscopy, so that the anterosuperior extension of the skin incision can follow. Using the interval between gluteus medius and tensor fascia lata, the hip capsule is cleared of pericapsular fat and the lateral aspect of the proximal femur is exposed subperiosteally using a sub-vastus approach. Capsular exposure is helped by division of the rectus femoris tendon and retraction of the iliopsoas. The position of the femoral head and neck is confirmed by external

a

b

c

Fig. 18.26  Surgical approach for the McHale procedure. (a) Surgical incision. The greater trochanter and anterior superior iliac spine are also marked. (b) Exposure of the proximal femur through an extended Watson-Jones (i.e. antero-lateral approach). Note the severe femoral head dysplasia and cartilage degeneration. The residual cavity post concomitant removal of a previous blade plate is evident. (c) Immediate post-operative view following femoral head resection and valgus osteotomy of the proximal femur. (Courtesy Jason Howard)

rotation of the hip. A wide capulotomy is then performed and the hip gently dislocated anteriorly by external rotation and adduction (Fig. 18.26b). Occasionally, the use of a bone hook around the femoral neck is used to aid in dislocation.

18  The Hip in Cerebral Palsy

Most hips requiring the McHale procedure are dislocated and it is relatively easy to deliver the head and neck into view by gentle external rotation, sometimes with adduction. However, if the hip is stiff and enlocated, it may be difficult to dislocate and there is a real risk of fracturing the femoral shaft. In such cases, it may be safer to divide the neck along the intertrochanteric line and remove the femoral head and neck fragment piecemeal. The damaged head and neck are excised using an oscillating saw along the inter-trochanteric line. Bleeding from the cut surface

521

can be reduced by bone wax which may also reduce HO. Attention is now given to the valgus osteotomy. A 6–8 hole pelvic reconstruction plate is a useful implant for this as it is easily bent and may reduce the stress shielding sometimes seen with locking plate fixation (Fig.  18.27). A proximal femoral valgus correction of 60–75° is required to tilt the cut surface of the femoral neck away from the acetabulum, allow adequate abduction for perineal hygiene, and to medialize the distal femur, which is a major advantage when compared to the Castle procedure.

a

b

c

d

Fig. 18.27  McHale procedure for painful recurrence of right hip displacement in a 10 year-old with spastic quadriplegia (GMFCS V) 5 years post unilateral varus derotational osteotomy and medial soft tissue release. (a) Subluxated right hip with a migration percentage of approximately 75% and severe femoral head and acetabular dysplasia. (b) Immediate post-operative radiograph showing bicortical fixation and minimal prominence of

proximal locking screw tips at intact base of femoral neck. (c) Substantial resorption of the base of femoral neck with uncovering of both proximal locking screw tips and abutment of the superior edge of the acetabulum at 8 months post-operatively, resulting in severe pain. (d) Internal fixation was removed 10  months after the index surgery resulting in a reduction, but not elimination of, the patient’s pain. (Courtesy Jason Howard)

J. J. Howard et al.

522

The plate is bent and pre-contoured to the proximal femur and 2–3 screws are inserted in the proximal fragment. The position for the osteotomy is marked on the femur and the plate and screws removed. A laterally based wedge is removed to permit valgus of greater than 60°. Fixation is started proximally and continued distally using a reduction clamp to hold the distal femur against the plate. Fluoroscopy is used to assess fixation, alignment, stability and movement of the femur into both abduction and adduction. Following completion of the osteotomy fixation, identify and suture the remnants of the iliopsoas tendon to the ligamentum teres or medial capsule. The goal is to keep the lesser trochanter directed towards the acetabulum whilst also maintaining soft tissue interposition between proximal femur and pelvis. This also reduces the chance of lateral displacement of the proximal femur, threatening the healing of the lateral skin incision, which is a greater risk with the Castle procedure. Partial closure of the capsule helps interpose soft tissue between the proximal femur and acetabulum. Following irrigation with normal saline containing antibiotic solution, the incision is closed in layers. Neither traction nor casting are necessary or helpful.

Operative Pearls and Pitfalls: McHale Procedure

• Assess the spine, pelvis and both hips • Optimize tone management: possibly intrathecal baclofen • Preserve adequate soft tissue attachment and blood supply to the greater trochanter • Protest the femoral nerve and vessels by careful retraction of iliopsoas • Dislocate the hip gently, to avoid femoral fracture • Continue adequate pain and tone management for a minimum of 6 weeks after surgery • There is usually little external callus, union can be slow and hardware problems are common

Essential Surgical Techniques

Preventative • Non-ambulators –– Soft tissue release (STR): adductor longus, gracilis, iliopsoas at lesser trochanter, +/− adductor brevis (partial) –– Varus derotational osteotomy (VDRO; for MP  >  50% or progressive post STR) • Ambulators –– STR: adductor longus, gracilis, psoas at the pelvic brim Reconstructive • Non-ambulators –– STR: adductor longus, gracilis, iliopsoas at lesser trochanter, +/− adductor brevis (partial) –– VDRO –– San Diego Acetabuloplasty • Ambulators –– STR: adductor longus, gracilis, psoas at the pelvic brim –– Need for VDRO rare except in Type IV hemiplegia Salvage • Non-ambulators –– McHale femoral head resection  +  valgus osteotomy (pre-operative tone management essential) • Ambulators –– Total hip arthroplasty

Classic Papers Phelps WM.  Prevention of acquired ­dislocation of the hip in cerebral palsy. JBJS-A 1959;41(3):440–448. One of the earliest papers focused on hip displacement in CP, Phelps implicated abductor insufficiency and muscle imbalance as being causative of hip displacement rather than the actions of spastic hip adductor and flexor muscles on their own. In addition, he surmised that the development of coxa valga was second-

18  The Hip in Cerebral Palsy

ary to delayed weight-bearing and a lack of normal joint forces which led to this relative abductor insufficiency. His theories are remarkably consistent with current thinking. Reimers J. The stability of the hip in children. A radiological study of the results of muscle surgery in cerebral palsy. Acta orthop Scand. 1980;184(Supp):1–100. This monograph summarised the state of the evidence known at the time and used the radiographic measure of hip displacement to document changes that occur in CP. The use of the ‘Migration Index’ has become the standard in hip surveillance in CP. Howard CB, McKibbin B, Williams LA, Mackie I.  Factors affecting the incidence of hip dislocation in cerebral palsy. JBJS-B. 1985;67(4):530–2. An observational study documented the incidence of hip displacement in the local population and noted the relationship between ambulatory status and incidence of hip dislocation. This was the first study to consider functional limitations as a contributing factor. Scrutton D, Baird G. Surveillance measures of the hips of children with bilateral cerebral palsy. Arch Dis Child. 1997;76(4):381–4. This was a population based study that highlighted the benefits of regular hip surveillance in early detection. The protocol for the identification of hip displacement included multi-­disciplinary clinical monitoring, documentation of motor milestones, and radiographic measurements; all of which now form the mainstay of hip surveillance. Miller F, Cardoso Dias R, Dabney KW, Lipton GE, Triana M.  Soft-tissue release for spastic hip subluxation in cerebral palsy. J Pediatr Orthop. 1997;17(5):571–84. This study looked at the outcome for the hip after soft tissue releases. It considered as important, the role of the spastic muscles in the etiology of hip displacement and suggested that soft tissue releases would be protective and prevent further hip displacement. This interpretation was limited however, by the duration of follow-up; only 39 months rather than to skeletal maturity. Mubarak SJ, Valencia FG, Wenger DR.  One-stage correction of the spastic dislocated hip. Use of pericapsular acetabuloplasty to improve coverage. JBJS-A. 1992;74(9):1347–57. The first study to describe

523

the comprehensive surgical management of the hip in CP, including soft tissue releases, femoral osteotomies, and acetabular reconstruction. The study’s 6 year follow up confirmed protection of the hips from further displacement by using this approach. Miller F, Dabney KW, Rang M. Complications in cerebral palsy treatment. In: Epps CH Jr, ed. Complications in pediatric orthopaedic surgery. Philadelphia, etc: JB Lippincott Company, 1995:pp 477–544. This classic work from Epps’ book was first to classify CP hip procedures into preventative (soft tissue), reconstructive (bony osteotomies), and salvage (femoral excision and interposition) categories. The philosophies outlined in this chapter are required reading to provide the basis for a sound rationale to the treatment of hip displacement in CP.

Key Evidence Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B.  Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39(4):214–23. This study reported the development of a standardized five level classification system for gross motor function in children with CP, now known as the GMFCS.  Their statement that the classification has application for clinical practice, research, teaching and administration has stood the test of time and has been subsequently shown to reliably predict the risk of hip displacement based on GMFCS level, in addition to other musculoskeletal manifestations, such as scoliosis. McNerney NP, Mubarak SJ, Wenger DR. One-stage correction of the dysplastic hip in cerebral palsy with the San Diego acetabuloplasty: results and complications in 104 hips. J Pediatr Orthop. 2000;20(1):93–103. The second study to come from the San Diego group detailing their step by step approach to deal with all aspects of hip pathology associated with displacement: the soft tissues including the capsule, the femoral deformity, and the change in acetabular shape. Results were assessed at a mean of 6.9 years with good outcomes.

524

Gordon GS, Simkiss DE.  A systematic review of the evidence for hip surveillance in children with cerebral palsy. JBJS-B. 2006;88(11):1492–6. This systematic review supports the routine use of hip surveillance for early detection of hip displacement. It also showed that hip surveillance programs have reduced the need for surgery for dislocated hips. Soo B, Howard JJ, Boyd RN, Reid SM, Lanigan A, Wolfe R, et al. Hip displacement in cerebral palsy. JBJS-A. 2006;88(1):121–9. A population based study showing the direct relationship between GMFCS level and incidence of hip displacement. This study helped facilitate the assessment of risk of displacement in a child with CP; allowing for: the design of hip surveillance screening programs, parental counselling with respect to prognosis, and the most appropriate allocation of resources. Hägglund G, Lauge-Pedersen H, Wagner P. Characteristics of children with hip displacement in cerebral palsy. BMC Musculoskelet Disord. 2007;8:101. Similar findings to that of Soo and colleagues [13] from an European population based study. Hip range of motion assessment to identify the presence or absence of adduction contractures was found to be a poor indicator of the risk of hip displacement. Recommendations were made for hip surveillance based on the child’s age and GMFCS level. Robin J, Graham HK, Selber P, Dobson F, Smith K, Baker R. Proximal femoral geometry in cerebral palsy: a population-based crosssectional study. JBJS-B. 2008;90(10):1372–9. This population-based study showed that there is a strong relationship between changes in proximal femoral shape (specifically, NSA and femoral anteversion) and hip displacement. The persistence of high anteversion and increases in coxa valga were shown to be directly related to GMFCS level. Gose S, Sakai T, Shibata T, Murase T, Yoshikawa H, Sugamoto K.  Morphometric analysis of the femur in cerebral palsy: 3-dimensional CT study. J Pediatr Orthop.

J. J. Howard et al.

2010;30(6):568–74. Cross-sectional imaging using CT confirmed the pathologic changes in the shape of the proximal femur and acetabulum and reinforced the direct relationship of these changes with the functional level as determined by the GMFCS. Wynte M, Gibson N, Kentish M, Love S, Thomason P, Graham HK.  The development of Australian Standards of Care for Hip Surveillance in Children with Cerebral Palsy: how did we reach consensus? J Pediatr Rehabil Med. 2011;4(3):171–82. This article described the process of consultation and review of the available evidence that led to the publication and acceptance of hip surveillance guidelines in Australia. Shore BJ, Yu X, Desai S, Selber P, Wolfe R, Graham HK.  Adductor surgery to prevent hip displacement in children with cerebral palsy: the predictive role of the Gross Motor Function Classification System. JBJS-A. 2012;94(4):326–34. In this population-based study, walking ability was found to be a good predictor of the surgical success of adductor releases for the treatment of hip displacement. Willoughby K, Jachno K, Ang SG, Thomason P, Graham HK. The impact of complementary and alternative medicine on hip development in children with cerebral palsy. Dev Med Child Neurol. 2013;55(5):472–9. A warning that delay in the institution of appropriate surgical treatment—in favour of unproven alternative therapies—may be deleterious to the outcome for hip displacement. Rutz E, Vavken P, Camathias C, Haase C, Junemann S, Brunner R.  Long-term results and outcome predictors in one-stage hip reconstruction in children with cerebral palsy. JBJS-A. 2015;97(6):500–6. At a mean follow-­up of 7.3  years, 168 patients underwent hip reconstructions, resulting in significant reductions in pain and clinical scores. The pre-­operative migration percentage was the only factor that had an impact on surgical outcomes.

18  The Hip in Cerebral Palsy

Shore BJ, Zurakowski D, Dufreny C, Powell D, Matheney TH, Snyder BD. Proximal Femoral Varus Derotation Osteotomy in Children with Cerebral Palsy: The Effect of Age, Gross Motor Function Classification System Level, and Surgeon Volume on Surgical Success. JBJS-A. 2015;97(24):2024– 31. This study with mean 8.3  year follow-up highlighted the importance of the factors such as older age, lower GMFCS, and increased surgeon volume, as strong predictors of surgical success. DiFazio R, Vessey JA, Miller P, Van Nostrand K, Snyder B.  Postoperative Complications After Hip Surgery in Patients With Cerebral Palsy: A Retrospective Matched Cohort Study. J Pediatr Orthop. 2016;36(1):56–62. This large study showed that medical complications after surgery were common and easier to treat than surgical ones. Pruszczynski B, Sees J, Miller F.  Risk Factors for Hip Displacement in Children With Cerebral Palsy: Systematic Review. J Pediatr Orthop. 2016;36(8):829–33. This systematic review confirmed that application of a practical surveillance program can prevent hip dislocation, allow for the provision of early treatment which leads to consistently better outcomes than for neglected hip dislocations. Wawrzuta J, Willoughby KL, Molesworth C, Ang SG, Shore BJ, Thomason P, et al. Hip health at skeletal maturity: a population-­based study of young adults with cerebral palsy. Dev Med Child Neurol. 2016;58(12):1273–80. Limited access to hip surveillance and delayed surgery (too little, too late) was likely to result in a hip that had poor morphology that was more likely to be painful. DiFazio R, Shore B, et  al. Effect of Hip Reconstructive Surgery on HealthRelated Quality of Life of Non-Ambulatory Children with Cerebral Palsy. JBJS-A. 2016:98(14):1190–8. The repeated assessment of health related quality of life (HRQoL) using the CP-CHILD measurement tool confirmed the effectiveness of hip reconstructive surgery for the treatment of displacement in non-ambulatory children.

525

Take Home Messages

• Hip displacement is common in children with CP and may lead to dislocation, pain, loss of function, and impaired HRQoL. • The risk of displacement can be predicted by the GMFCS and detected by regular radiographic examination (with MP as the primary radiographic measure) via a standardized hip surveillance program. • Hip displacement is related to both adductor contracture/muscle imbalance and proximal femoral deformity (femoral anteversion and coxa valga). • Adductor releases have a high success rate in ambulatory children with CP and a low success rate in non-ambulatory children. • Proximal femoral osteotomy (VDRO) is the key intervention for the prevention of hip displacement and as part of a comprehensive bony reconstruction for the surgical management of a subluxated or dislocated hip. • San Diego Acetabuloplasty (SDA) is an effective adjunct to hip reconstruction, in combination with adductor releases and VDROs. • Hip morphology is related to pain and HRQoL, as measured by the CPCHILD™. • Salvage surgery should be avoided because it has unpredictable clinical outcomes and high revision rates. • Total hip arthroplasty has the best reported outcomes of all the options for salvage surgery but may be applicable to only a small subset of young adults with CP, who are ambulatory.

References 1. Osler W. The cerebral palsies of children. Philadelphia, PA: Blakiston; 1889. 2. Little WJ.  On the nature and treatment of the deformities of the human frame: being a course of lectures

526 delivered at the Royal Orthopaedic Hospital in 1843. London: Longman, Brown, Green; 1853. 3. Mutch L, Alberman E, Hagberg B, Kodama K, Perat MV. Cerebral palsy epidemiology: where are we now and where are we going? Dev Med Child Neurol. 1992;34:547–51. 4. Rosenbaum P, et al. A report: the definition and classification of cerebral palsy. Dev Med Child Neurol Suppl. 2006;109:8–14. 5. Narayanan UG, Fehlings D, Weir S, Knights S, Kiran S, Campbell K.  Initial development and validation of the caregiver priorities and child health index of life with disabilities (CPCHILD). Dev Med Child Neurol. 2006;48:804–12. https://doi.org/10.1017/ s0012162206001745. 6. World Health Organization. International classification of functioning, disability, and health: children & youth version: ICF-CY.  Geneva: World Health Organization; 2007. 7. Grzegorzewski A, Jóźwiak M, Pawlak M, Modrzewski T, Buchcic P, Masłoń A. Hip joint pain in children with cerebral palsy and developmental dysplasia of the hip: why are the differences so huge? BMC Musculoskelet Disord. 2014;15:96. 8. Miller F, Dabney KW, Rang M.  Complications in cerebral palsy treatment. In: Epps Jr CH, editor. Complications in pediatric orthopaedic surgery. Philadelphia, PA: JB Lippincott Company; 1995. p. 477–544. 9. Carter DR, Tse B.  The pathogenesis of osteoarthritis in cerebral palsy. Dev Med Child Neurol. 2009;51(Suppl 4):79–83. https://doi. org/10.1111/j.1469-8749.2009.03435.x. 10. DiFazio R, Shore B, Vessey JA, Miller PE, Snyder BD.  Effect of hip reconstructive surgery on health-­ related quality of life of non-ambulatory children with cerebral palsy. J Bone Joint Surg Am. 2016;98: 1190–8. 11. Kolman SE, Ruzbarsky JJ, Spiegel DA, Baldwin KD. Salvage options in the cerebral palsy hip: a systematic review. J Pediatr Orthop. 2016;36:645–50. 12. Graham HK, Rosenbaum P, Paneth N, et al. Cerebral palsy. Nat Rev Dis Primers. 2015. https://doi. org/10.1038/nrdp.2015.82. 13. Soo B, Howard JJ, Graham HK, et  al. Hip dis placement in cerebral palsy. J Bone Joint Surg Am. 2006;88:121–9. 14. Cornell MS. The hip in cerebral palsy. Dev Med Child Neurol. 1995;37:3–18. 15. Flynn JM, Miller F. Management of hip disorders in patients with cerebral palsy. J Am Acad Orthop Surg. 2002;10:198–209. 16. Shore BJ, Yu X, Desai S, Selber P, Wolfe R, Graham HK.  Adductor surgery to prevent hip displacement in children with cerebral palsy: the predictive role of the gross motor function classification system. J Bone Joint Surg Am. 2012;94:326–34. https://doi. org/10.2106/jbjs.j.02003. 17. Howard JJ, Soo B, Graham HK, Boyd RN, Reid S, Lanigan A, Wolfe R, Reddihough DS. Cerebral palsy

J. J. Howard et al. in Victoria: motor types, topography and gross motor function. J Paediatr Child Health. 2005;41:479–83. 18. Bax MC.  Terminology and classification of cerebral palsy. Dev Med Child Neurol. 1964;6:295–7. 19. Graham HK, Selber P. Musculoskeletal aspects of cerebral palsy. J Bone Joint Surg Br. 2003;85-B:157–66. 20. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B.  Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39:214–23. 21. Howard CB, McKibbin B, Williams LA, Mackie I. Factors affecting the incidence of hip dislocation in cerebral palsy. J Bone Joint Surg Br. 1985;67:530–2. 22. Robin J, Graham HK, Selber P, Dobson F, Smith K, Baker R.  Proximal femoral geometry in cerebral palsy: a population-based cross-sectional study. J Bone Joint Surg Br. 2008;90-B:1372–9. 23. Henriksson L. Measurement of femoral neck anteversion and inclination: a radiographic study in children. Acta Orthop Scand Suppl. 1980;186:1–59. 24. Gose S, Sakai T, Shibata T, Murase T, Yoshikawa H, Sugamoto K.  Morphometric analysis of the femur in cerebral palsy: 3-dimensional CT study. J Pediatr Orthop. 2010;30:568–74. 25. Lapalza FJ, Root L.  Femoral anteversion and neck-­ shaft angles in hip instability in cerebral palsy. J Pediatr Orthop. 1994;14:719–23. 26. Schoenecker J.  Pathologic hip morphology in cerebral palsy and down syndrome. J Pediatr Orthop. 2013;33:S29–32. 27. Shefelbine SJ, Carter DR. Mechanobiological predictions of femoral anteversion in cerebral palsy. Ann Biomed Eng. 2004;32:297–305. 28. Tönnis D. Congenital dysplasia and dislocation of the hip. Berlin: Springer; 1987. 29. Lee W, Hsuan-Kai K, Yang W, Ho P, Chang C. Guided growth of the proximal femur for hip displacement in children with cerebral palsy. J Pediatr Orthop. 2016;36:511–5. 30. McCarthy JJ, Noonan KJ, Nemke BS, Markel M.  Guided growth of the proximal femur: a pilot study in the lamb model. J Pediatr Orthop. 2010;30: 690–4. 31. Portinaro N, Turati M, Cometto M, Bigoni M, Davids J, Panou A.  Guided growth of the proximal femur for the management of hip dysplasia in children with cerebral palsy. J Pediatr Orthop. 2017. https://doi. org/10.1097/BPO.0000000000001069. 32. Lee KM, Kang JY, Chung CY, Kwon DG, Lee SH, Choi IH, Cho T, Yoo WJ, Park MS. Clinical relevance of valgus deformity of proximal femur in cerebral palsy. J Pediatr Orthop. 2010;30:720–5. 33. Phelps WM.  Prevention of acquired dislocation of the hip in cerebral palsy. J Bone Joint Surg Am. 1959;41:440–8. 34. Tachdjian MO, Minear WL. Hip dislocation in cerebral palsy. J Bone Joint Surg Am. 1956;38(6):1358–64. 35. Abel MF, Wenger DR, Mubarak SJ, Sutherland DH. Quantitative analysis of hip dysplasia in cerebral

18  The Hip in Cerebral Palsy palsy: a study of radiographs and 3-D reformatted images. J Pediatr Orthop. 1994;14:283–9. 36. Beck M, Woo A, Leunig M, Ganz R. Gluteus minimus induced femoral head deformation in dysplasia of the hip. Acta Orthop Scand. 2001;72:13–7. 37. Graham HK, Narayanan UG. Salvage hip surgery in severe cerebral palsy: some answers, more questions? Bone Joint J. 2014;96-B:567–8. 38. Dobson F, Boyd RN, Parrott J, Nattrass GR, Graham HK. Hip surveillance in children with cerebral palsy: impact on the surgical management of spastic hip disease. J Bone Joint Surg Br. 2002;84-B:720–6. 39. Hägglund G, Andersson S, Düppe H, Lauge-Pedersen H, Nordmark E, Westbo L. Prevention of dislocation of the hip in children with cerebral palsy: the first ten years of a population-based prevention programme. J Bone Joint Surg Br. 2005;87-B:95–101. 40. Brunner R, Picard C, Robb J. Morphology of the acetabulum in hip dislocations caused by cerebral palsy. J Pediatr Orthop B. 1997;6:207–11. 41. Buckley SL, Sponseller PD, Magid D. The acetabulum in congenital and neuromuscular hip instability. J Pediatr Orthop. 1991;11:498–501. 42. Miller F, Slomczykowski M, Cope R, Lipton GE.  Computer modeling of the pathomechanics of spastic hip dislocation in children. J Pediatr Orthop. 1999;19:486–92. 43. Heinrich SD, MacEwen GD, Zembo MM. Hip dysplasia, subluxation, and dislocation in cerebral palsy: an arthrographic analysis. J Pediatr Orthop. 1991;11:488–93. 44. Chung CY, Park MS, Choi IH, Cho TJ, Yoo WJ, Lee KM.  Morphometric analysis of acetabular dysplasia in cerebral palsy. J Bone Joint Surg Br. 2006;88-B:243–7. 45. Chung MK, Zulkarnain A, Lee JB, Cho BC, Chung CY, Lee KY, Sung KH, Park MS.  Functional status and amount of hip displacement independently affect acetabular dysplasia in cerebral palsy. Dev Med Child Neurol. 2017;59:743–9. 46. Kim HT, Wenger DR.  Location of acetabular deficiency and associated hip dislocation in neuromuscular hip dysplasia: three-dimensional computed tomographic analysis. J Pediatr Orthop. 1997;17:143–51. 47. Chung CY, Choi IH, Cho TJ, Yoo WJ, Lee SH, Park MS.  Morphometric changes in the acetabulum after Dega osteotomy in patients with cerebral palsy. J Bone Joint Surg Br. 2008;90-B:88–91. 48. Karlen JW, Skaggs DL, Ramachandran M, Kay RM.  The Dega osteotomy: a versatile osteotomy in the treatment of developmental and neuromuscular hip pathology. J Pediatr Orthop. 2009;29:676–82. 49. Gugenheim JJ, Gerson LP, Sadler C, Tullos HS. Pathologic morphology in paralytic and congenital hip instability. J Pediatr Orthop. 1982;2:397–400. 50. Granzier HL, Labeit S. The giant muscle protein titin is an adjustable molecular spring. Exerc Sport Sci Rev. 2006;34:50–3. 51. Smith LR, Lee KS, Ward SR, Chambers HG, Lieber RL.  Hamstring contractures in children with spastic cerebral palsy result from a stiffer extracellular matrix

527 and increased in  vivo sarcomere length. J Physiol. 2011;589:2625–39. 52. Larkin-Kaiser KA, Joumaa V, Leonard T, Howard JJ, Herzog W.  Larger isoforms of titin are associated with increased sarcomere lengths and severity of hip displacement in cerebral palsy. In: Proceedings of the International Society of Biomechanics; 2015; Glasgow, UK. https://isbweb.org/images/conferences/isbcongresses/2015/isb_2015_abstract_book_final.pdf. 53. Gordon AM, Huxley AF, Julian FJ. Tension development in highly stretched vertebrate muscle fibres. J Physiol. 1966;184(1):143–69. 54. Leonard TR, Howard JJ, Larkin-Kaiser KA, Joumaa V, Gauthier L, Logan K, Orlik B, El-Hawary R, Herzog W. Stiffness of Hip adductor myofibrils is decreased for children with spastic cerebral palsy. J Biomech. 2019;87:100–106. https://doi. org/10.1016/j.jbiomech.2019.02.023. 55. Ramstada K, Terjesen T. Hip pain is more frequent in severe hip displacement: a population-based study of 77 children with cerebral palsy. J Pediatr Orthop B. 2016;25:217–21. 56. Wawrzuta J, Willoughby KL, Molesworth C, Ang SG, Shore BJ, Thomason P, Graham HK.  Hip health at skeletal maturity: a population-based study of young adults with cerebral palsy. Dev Med Child Neurol. 2016;58:1273–80. 57. Masłon A, Marek Józwiak M, Pawlak M, Modrzewski T, Grzegorzewski A.  Hip joint pain in spastic dislocation: aetiological aspects. Dev Med Child Neurol. 2011;53:1019–23. 58. Valencia FG. Management of hip deformities in cerebral palsy. Orthop Clin N Am. 2010;41:549–59. 59. Smith RL, Thomas KD, Schurman DJ, Carter DR, Wong M, van der Meulen MC. Rabbit knee immobilization: bone remodeling precedes cartilage degradation. J Orthop Res. 1992;10:88–95. 60. Lundy DW, Ganey TM, Ogden JA, Guidera KJ.  Pathologic morphology of the dislocated proximal femur in children with cerebral palsy. J Pediatr Orthop. 1998;18:528–34. 61. Viehweger E.  Importance of hip problems in daily activities for cerebral palsy patients. J Child Orthop. 2013;7:401–6. 62. O’Brien JJ, Sirkin RB. The natural history of the dislocated hip in cerebral palsy. Dev Med Child Neurol. 1978;20:241. 63. Pritchett JW. The untreated unstable hip in severe cerebral palsy. Clin Orthop Relat Res. 1983;173:169–72. 64. Letts M, Shapiro L, Mulder K, Klassen O. The windblown of children with hip displacement in cerebral palsy. J Pediatr Orthop. 1984;4:55–62. 65. Samilson RL, Tsou P, Aamoth G, Green W. Dislocation and subluxation of the hip in cerebral palsy. J Bone Joint Surg Am. 1972;54:863–73. 66. Difazio RL, Vessey JA, Zurakowski D, Snyder BD.  Differences in health-related quality of life and caregiver burden after hip and spine surgery in non-­ ambulatory children with severe cerebral palsy. Dev Med Child Neurol. 2016;58:298–305.

528 67. Jung NH, Pereira B, Nehring I, Brix O, Bernius P, Schroeder SA, Kluger GA, Koehler T, Beyerlein A, Weir S, von Kries R, Narayanan UG, Berweck S, Mall V.  Does hip displacement influence health-­ related quality of life in children with cerebral palsy? Dev Neurorehabil. 2014;17(6):420–5. 68. Penner M, Xie WY, Binepal N, Switzer L, Fehlings D. Characteristics of pain in children and youth with cerebral palsy. Pediatrics. 2013;132:e407–13. 69. Terjesen T.  The natural history of hip develop ment in cerebral palsy. Dev Med Child Neurol. 2012;54:951–7. 70. Ramstad K, Jahnsen RB, Terjesen T. Severe hip displacement reduces health-related quality of life in children with cerebral palsy: a population-­based study of 67 children. Acta Orthop. 2017;88(2):205–10. 71. DiFazio R, Vessey JA, Miller P, Van Nostrand K, Snyder B.  Postoperative complications after hip surgery in patients with cerebral palsy: a retrospective matched cohort study. J Pediatr Orthop 2016;36:56–62. 72. Bagg MR, Farber J, Miller F. Long-term follow-­up of hip subluxation in cerebral palsy patients. J Pediatr Orthop. 1993;13:32–6. 73. Westbom L, Hagglund G, Nordmark E. Cerebral palsy in a total population of 4–11 year olds in southern Sweden. Prevalence and distribution according to different CP classification systems. BMC Pediatr. 2007;7:41–9. 74. Graham HK.  Classifying cerebral palsy. J Pediatr Orthop. 2005;25:127–8. 75. Larnert P, Risto O, Hagglund G, Wagner P. Hip displacement in relation to age and gross motor function in children with cerebral palsy. J Child Orthop. 2014;8:129–34. 76. Pruszczynski B, Sees J, Miller F. Risk factors for hip displacement in children with cerebral palsy: systematic review. J Pediatr Orthop. 2016;36:829–33. 77. Elkamil AI, Andersen GL, Hägglund G, Lamvik T, Skranes J, Vik T. Prevalence of hip dislocation among children with cerebral palsy in regions with and without a surveillance programme: a cross sectional study in Sweden and Norway. BMC Musculoskelet Disord. 2011;12:284–95. 78. Reimers J. The stability of the hip in children. A radiological study of the results of muscle surgery in cerebral palsy. Acta Orthop Scand 1980;184(Supp):1–100. 79. Shore B, Spence D, Graham HK.  The role for hip surveillance in children with cerebral palsy. Curr Rev Musculoskelet Med. 2012;5:126–13. 80. Wynter M, Gibson N, Kentish M, et al. The consensus statement on hip surveillance for children with cerebral palsy: Australian standards of care. J Pediatr Rehabil Med. 2011;4:183–95. 81. Yam WK, Leung MS.  Interrater reliability of modified Ashworth scale and modified Tardieu scale in children with spastic cerebral palsy. J Child Neurol. 2006;21:1031–5. 82. Ruwe PA, Gage JR, Ozonoff MB, DeLuca PA. Clinical determination of femoral anteversion: a comparison with established techniques. J Bone Joint Surg Am. 1992;74-A:820–30.

J. J. Howard et al. 83. Tamari K, Tinley P, Briffa K, Breidahl W.  Validity and reliability of existing and modified clinical methods of measuring femoral and tibiofibular torsion in healthy subjects: use of different reference axes may improve reliability. Clin Anat. 2005;18:46–55. 84. Parrott J, Boyd RN, Dobson F, Lancaster A, Love S, Oates J, Wolfe R, Nattrass GR, Graham HK. Hip displacement in spastic cerebral palsy: repeatability of radiologic measurement. J Pediatr Orthop. 2002;22:660–7. 85. Pons C, Remy-Neris O, Medee B, Brochard S.  Validity and reliability of radiological methods to assess proximal hip geometry in children with cerebral palsy: a systematic review. Dev Med Child Neurol. 2013;55:1089–102. 86. Robin J, Graham HK, Baker R, et al. A classification system for hip disease in cerebral palsy. Dev Med Child Neurol. 2009;51:183–92. 87. Faraj S, Atherton WG, Stott NS.  Inter- and intra-­ measurer error in the measurement of Reimers’ hip migration percentage. J Bone Joint Surg. 2004;86-B:434–7. 88. Brunner R, Robb JE. Inaccuracy of the migration percentage and center-edge angle in predicting femoral head displacement in cerebral palsy. J Pediatr Orthop. 1996;5:239–41. 89. Selva G, Miller F, Dabney K.  Anterior hip dislocation in children with cerebral palsy. J Pediatr Orthop. 1998;18:54–61. 90. Hilgenreiner H. Zur [early diagnosis and early treatment of congenital hip dislocation] (in German). Med Klin. 1925;21:1385–429. 91. Chung CY, Lee KY, Park MS, Lee SH, Choi IH, Cho TJ. Validity and reliability of measuring femoral anteversion and neck-shaft angle in patients with cerebral palsy. J Bone Joint Surg Am. 2010;92:1195–205. 92. Foroohar A, McCarthy JJ, Yucha D, Clarke S, Brey J.  Head-shaft angle measurement in children with cerebral palsy. J Pediatr Orthop. 2009;29(3):248–50. 93. Chougule S, Dabis J, Petrie A, Daly K, Gelfer Y.  Is head–shaft angle a valuable continuous risk factor for hip migration in cerebral palsy? J Child Orthop. 2016;10:651–6. 94. Burns F, Stewart R, Reddihough D, Scheinberg A, Ooi K, Graham HK.  The cerebral palsy transition clinic: administrative chore, clinical responsibility, or opportunity for audit and clinical research? J Child Orthop. 2014;8:203–13. 95. Ward WT, Vogt M, Grucziak JS, Tumer Y, Cook C, Fitch RD.  Severin classification system for evaluation of the results of operative treatment of congenital dislocation of the hip. A study of intraobserver and interobserver reliability. J Bone Joint Surg Am. 1997;79:656–63. 96. Murnaghan ML, Simpson P, Robin JG, Shore BJ, Selber P, Graham HK. The cerebral palsy hip classification is reliable: an inter- and intra-­observer reliability study. J Bone Joint Surg Br. 2010;92-B:436–41. 97. Kim YJ, Jaramillo D, Millis MB, et  al. Assessment of early osteoarthritis in hip dysplasia with delayed

18  The Hip in Cerebral Palsy gadolinium-enhanced magnetic resonance imaging of cartilage. J Bone Joint Surg Am. 2003;85-A:1987–92. 98. Shore BJ, Powell D, Miller PE, Matheney TH, Snyder BD.  Acetabular and femoral remodeling after varus derotational osteotomy in cerebral palsy: the effect of age and gross motor function classification level. J Pediatr Orthop B. 2016;25:322–30. 99. Gough M.  Continuous postural management and the prevention of deformity in children with cerebral palsy: an appraisal. Dev Med Child Neurol. 2009;51:105–10. 100. Miller SD, Juricic M, Hesketh K, McLean L, Magnuson S, Gasior S, Schaeffer E, O’Donnell M, Mulpuri K. Prevention of hip displacement in children with cerebral palsy: a systematic review. Dev Med Child Neurol. 2017;59:1130–9. 101. Novak I, McIntyre S, Morgan C, Campbell L, Dark L, Morton N, Stumbles E, Wilson S, Goldsmith S. A systematic review of interventions for children with cerebral palsy: state of the evidence. Dev Med Child Neurol. 2013;55:885–910. 102. Graham HK, Boyd R, Carlin JB, et al. Does botulinum toxin a combined with bracing prevent hip displacement in children with cerebral palsy and “hips at risk”? A randomized, controlled trial. J Bone Joint Surg Am. 2008;90:23–33. 103. Willoughby K, Ang SG, Thomason P, Graham HK. The impact of botulinum toxin a and abduction bracing on long-term development in children with cerebral palsy. Dev Med Child Neurol. 2012;54:743–7. 104. Lundy C, Doherty GM, Fairhurst CB.  Botulinum toxin type A injections can be an effective treatment for pain in children with hip spasms and cerebral palsy. Dev Med Child Neurol. 2009;51:705–10. 105. Miller F, Giraedi H, Lipton G, Ponzio R, Klaumann M, Dabney KW.  Reconstruction of the dysplastic spastic hip with peri-iliac pelvic and femoral osteotomy followed by immediate mobilization. J Pediatr Orthop. 1995;15:672–6. 106. Rutz E, Vavken P, Camathias C, Haase C, Jünemann S, Brunner R.  Long-term results and outcome predictors in one-stage hip reconstruction in children with cerebral palsy. J Bone Joint Surg Am. 2015;97:500–6. 107. Winters TF, Gage JR, Hicks R. Gait patterns in spastic hemiplegia in children and young adults. J Bone Joint Surg Am. 1987;69:437–41. 108. Rutz E, Passmore E, Baker R, Graham HK. Multilevel surgery improves gait in spastic hemiplegia but does not resolve hip dysplasia. Clin Orthop Relat Res. 2012;470:1294–302. 109. Khot A, Sloan S, Desai S, Harvey A, Wolfe R, Graham HK.  Adductor release and chemodenervation in children with cerebral palsy: a pilot study in 16 children. J Child Orthop. 2008;2:293–9. 110. Bialik G, Pierce R, Dorociak R, Lee TS, Aiona MD, Sussman MD.  Iliopsoas Tenotomy at the lesser trochanter versus at the pelvic brim in ambulatory children with cerebral palsy. J Pediatr Orthop. 2009;29:251–5.

529 111. Skaggs DL, Kaminsky CK, Eskander-Rickards E, Reynolds RAK, Tolo VT, Bassett GS.  Psoas over the brim lengthenings: anatomic investigation and surgical technique. Clin Orthop Relat Res. 1997;339:174–9. 112. Rutz E, Brunner R. The pediatric LCP hip plate for fixation of proximal femoral osteotomy in cerebral palsy and severe osteoporosis. J Pediatr Orthop. 2010;30:726–31. 113. Zhou L, Camp M, Gahukamble A, Khot A, Graham HK.  Cannulated, locking blade plates for proximal femoral osteotomy in children and adolescents. J Child Orthop. 2015;9:121–7. 114. Canavese F, Emara K, Sembrano JK, Bialik V, Aiona MD, Sussman MD.  Varus Derotation osteotomy for the treatment of hip subluxation and dislocation in GMFCS level III to V patients with unilateral hip involvement. Follow-up at skeletal maturity. J Pediatr Orthop. 2010;30:357–64. 115. Shore BJ, Zurakowski D, Dufreny C, Powell D, Matheney TH, Snyder BD.  Proximal femoral Varus Derotation osteotomy in children with cerebral palsy: the effect of age, gross motor function classification system level, and surgeon volume on surgical success. J Bone Joint Surg Am. 2015;97: 2024–31. 116. Ruzbarsky JJ, Beck NA, Baldwin KD, Sankar WN, Flynn JM, Spiegel DA. Risk factors and complications in hip reconstruction for nonambulatory patients with cerebral palsy. J Child Orthop. 2013;7:487–500. 117. McGillion S, Clarke NM.  Lateral growth arrest of the proximal femoral physis: a new technique for serial radiological observation. J Child Orthop. 2011;5:201–7. 118. Torode IP, Young JL. Caput valgum associated with developmental dysplasia of the hip: management by transphyseal screw fixation. J Child Orthop. 2015;9:371–9. 119. McNerney NP, Mubarak SJ, Wenger D.  One-stage correction of the dysplastic hip in cerebral palsy with the San Diego Acetabuloplasty: results and complications in 104 hips. J Pediatr Orthop. 2000;20:93. 120. Mubarak SJ, Valencia FG, Wenger DR.  One-stage correction of the spastic dislocated hip: use of pericapsular acetabuloplasty to improve coverage. J Bone Joint Surg Am. 1992;74:1347–57. 121. Salter RB. Innominate osteotomy in the treatment of congenital dislocation and subluxation of the hip. J Bone Joint Surg Br. 1961;43:518–39. 122. Stasikelis PJ, Ridgeway SR, Pugh LI, Allen BL. Epiphyseal changes after proximal femoral osteotomy. J Pediatr Orthop B. 2001;10:25–9. 123. Hesketh K, Leveille L, Mulpuri K.  The frequency of AVN following reconstructive hip surgery in children with cerebral palsy: a systematic review. J Pediatr Orthop. 2016;36:e17–24. 124. Dega W.  Transiliac osteotomy in the treatment of congenital hip dysplasia [polish]. Chir Narz Ruchu Ortop Pol. 1974;39:601–13.

530 125. Dietz FR, Knutson LM. Chiari pelvic osteotomy in cerebral palsy. J Pediatr Orthop. 1995;15:372–80. 126. Zuckerman JD, Staheli LT, McLaughlin JF. Acetabular augmentation for progressive hip subluxation in cerebral palsy. J Pediatr Orthop. 1984;4:436–42. 127. Sankar WN.  Complete Redirectional Acetabular osteotomies for neurogenic and Syndromic hip dysplasia. J Pediatr Orthop. 2013;33:S39–44. 128. MacDonald SJ, Hersche O, Ganz R. Periacetabular osteotomy in the treatment of neurogenic acetabular dysplasia. J Bone Joint Surg Br. 1999;81:975–8. 129. Rebello G, Zilkens C, Dudda M, Matheney T, Kim YJ. Triple pelvic osteotomy in complex hip dysplasia seen in neuromuscular and teratologic conditions. J Pediatr Orthop. 2009;29(6):527–34. 130. Wright PB, Ruder J, Birnbaum MA, Phillips JH, Herrera-Soto JA, Knapp DR. Outcomes after salvage procedures for the painful dislocated hip in cerebral palsy. J Pediatr Orthop. 2013;33:505–10. 131. Girdlestone GR. Acute pyogenic arthritis of the hip. An operation giving free access and effective drainage. Lancet. 1943;1:419–21. 132. Baxter MP, D’Astous JL.  Proximal femoral resection-­ interposition salvage hip arthroplasty: surgery for the severely disabled child with cerebral palsy. J Pediatr Orthop. 1986;6:681–5. 133. Koffman M.  Proximal femoral resection or total hip replacement in severely disabled cerebralspastic patients. Orthop Clin North Am. 1981;12: 91–100. 134. Leet AI, Chor K, Launay F, Kier York J, Sponseller PD.  Femoral head resection for painful hip subluxation in cerebral palsy: is valgus osteotomy in conjunction with femoral head resection preferable to proximal femoral head resection and traction? J Pediatr Orthop. 2005;25:70–3. 135. Root L. Surgical treatment for hip pain in the adult cerebral palsy patient. Dev Med Child Neurol. 2009;51(Suppl):84–91. 136. Castle ME, Schneider C. Proximal femoral resection-­ interposition arthroplasty. J Bone Joint Surg Am. 1978;60-A:1051–4. 137. McCarthy RE, Simon S, Douglas B, Zawacki R, Reese N. Proximal femoral resection to allow adults who have severe cerebral palsy to sit. J Bone Joint Surg Am. 1988;70-A:1011–6.

J. J. Howard et al. 138. Patel NK, Sabharwal S, Gooding CR, Hashemi-­ Nejad A, Eastwood DM.  Proximal femoral excision with interposition myoplasty for cerebral palsy patients with painful chronic hip dislocation. J Child Orthop. 2015;9:263–71. 139. Egermann M, Döderlein L, Schläger E, Müller S, Braatz F.  Autologous capping during resection arthroplasty of the hip in patients with cerebral palsy. J Bone Joint Surg Br. 2009;91-B:1007–12. 140. Gabos PG, Miller F, Galban MA, Gupta GG, Dabney K.  Prosthetic interposition arthroplasty for the palliative treatment of end-stage spastic hip disease in nonambulatory patients with cerebral palsy. J Pediatr Orthop. 1999;19:796–804. 141. Schanz A.  Zur Behandlung der veralteten angeborenen Hüftverrenkung. Z Orthop. 1921;42:442–4. 142. McHale KA, Bagg M, Nason SS.  Treatment of the chronically dislocated hip in adolescents with cerebral palsy with femoral head resection and subtrochanteric valgus osteotomy. J Pediatr Orthop. 1990;10:504–9. 143. Root L, Goss JR, Mendes J. The treatment of the painful hip in cerebral palsy by total hip replacement or hip arthrodesis. J Bone Joint Surg. 1986;68A:590–7. 144. Boldingh EJ, Bouwhuis CB, van der Heijden-­ Maessen HC, Bos CF, Lankhorst GJ. Palliative hip surgery in severe cerebral palsy: a systematic review. J Pediatr Orthop B 2014;23:86–92. 145. de Souza RC, Mansano MV, Bovo M, Yamada HH, Rancan DR, de Moraes Barros Fucs PM, Svartman C, César de Assumpção RM. Hip salvage surgery in cerebral palsy cases: a systematic review. Rev Bras Ortop 2015;50:254–259. 146. Howell K, Selber P, Graham HK, Reddihough D.  Botulinum neurotoxin A: an unusual systemic effect. J Paediatr Child Health. 2007;43:499–501. 147. Barney C, Merbler A, Frenn K, Stansbury J, Krach L, Partington M, Graupman P, Kim P, Song D, Symons FA. Prospective study of pain pre- and post-­ intrathecal baclofen pump implant in children with cerebral palsy. Dev Med Child Neurol. 2017;59:98. 148. Dartnell JL, Paterson JM, Magill N, Norman-Taylor F.  Proximal femoral resection for the painful dis-­ located hip in cerebral palsy: does indomethacin prevent heterotopic ossification? J Pediatr Orthop. 2014;34:295–9.

The Hip in Myelomeningocele

19

Emmanouil Morakis, Jason J. Howard, and James Wright

Introduction Myelomeningocele (MMC) is a congenital malformation that belongs to the family of neural tube defects. These include anencephaly, exencephaly, encephaloceles and meningoceles. MMC results from incomplete closure of the neural tube around the fifth and sixth weeks of gestation. It is the most common among the neural tube defects (NTD). MMC is characterized by a failure of formation of the dorsal vertebral elements, a defect of the overlying skin with exposure of the meninges and spinal cord. It results in bowel, bladder, motor and sensory paralysis distal to the level of the malformation. Musculoskeletal manifestations associated with MMC are common; often resulting in significant functional impairments relating to gait abnormalities, seating imbalance, and skin ulceration. The most common surgically treated deformities include talipes equinovarus, knee

flexion contractures, and occasionally scoliosis and kyphosis. Hip dislocation is also common and was treated aggressively in the past due to a then commonly held view that reducing these hips would lead to better outcomes. Over the last 20  years, this view has been challenged, with best evidence suggesting that the risks of treatment for hip dislocation in spina bifida far outweigh the benefits (Fig.  19.1) [1, 2]. This chapter will review the evidence relating to the treatment of the hip in spina bifida, emphasizing a measured approach when addressing the associated deformities, particularly with respect to dislocation.

E. Morakis (*) Royal Manchester Children’s Hospital, Manchester, UK e-mail: [email protected] J. J. Howard Weill Cornell Medicine, Chief of Orthopaedic Surgery, Sidra Medicine, Doha, Qatar J. Wright Oxford University Hospitals NHS Foundation Trust, Oxford, UK e-mail: [email protected] © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_19

Fig. 19.1  Eleven year-old girl with L3 myelomeningocele and bilateral hip dislocations. She had pain-free, mobile hips which allowed her to utilize a walker for short distances following reconstruction of her rigid talipes equinovarus deformities. The hips were not an impediment to her ambulation. (Courtesy Jason J. Howard) 531

E. Morakis et al.

532

Pathophysiology Embryology The embryologic process that forms the brain and most of the spinal cord is called primary neurulation. It occurs between 17 and 27 days after conception. The neural plate, a plate of thickened ectoderm, forms at the beginning of the third week of embryonic life. The lateral edges of the neural plate elevate to form the neural folds. The neural folds fuse and form the neural tube, starting at the cervicomedullary junction and extending in both directions. By the 28th day after conception the caudal neuropore closes. MMC results either from ­failure of the neural tube to close or from a later pathologic reopening of the neural tube.

Neuropathology Nicholaas Tulp (1593–1674), a Dutch physician and anatomist, first illustrated in modern history in his textbook entitled “Observationes Medicae”, a case of spinal dysraphism [3]. He described a case of a child with a large lumbar MMC. He was first to use the term spina bifida. This term is still used, commonly referring to the entire spectrum of neural tube defects, with the exception of spina bifida occulta. MMC lesions are formed by combined malformations of the vertebral column and the spinal cord. The lesion consists of a vertebral column bony defect through which the meningeal membranes that cover the spinal cord—and part of the spinal cord—protrude. It most often involves the posterior lumbosacral region. The neural placode (i.e. malformed spinal cord) is a flat plaque of neural tissue merging into the malformed meningeal coverings. Both ventral and dorsal spinal roots exit from the ventral aspect of the placode. There is an intact subarachnoid space ventral to the placode. The placode is covered with pia and arachnoid without any dura or skin overlying. Under the placode protrudes a membranous sac containing meninges, cerebrospinal fluid and nerve roots. At the lateral edges, the lesion is attached to the dysplastic meninges and skin.

At birth, neurologic deficits are ubiquitous in patients with MMC. The neural elements are damaged primarily by the neural tube malformation and secondarily by the exposure to the amniotic fluid in utero, mechanical trauma during passage through the birth canal, a lack of vascular support, or scarring from surgical closure of the lesion. The associated neurologic deficits can include sensory and motor paralysis of the trunk and lower limbs, as well as urinary and fecal incontinence. Often, there are other conditions that can have an effect on the neurological status of patients with MMC.  These include hydrocephalus, a Chiari II malformation, a syrinx or a diastematomyelia. These can cause brainstem dysfunction with strabismus, swallowing difficulties, vocal cord paresis and apnoeic episodes. Severe forms of the Chiari malformation can cause weakness of the upper extremities and indeed most children with spina bifida have some level of upper limb dysfunction.

 eurosegmental Levels in Spina N Bifida Children with MMC present with motor and sensory paralysis in the lower limbs. The extent of involvement depends on the level of the lesion. Patients have been traditionally classified into neurosegmental levels based on the clinically determined strength of specific muscle groups. The neurologic level of involvement is based on the lowest level of antigravity strength on the patients’ best site [4, 5] (Table 19.1). Table 19.1  Neurosegmental levels Definition of neurological level Level Function Thoracic No antigravity strength in muscles of lower extremities L1–L2 Hip flexion or adduction L3 Knee extension L4 Knee flexion L5 Ankle dorsiflexion Sacral Ankle plantarflexion

19  The Hip in Myelomeningocele

The neurosegmental level does not always correspond though to the anatomic level of the lesion. The level of involvement can be asymmetric and the motor and sensory deficit can correspond to different neurosegmental levels [6]. The neurologic impairment can be the result of upper and/or lower motor neuron dysfunction usually presenting with flail or, occasionally, spastic paralysis [7]. Sharrard described the different levels of neurosegmental involvement in children with MMC using information from electrical stimulation of the nerve roots during surgical closure of the spinal defect and physical assessment of patients with MMC [5]. The International Myelodysplasia Study Group described more standardized criteria for assigning levels of neurological involvement (Table 19.2) [8].

533

Despite this sophisticated classification system, more practically, children with MMC can be grossly classified into four groups based on their neurosegmental level of involvement: thoracic, upper lumbar (L1–2), lower lumbar (L3–5) and sacral. These more simplified groupings are perhaps better reflective of what is seen clinically, with the function of few children with MMC being accurately defined by the motor distribution of a single nerve root level (e.g. L3). “Children with MMC can be grossly classified into four groups based on their neurosegmental level of involvement: thoracic, upper lumbar (L1-2), lower lumbar (L3-5) and sacral”. Children with a thoracic level of involvement have no muscle activity in their lower extremities.

Table 19.2  International Myelodysplasia Study Group criteria for assigning neurosegmental levels T10 or above T11a T12 L1 L1–2a L2 L3 L3–4a L4 L4–5a L5

L5–S1a S1

S1–2a S2 S2–3 No loss L1–3 L2–4

Determine based on sensory level and/or palpation of abdominals

Some pelvic control in sitting or when supine, which may originate in the abdominals or back; hip hiking may be noted Weak iliopsoas (grade 2) lliopsoas, sartorius, adductors all grade 3 or higher Meet criteria for L2 and quadriceps are grade 3 or higher Meet criteria for L3 and have: medial hamstrings grade 3 or higher; anterior tibialis grade 3 or higher (may also have weak peroneus tertius) Meet criteria for L4 and have lateral hamstrings grade 3 or higher plus one of the following:  Gluteus medius grade 2 or higher  Peroneus tertius grade 4 or higher  Posterior tibialis grade 3 or higher Meet criteria for L5 plus two of the following:  Gastrocnemius-soleus complex grade 2 or higher  Gluteus medius grade 3 or higher  Gluteus maximus grade 2 or higher (puckering of buttocks) Meet criteria for S1 and gastrocnemius-soleus complex grade 3 or higher and gluteus medius and maximus grade 4 or higher All muscles in lower extremities of normal strength, though may be grade 4 in one or two muscle groups; also includes normal-looking infants too young to be bowel and bladder trained Meets criteria for S2–3 and has no bowel/bladder loss Weak hip flexors, adductors, and weak quadriceps (weak innervation through L3) Strong hip flexors, but weak knee flexion and extension

Assign to this level when the child exceeds the criteria for the preceding level but does not meet the criteria for next lower level

a

534

They often develop scoliosis, hip dislocation and fixed contractures of their knees and hips. These children often manage to become therapeutic or household ambulators with the help of extensive bracing and walking aids. Most, however, lose this ability as adolescents or young adults and rely on a wheelchair for their mobility [9–12]. Children with an upper lumbar level of involvement present with intact flexor power of the iliopsoas and hip adductors. As such, they usually develop hip and knee flexor contractures. Similar to the previous group, extensive bracing is required for ambulation and many of these children lose their ability of functional ambulation as adults [4, 9, 13–15]. Children with a lower lumbar level of involvement have retained quadriceps strength which allows for active knee extension. Most are functional ambulators, using orthotics to help them control their foot and ankle position. They often use a wheelchair for longer distances as an adult [11, 16]. Many of these patients develop joint (hip, knee) contractures and limb deformities (external tibial torsion, ankle valgus, foot deformities) that may require surgical management to help them maintain their functional walking ability. Finally, most children with a sacral level of involvement can walk without assistive devices. They may need the help of orthotics to avoid crouch gait (weak gastrosoleus). They may walk with an abductor lurch (increased active pelvic rotation with stance phase hip abduction). Most of them maintain their independent ambulation as adults [4, 17]. Associated foot deformities, knee flexion contractures, hip displacement, femur and/or tibia torsional deformities, and crouch gait may need to be addressed to allow them to remain independently mobile. Pressure sores in the foot and ankle, musculoskeletal infections and a tethered cord can contribute to a loss of ambulatory capacity in these children [18].

Etiology The etiology of spina bifida is multifactorial in most patients. MMC usually presents as an isolated malformation, but can be part of syndromic

E. Morakis et al.

conditions (Meckel-Gruber, Roberts, Jarcho-­ Levin, HARD syndrome) or chromosomal anomalies (trisomy 13 or 18). Genetic and environmental factors play a role in the pathogenesis of neural tube defects (NTD) and MMC.  Studies have indicated a higher risk of NTD in a family of a person with a NTD [19–21]. The risk is higher among first and second degree relatives. The risk is 1 in 20–30 times higher than the general population in subsequent pregnancies. Mothers with children born with NTD were more likely to carry polymorphic variants in genes of the folate and homocysteine pathways (C677T variant of MTHFR gene, VANGL1 gene) [22]. Other risk factors identified include: pesticide exposure, nitrosatable drugs, clomiphene, obesity and poorly controlled pregestational diabetes mellitus [23–26]. Hyperthermia during the first trimester of pregnancy has been correlated with increased risk for NTD [27]. Epidemiological studies in the United Kingdom in the 1960s demonstrated an increased incidence of NTD in deprived areas. Mothers that gave birth to babies with NTD had lower levels of several vitamins [28, 29]. Since then, several studies indicated that a significant percentage of NTD are caused by folate deficiency, in combination with genetic and environmental factors. Women that gave birth to children with NTD had lower concentrations of folate in their blood [30]. The use of folic acid antagonists increases the risk of NTD [31]. Randomized control studies have demonstrated that folate supplementation reduces the risk of NTD [32, 33]. Because folate should be initiated before conception, in 1998 the FDA has mandated fortification with folic acid of all cereal and grain products. This lead to a 23% decrease of NTD birth prevalence between 1995–1996 and 2003–2004. It is estimated that 50–70% of NTD are folic acid preventable [33]. In addition, all women of reproductive potential are advised to take 0.4–0.8 mg of folic acid daily [34]. “It is estimated that 50-70% of neural tube defects are folic acid preventable”.

19  The Hip in Myelomeningocele

Pathophysiology of Hip Deformities Several deformities of the hip can be seen in patients with MMC.  These include: teratogenic hip dislocation present at birth, progressive paralytic subluxation and dislocation, and soft tissue contractures (adduction, abduction, external rotation, flexion). The etiology and pathophysiology behind the development of these deformities is multifactorial and is, to a certain extent, unknown. Muscle imbalance was hypothesized to be a driving force behind the development of paralytic hip subluxation/dislocation first by Sharrard [5]. This muscle imbalance can be a result of a lower motor neuron flail paralysis, an upper motor neuron spastic paralysis or a combination of both. The unopposed hip flexor and adductor muscle function was hypothesized to gradually lead to a flexion, adduction, and lateral rotation deformity of the hip. Muscle imbalance cannot completely explain the development of hip instability and hip contractures in these patients (see section Operative Management). Patients with thoracic lesions have flail paralysis of all hip muscles and despite that they often develop hip contractures and hip dislocation [1, 35]. “Muscle imbalance cannot completely explain the development of hip instability and hip contractures in these patients”. Muscle imbalance is also thought to be involved in the development of hip contractures. With the exception of children at the sacral level, the physiologic hip flexion contracture of the newborn fails to resolve due to paralysis of the hip extensors [35, 36]. The contracted iliopsoas may be “bowstrung” across the femoral head, causing posterior and lateral displacement of the femoral head and deformity (flattening, formation of groove) [36]. Patients with thoracic or upper lumbar lesions have flail paralysis of their legs. They assume a resting position with their hips in abduction and external rotation. The habitual posturing of the hips in this position may contribute in the development of contractures of the tensor fascia lata

535

(TFL), associated with a tight posterior hip capsule and short external rotators [35, 37, 38]. Muscle imbalance in utero may contribute to the development of “arthrogrypotic” type of rigid hip contractures that are present since birth in some children with MMC [38]. Muscle imbalance and hip contractures in a growing child may lead to secondary bony deformities of the acetabulum. The malpositioned femoral head can gradually cause dysplasia and deficiency of the acetabulum [36]. The acetabular deficiency can be anterior, posterior and superior. The posterior acetabular deficiency seems to be significantly greater [39, 40]. Coxa valga, laxity of the capsule of the joint and increased femoral neck anteversion are secondary deformities that may also contribute to the progression of hip subluxation and eventually dislocation. Jones [41] and Somerville [42] emphasized the importance of coxa valga in the pathogenesis of paralytic hip instability. The coxa valga deformity is hypothesized to be a result of the iliopsoas predominance [36, 43]. The iliopsoas action moves the head of the femur proximally and posteriorly in a way that the neck is aligned with the diaphysis. Dias and colleagues described two discrete forms of femoral neck abnormalities in children with thoracic level of involvement and abduction contractures [44]. In the first type of deformity, they observed widening of the physis with a varus deformity of the femoral neck. The authors attributed this deformity to micro-trauma from stretching exercises performed to correct the abduction contractures. The second type of deformity described consisted of narrowing of the femoral neck giving a “mushroom” appearance of the femoral head. The cause of this deformity is thought to be the lack of stimulation of the greater trochanter physis due to gluteus medius and minimus paralysis. No direct causative relation between hip instability and pelvic obliquity has been identified, though many authors have previously made this connection [45, 46]. Keggi and colleagues investigated the relationship between unilateral hip dislocation and pelvic obliquity with scoliosis [47]. They observed no correlation between hip

E. Morakis et al.

536

instability and pelvic obliquity, although there was a correlation between scoliosis and pelvic obliquity. Most likely pelvic obliquity and hip instability are co-morbid factors without a causative relationship. On the contrary, development of pelvic obliquity has been described after successful reduction of a dislocated hip [48].

Natural History

Neurologic function can deteriorate over time. This can be the result of ventriculoperitoneal shunt malfunction, a clinically relevant (rather than MR identification alone) tethered cord, or compression of the cervical canal from a Chiari II malformation or hydromyelia [60].

Prognosis for Ambulation Early ambulation should be encouraged in all children with MMC as it may provide significant benefits long-term, even if they lose their ambulatory ability in the future. Patients with high level MMC that were encouraged to walk during childhood had fewer fractures and pressure sores as compared to patients who had used a wheelchair from early in life. They were also more independent and better able to transfer [61].

Modern medical practice has dramatically changed the survival rates in children born with MMC. In the past, most children with MMC died during infancy secondary to untreated hydrocephalus or meningitis. The children that survived died later in life as a result of renal failure [49]. The introduction of antibiotics, the development of the ventriculoperitoneal (VP) shunt, early surgical closure of the lesion, procedures for urinary diversion, and intermittent catheterization, have improved survival rates dramatically; with survival rates to adulthood reaching 75% [50–52]. Currently, the leading cause of mortality in these patients remains renal failure [53]. The long term prognosis of these patients seems to correlate with level of the lesion, the presence of hydrocephalus and medical management of the several comorbidities. Most patients without hydrocephalus or with hydrocephalus and low lumbar/sacral lesions are independent in most activities of daily living. There is ample evidence that life satisfaction in adulthood is associated with employment, participation in the community, social integration and physical health status. Several studies have shown that patients with MMC have impaired health-related quality of life (HRQOL) [54–56]. Their HRQOL tends to decline from adolescence to young adulthood, due to difficulties in the transition to adult health care [57]. Young adults with MMC are less likely to attend college or be employed comparing to healthy young people [58, 59].

Several factors have been shown to be associated with the potential for ambulation in patients with MMC, including scoliosis, pelvic obliquity, hip flexion contractures [11], upper extremity function [62], knee extensor muscle strength and spasticity of the hip and knee joint muscles groups [4, 14, 63–65]. The most significant prognostic factor for ambulation seems to be the level of neurologic involvement [4, 13, 16, 66]. Almost all patients with L5 and sacral neurosegmental level of involvement are community ambulators, some patients with L4 level of involvement are community ambulators and most patients with higher levels of involvement are non-functional ambulators. Amongst the children that manage to become functional ambulators during childhood, there is a group that maintains this ability into adulthood. Strong predictors for that has been to shown to be their quadriceps strength, absence of spasticity and balance disturbance [14, 67–69].

“Several studies have shown that patients with MMC have impaired health-related quality of life”.

“The most significant prognostic factor for ambulation seems to be the level of neurologic involvement”.

“Early ambulation should be encouraged in all children with MMC as it may provide significant benefits long-term, even if they lose their ambulatory ability in the future”.

19  The Hip in Myelomeningocele

Fig. 19.2  This 5 year-old girl with high lumbar myelomeningocele underwent bilateral hip flexor releases (including psoas, rectus femoris, sartorius) for severe hip flexion contractures. Post-operatively, she was able to use a reciprocating gait orthosis for functional ambulation. Both of her hips were dislocated but no attempts were made to reduce them. (Courtesy Jason J. Howard)

Facilitation of functional ambulation is difficult in children with high-level MMC but a worthy goal for the reasons discussed above. The use of a reciprocating gait orthosis (RGO) can help facilitate walking for these patients with good truncal control and mobile hips (Fig.  19.2). At the very least, the use of a standing frame should be encouraged to allow for weight-bearing and the preservation of bone stock. With respect to the hip, flexion contractures greater than 30–40° can preclude the use of these assistive devices, necessitating surgical release to allow the patient to avail of their benefits.

Natural History of Hip Deformities Contrary to previously held beliefs, the natural history of hip dislocation is not as dire as was

537

commonly thought. Previous work using three-­ dimensional gait analysis (3DGA) in ambulatory children with MMC has shown that the presence or absence of unilateral hip dislocation had no impact on gait function or symmetry. Gait asymmetry was instead related to the presence of unilateral soft tissue contracture and the authors recommended its treatment rather than hip reduction [70]. In a recent evidence-based review, ten Level II studies investigated the factors relating to ambulatory ability in MMC [2]. Those factors included cognitive ability, parental involvement, physiotherapy, neurosegmental level, clubfoot, hip and knee contractures, back pain, scoliosis, lack of motivation, and age. The presence of hip dislocation was not found to be related to walking ability. Alman and colleagues reviewed the outcomes of patients with L3 and L4 MMC and hip dislocation and found no difference in walking or functional ability but high rates of re-dislocation and arthrofibrosis for those who underwent operative reduction [71]. In summary, though common across neurosegmental levels, the best evidence to date suggests that hip dislocation does not reliably result in clinical dysfunction and does not significantly impact ambulatory capacity [2]. Thus, in addition to the fact that these hips are rarely painful and that the natural history of the untreated hip dislocation in MMC is not positively impacted by surgical intervention, these hips should be left alone in most cases (Fig.  19.1). The treatment of hip dislocation in sacral level MMC remains somewhat controversial but, due to limb length discrepancy and likely gait asymmetry, unilateral cases should likely be treated in these higher functioning children [72]. “Hip dislocation does not reliably result in clinical dysfunction and does not significantly impact ambulatory capacity”.

Epidemiology The incidence of MMC is variable and depends upon ethnic, geographic, and nutritional factors. It usually ranges from 0.2 to 4 per 1000 live

538

births [73–76]. The average worldwide incidence of spina bifida is 1 case per 1000 live births. It is more common in Hispanic than non-Hispanic white people and even more rare in African-­ Americans [77, 78]. The prevalence rate of MMC is slightly higher in females than in males [79]. In developed countries, there has been a decreasing trend in the incidence of neural tube defects and MMC.  The two main contributing factors have been prenatal screening with elective termination of affected pregnancies and folate administration to women before and during pregnancy. Maternal screening of serum alpha fetoprotein (AFP) levels in the second trimester of pregnancy can identify the risk of neural tube defects. Ultrasound findings can confirm the diagnosis in utero. Patents with MMC usually have normal intelligence but they do exhibit learning difficulties. Cognitive function seems to be affected by the presence of Chiari II malformation, hydrocephalus and central nervous system infections [80, 81]. Almost all patients with MMC have bowel and bladder dysfunction. The presence of neurogenic bladder requires clean intermittent catheterization, while some may develop chronic renal disease [82, 83]. Sixty to 70% of the patients demonstrate bowel incontinence due to dysmotility and poor sphincter control [84]. Both bowel and bladder dysfunction can have a significant effect in patients social interactions and self-­ esteem [85]. Pressure sores represent a significant problem in patients with MMC.  They have a significant negative impact in their quality of life and cost of their medical care [86]. The most common locations are the sacrum, over the ischial tuberosity, and on the foot. Contributing factors include the absence of sensation, lack of mobility, bowel and bladder incontinence and bony prominences as a result of spine or limb deformities. Meticulous inspection of all pressure areas on routine basis is of paramount importance to detect early indications of pressure. Pathological fractures are common in patients with MMC [87, 88]. Their incidence correlates with the level of the lesion [88]. Osteopenia com-

E. Morakis et al.

bined with absent sensation, joint contractures predispose to fractures. They can present with swelling and redness resembling an infection. Physeal injuries can lead to growth disturbance.

Epidemiology of Hip Deformities The incidence and natural history of hip deformities has been shown to correlate with the neuro-­ segmental level of involvement. Hip instability and contractures are common findings in patients with higher levels of neurological involvement and quite rare in patients with sacral lesions. “Hip instability and contractures are common findings in patients with higher levels of neurological involvement”. The exact incidence of hip instability of a new-born with MMC is not known as it is difficult to extract accurate data from the literature. The incidence seems to be 10–20% in patients with thoracic-level of involvement, 33–50% in patients with upper-lumbar level of involvement, 33–66% in patients with involvement at the L3 and L4 levels, 20% at the L5 level and quite rare in patients with sacral-level [89]. Paralytic hip instability seems to increase in the first 3 years of life, especially in the children with L1–L5 level of involvement [90]. Broughton and colleagues, retrospectively reviewed the records and X-rays of 1061 children over a period of 17  years from two centres (Melbourne and Seattle) [1]. They reported a correlation between hip instability and age for each neurosegmental level. In children with L3 level of involvement, they noticed no hip dislocation after the age of 3 years. On the other hand, children with thoracic and L1/2 levels of involvement continued to develop hip dislocation even after the age of 10 years. They did not identify any other factors, within each group of patients with the same level of involvement, that would likely contribute in the development of hip dislocation. They concluded that hip dislocation is not inevitable but unpredictable [38]. In a retrospective study of 127 patients with MMC that had no surgical treatment for their hips,

19  The Hip in Myelomeningocele

Samuelsson and Eklöf reported a 43% rate of hip instability (subluxation or dislocation) at a mean follow up of 15 years [91]. The rate of dislocation or subluxation depending on the level of involvement was 25% dislocated and 45% subluxated for thoracic level; 20% dislocated and 20% subluxated for L1/2 levels; 29% dislocated and 37% subluxated for L3 level; 21% dislocated and 41% subluxated for L4 level; no dislocation and 41% subluxated for L5 level and no dislocation and 11% subluxated for sacral level of involvement. The long term incidence of hip instability into adulthood is unknown as there are few data in the literature. Within a group of 29 adults with MMC and average age of 28, Barden and colleagues reported a hip dislocation rate of 24% (14/58 hips) in 9/29 patients [92]. All patients had L2/3 level of involvement. Few of them had treatment, mainly closed reduction and osteotomies. There is suggestion that developing spasticity may contribute in the development of hip dislocation in adult patients [91]. Newly developing spasticity can be a result of a tethered cord, hydromyelia or a ventriculoperitoneal shunt malfunction [93, 94]. Different forms of hip contractures can develop in children with MMC including flexion, adduction with or without flexion, external rotation with or without flexion-abduction, and abduction contractures. The incidence and severity of these hip contractures in patients with MMC seems to correlate with age and the level of involvement [38]. Shurtleff et al. reported on the incidence and natural history of hip flexion contractures of 966 patients with MMC [35]. The incidence of hip flexion contractures more than 21° was 40% for patients with thoracic level of involvement, 54% with L1–2 level of involvement, 25% with L3 level of involvement, 21% with L4–5 level of involvement and 6% with sacral level of involvement. They noticed a decrease in the severity of this deformity in the first 27  months of life in all children with lumbar and sacral level of involvement. With increasing age after the first 27 months of life, there was a deterioration of this deformity in patients with thoracic and L4–5 level of involvement. They reported that patients with higher level of involvement (thoracic mainly) demonstrated a tendency to retain the physi-

539

ologic hip flexion position in the first years of life and gradually develop more severe hip flexion contractures with time. Other patients, mainly the ones with L4–5 level of involvement, lost the physiologic hip flexion position in the first years of life, but developed hip contractures with age. Patients with sacral level of involvement rarely developed this deformity (after resolution of the hip flexion contracture typically present at birth). Children with higher level of involvement (thoracic, L1–2) developed more severe deformities over time compared with children with lower lumbar and sacral level of involvement [1]. Hip abduction contractures usually form in children with in higher neurosegmental levels. External rotation contractures are also seen in this group and can be combined with flexion contractures [95, 96]. Posturing with the legs in flexed-abducted-external rotated position in combination with prolonged sitting may predispose to the development of hip contractures. Finally, hip adduction contractures can develop in combination with other deformities; usually with hip flexion contractures and hip displacement [45].

Clinical Presentation The role of the orthopaedic surgeon as part of a multidisciplinary team in the management of children with MMC is crucial. The physical examination of these children is his important diagnostic tool. This starts with assessment of a new-born baby with MMC. The social circumstances of the family are extremely important as they may make the management of children with complex needs challenging for the different health professionals involved into the care of this child. The physical assessment of the newborn by the orthopaedic surgeon is mainly focused on the evaluation of the spinal cord function and the diagnosis of congenital deformities. The posture that the baby assumes can give clues regarding the level of neurological involvement: a baby with a thoracic level of involvement would usually lay with the legs in external rotated position without any voluntary activity, while a baby with a lower lumbar level of

540

involvement would demonstrate flexion of the hips and knees with ­calcaneus posture of the feet. With the child calm and warm but not sleepy, a first assessment of the level of normal sensation in the lower limbs is undertaken starting with the most distal dermatomes in the perianal region and progressing proximally. The presence of intact sensation is usually indicated by facial expressions, a cry or a Moro response from the baby upon stimulation. Then the motor function is assessed in a stimulated or crying baby, looking for voluntary movement of each limb. The muscle power with gravity eliminated is tested for each muscle. The determination of motor level is unreliable until after age 4 years, after which a formal strength assessment can be performed in a cooperative child. A thorough examination looking for congenital deformities (e.g. clubfoot, kyphosis) is performed. Finally, the range of motion of the joints is tested and recorded. Following the initial orthopaedic evaluation, children with MMC should be reviewed regularly by the orthopaedic surgeon who makes part of a multidisciplinary team; in addition to urology, neurosurgery, nursing, social work, physiotherapy and occupational therapy. The periodic orthopaedic assessments include an evaluation of the sensory and motor function of all limbs and the examination for developing spine or limb deformities. A deterioration in the neurologic status or mobility of a growing child could indicate a development of a tethered spinal cord. The inspection of the insensate skin is critical, as the review of the current orthotics that the child is using. These should maximize the mobility of the child without causing skin problems.

E. Morakis et al.

objective evaluation of gait patterns which may assist in brace selection and surgical planning. Fixed and dynamic hip deformities can affect the gait in patients with MMC. Fixed hip deformities include hip subluxation or dislocation, hip contractures and pelvic obliquity. Dynamic deformities include muscle weakness, imbalance, spasticity, secondary gait compensations for muscle weakness [98]. Patients with lower lumbar level of involvement demonstrate a “hip abductor avoidance gait” due to the hip abductor weakness [98]. This is characterized by lateral movement of the trunk over the stance-phase limb, pelvic external rotation and medial translation of the hip (i.e. Trendelenburg gait). This gait pattern provides stability in stance and a means of progression in swing. Coronal-plane hip motion is out of phase and there is increased range of motion. The hip is in significant abduction during stance and goes into adduction during swing [99]. The contralateral side of the pelvis is “hiked” by the trunk during swing to achieve progression of the limb and clearance during swing [99]. The ground reaction force passes lateral to the hip and knee applying an external abduction moment to the hip and valgus moment to the knee. The increased range motion noted during this gait pattern may be hindered if the hip becomes stiff, as may occur following a surgical hip reduction, and gait may deteriorate. In the sagittal plane, posterior lean of the trunk provides stance phase stability to counteract typically weak hip extensors. Some children use posterior trunk shift to decrease the anterior pelvic tilt before toe-off to help advance the limb into swing phase [99].

 valuation of Gait in Patients E with Myelomeningocele

Non-ambulatory Hip Assessment

Distinct gait patterns have been described in these patients using instrumented 3DGA [97]. Compensatory movement patterns have been identified using the data from the gait analysis. Developing “normal” values for these patients can assist clinicians in the evaluation and management of the gait of ambulatory patients with MMC. Instrumented gait analysis can provide an

For children in higher neurosegmental levels, in addition to hip flexion contractures, abduction and external rotation contractures (typically due to a tight TFL) or adduction contractures are often coincident. In flail hips, the hips are often positioned in abduction and external rotation which over time can lead to fixed contractures. These deformities can compromise wheelchair seating and as such

19  The Hip in Myelomeningocele

may require surgical management. In addition to seating concerns due to pelvic obliquity, adduction contractures may also cause difficulties in toileting and perineal hygiene. The resulting seating imbalance associated with these contractures in high level lesions with insensate skin can also lead to pressure ulceration, a complication that should also be sought during the physical examination.

Essential Clinical Tests

• Thomas Test: To identify the presence of an iliopsoas contracture. The patient is placed supine with the contralateral hip flexed to align the ASIS-PSIS line in a vertical position and the resulting angle between the thigh and the horizontal for the ipsilateral hip is measured with a goniometer. • Ober Test: To identify the presence of a tensor fascia lata contracture. The patient is placed in the lateral decubitus position with the downside limb flexed at the hip and knee. The upside limb is flexed and abducted at the hip, then brought into extension and the thigh then slowly allowed to fall into adduction. Any residual abduction from the neutral is recorded. • Hip Abduction Range: To identify the presence of a hip adduction contracture. The hip and knee are flexed to 90° (testing the adductor longus primarily) followed by abduction of the hip. The amount of hip abduction loss is recorded bilaterally. With the hip and knee extended to neutral (testing the proximal hamstrings primarily), the abduction range is again recorded.

Imaging Imaging studies of the hips can occasionally assist the orthopaedic surgeon in the diagnosis and management of hip pathology in patients with MMC. A hip ultrasound scan in a new-born with MMC can diagnose hip dislocation or instability if the physical assessment is doubtful.

541

In older children, X-ray of the hips is a useful imaging study to identify hip pathology (Fig. 19.1). Routine hip X-rays in the form of a hip surveillance program are not warranted. The development of hip instability in children with MMC is less predictable than in children with cerebral palsy. Hip displacement does not seem to affect the function of most patients with MMC and preventive hip surgery is not justified [2]. If there is doubt from the physical examination, X-rays of the hips can identify hip displacement and dysplasia. X-rays of the hips can be useful to identify pathologic fractures (Fig.  19.3). Patients with MMC, especially non-functional ambulators, have a higher risk of pathologic insufficiency fractures. The presence of hip contractures, the loss of sensation and the use post-operative immobilization with a hip spica can contribute in the occurrence of a low-energy insufficiency fracture by an inattentive patient or caretaker [100–102]. New-born babies with thoracic or upper lumbar level of involvement and significant contractures have increased risk of birth fractures [103]. Dual-energy X-ray absorptiometry (DEXA) can be useful in patients with MMC to help identify patients with low bone mineral density and direct appropriate management to prevent pathologic fractures [104–106].

Essential Imaging

• Plain radiography is warranted for the diagnosis of hip dislocation and insufficiency fractures. • Hip surveillance programs using serial X-ray examinations at defined intervals are not warranted since hip displacement is not reliably predicted by motor level, dislocated hips are typically pain free, and surgical re-location does not lead to better outcomes than nihilistic treatment. • DEXA bone densitometry scanning is useful for non-ambulatory patients where osteopenia and insufficiency fractures are common.

542

Fig. 19.3  Fifteen year-old male with thoracic myelomeningocele, severe scoliosis, and right femoral shaft insufficiency fractures post contralateral hip reconstruction surgery. A Thomas splint was temporarily placed immedi-

Non-operative Management Multidisciplinary Team Approach The management of these complex patients with several comorbidities requires a multidisciplinary team approach. To achieve optimal outcomes for these patients the co-operation of several health professionals is required, including orthopaedic surgeon, urologist, neurosurgeon, occupational therapist, physical

E. Morakis et al.

ately post fracture. The right femoral shaft fracture was stabilized with flexible intramedullary nailing. A chronic femoral neck non-union is also evident. (Courtesy Jason J. Howard)

therapist, psychologist, orthotist, dietician, social worker and pediatrician. “To achieve optimal outcomes for these patients the co-operation of several health professionals is required”. Multidisciplinary clinics allow patients and their families to be evaluated during one clinic visit by several health professionals. Their care is better co-ordinated. It minimizes time off work and school, allows more effective communication

19  The Hip in Myelomeningocele

between health professionals and allows timely clinical decisions. It has been shown to provide better clinical outcomes and reduced morbidity for these complex patients [107]. The orthopaedic surgeon has an important role within this team to help a child with spina bifida to reach its maximum potential of function and independence in the society. Orthopedic surgical interventions should boost, rather than hinder, the potential of a child for physical and psychological development. “Orthopedic surgical interventions should boost, rather than hinder, the potential of a child for physical and psychological development”.

Non-operative Treatment of Hip Dislocation As discussed at length elsewhere in the chapter, there is little support for the surgical reconstruction of dislocated hips in MMC.  Instead, the focus should be on the prevention and treatment of peri-articular contractures by either physiotherapy, positioning, or both. As detailed in the Operative Management section, surgical management of functionally limiting soft tissue contractures, rather than re-location of the hip should be practiced. The only caveat to this involves the L5 or sacral level child with a unilateral dislocation, where surgical reduction might be a reasonable approach despite the lack of evidence.

 ositioning to Reduce Contracture P Development Although there is little evidence to support the practice, positioning programs which serve to avoid hip flexion, abduction, and external rotation may help prevent the development of fixed contractures. The use of a sleeve that holds the hips in adduction and internal rotation has been previously advocated for children with extensive paralysis [108].

543

 he Use of the Pavlik Harness for Hip T Dislocation in Myelomeningocele It may be reasonable to use a Pavlik harness to reduce an Ortolani positive hip in MMC for children at L5 and sacral levels. Despite the lack of evidence to support (or refute) this practice, our approach is to perform a trial of Pavlik harness hip reduction under ultrasound guidance in a similar manner to the practice for developmental hip dysplasia (DDH). Unlike DDH, however, if the hip does not readily reduce and stabilize using the Pavlik harness, no attempts to perform a closed or open reduction under general anesthesia are made.

Essential Non-operative Management

• Non-operative management of hip dislocation is the rule; the only caveat being the child with unilateral dislocation at an L5 or sacral level. • Positioning and physiotherapy to prevent fixed contractures may be useful; especially in the high lumbar and thoracic levels. • The use of a Pavlik harness in children in L5 and sacral levels with Ortolanipositive hips may be warranted but the evidence for this practice is lacking.

Operative Management General Considerations Given reasonably strong evidence against the surgical reduction of dislocated hips in MMC, the focus of management has shifted to lengthening of functionally limiting contractures of the hip flexors, adductors, and abductors [2, 109]. Addressing these soft tissue deformities can help to realize one of the primary goals of the orthopedic surgeon with respect to the treatment of children with MMC: to facilitate the use of orthoses for standing tolerance and/or functional ambulation.

544

Fig. 19.4  Six year-old girl with thoracic level myelomeningocele and right unilateral hip dislocation. This child had bilateral hip flexion contractures (Thomas test: right 80°, left 55°) which precluded the use of a standing frame. Note the increased anterior pelvic tilt apparent on X-ray. She underwent bilateral releases of her psoas tendon at the pelvic brim, rectus femoris, sartorius, and tensor fascia lata (achieving contracture reduction to 15° bilaterally) which allowed for standing frame use post-operatively. The right hip adduction and pelvic obliquity shown here was due to X-ray positioning rather than the presence of a fixed adduction contracture. (Courtesy Jason J. Howard)

Hip Flexion Contracture Hip flexion contractures greater than 30–40° usually interfere with walking (for children L4 and below) and the use of standing frames and/or RGOs (for higher level lesions); providing relative indications for surgical release [110]. Unlike cerebral palsy, these fixed flexion deformities are caused by more than just the iliopsoas, with sartorius, TFL, rectus femoris—and occasionally—the anterior hip capsule contributing to the deformity (Fig. 19.4).

 urgical Procedure: Hip Flexion S Contracture Release The patient is positioned supine with a folded surgical towel placed under the sacrum to raise the pelvis for access to the iliac crest laterally. A “Salter” bikini-style incision is made 1–2 fingerbreadths below the anterior superior iliac spine (ASIS) and the interval between tensor fascia lata (TFL) and sartorius is identified. The lateral femoral cutaneous nerve is identified and protected

E. Morakis et al.

throughout the procedure. The external oblique muscle is reflected off the iliac apophysis, which is subsequently split anteriorly for approximately 2 cm to allow for access to the direct and reflected heads of the rectus femoris deep to TFL-sartorius interval. The common tendon of rectus femoris is divided under direct vision. Subsequent to this, the psoas tendon is identified on the under surface of its muscle belly, rolling the muscle medially. The tendon is confirmed to have muscle fibres inserting into it and is stimulated with electrocautery (to confirm it is not the femoral nerve) and is released. The sartorius is often tight and should be mobilized and completely released from the ASIS as necessary. The tensor fascia lata can also be a substantial contributor to the flexion contracture, necessitating its lengthening. Extrafascial dissection over the TFL at the anterior third of the iliac crest laterally followed by incision of its overlying fascia (i.e. muscle recession) approximately three fingerbreadths below the iliac crest, is performed. If a significant residual hip flexion contracture is still evident following these releases, an anterior hip capsular release may be added but is rarely necessary. Post-operatively, prone lying and daily range of motion exercises is encouraged. No post-­ operative immobilization is typically required. Physiotherapy involvement is important to help reduce the risk of recurrence through patient and family education as to the importance of contracture prevention. The use of a standing frame can be commenced immediately post-operatively and an RGO introduced after the wounds have healed at 2 weeks post-operatively, as appropriate (Fig. 19.2).

Abduction-External Rotation Contracture For children with thoracic lesions, the hips are typically flail and tend to abduct and externally rotate which, over time, can result in fixed contracture and seating intolerance. Correction of the resulting abduction-external rotation deformity can be addressed by TFL release plus or minus a distal release of the iliotibial band. Appropriate post-operative positioning to avoid hip abduc-

19  The Hip in Myelomeningocele

tion/external rotation is encouraged via physiotherapy input and wheelchair adjuncts.

Adduction Contracture More common in higher level lesions, problems with perineal hygiene or sitting intolerance may occur secondary to the presence of a hip adduction contracture. These deformities are relatively easy to treat using open lengthening of the adductor longus and iliopsoas most commonly.

I s There Any Role for Reduction of the Hip in Myelomeningocele? Given the pain-free nature of these hips and the lack of functional improvement post-surgical hip reduction, the use of femoral and/or pelvic osteotomies is rarely indicated. In addition, there is a real risk of arthrofibrosis and heterotopic ossification following hip reconstruction in MMC which would further negate any surgical benefits and would likely make things worse [111]. Indeed, a stiff, located hip is much more functionally limiting than a mobile, dislocated hip. Community ambulators at the L5 or sacral levels with a unilateral hip dislocation, however, may benefit from surgical reduction to resolve their limb length discrepancy and—for sacral lesions—improve their Trendelenburg gait. Though this may seem to be reasonable, the literature is currently unclear as to whether this approach actually results in a functional improvement [2]. “Community ambulators at the L5 or sacral levels with a unilateral hip dislocation may benefit from surgical reduction to resolve their limb length discrepancy and—for sacral lesions—improve their Trendelenburg gait”.

 he Role of Muscle Rebalancing T Procedures In the past, muscle imbalance was thought to be a primary driver of hip dislocation in MMC and

545

tendon transfers were commonly performed in an attempt to prevent or treat hip dislocation. In a large combined series of MMC patients from Melbourne and Seattle, however, the prevalence of hip dislocation was unpredictable, with similar rates of dislocation evident in flail hips (thoracic) versus those with maximum muscle imbalance (L4 level) [1]. As such, rebalancing procedures such as the posterior iliopsoas transfer described by Sharrard [5] have proven unable to stabilize the hip and improve Trendelenburg gait. This is likely due to the lack of iliopsoas muscle activity during stance post transfer, confirmed by EMG studies previously [112]. For these reasons, tendon transfer procedures have little current role in the management of hip deformities in MMC. “Tendon transfer procedures have little current role in the management of hip deformities in MMC”.

Post-operative Complications Although the procedures most commonly indicated for treatment of the hip in MMC involve simple soft tissue releases, coincident comorbidities such as osteopenia, neurogenic bladder (with need for repeated catheterization), and a lack of protective sensation, require acknowledgment and mitigation to prevent post-operative complications. Associated with a lack of weight bearing in non-ambulatory children with MMC, these children are typically osteopenic and prone to iatrogenic fractures, particularly after hip spica casting (Fig. 19.3). As such, cast immobilization should be avoided in favour of custom moulded splinting and early post-operative mobilization [72]. In the rare cases where operative hip reduction is performed, rigid internal fixation should be used and early mobilization the rule. Well-­meaning physiotherapists and nursing staff should be educated about the risks of insufficiency fractures in the postoperative period to help avoid this complication. The use of casting for these children with insensate skin can predispose to ulceration with subsequent deep wound infection and, in some

E. Morakis et al.

546

cases, osteomyelitis. If casting is to be used, it should be well padded at all bony prominences and frequent skin checks instituted. Even without casting, frequent turning and skin inspection by nursing staff is required to avoid pressure ulceration in the early post-operative inpatient phase. Although the cause is not explicitly known—as a result of repeated exposure from multiple surgeries, diagnostic tests, bowel and bladder instrumentation—patients with MMC have high risk for allergic reactions to latex [113]. The prevalence of allergic reactions in these patients has been reported up to 22% [114–116]. Patients at higher risk have asthma or other allergies, history of multiple surgical procedures or presence of a shunt [117, 118]. Currently, it is recommended to strictly avoid exposure to latex from birth as a mean to prevent both allergic reactions and sensitization to latex [119]. One of the most common complications following soft tissue releases around the hip in MMC is contracture recurrence. For adductor releases, a foam hip abduction orthosis is often used for the first 6 weeks post-operatively and then at night for a further 6–12 weeks at the surgeon’s discretion. For hip flexor releases, a regular home program consisting of prone lying and positioning which avoids external rotation should be instituted to reduce the risk of recurrence.

Essential Surgical Techniques

• With the exception of children with motor levels below L4, dislocated hips in MMC should not be reduced surgically (good evidence). • Functionally limiting fixed soft tissue contractures should be addressed surgically if impeding gait, standing, or orthotic use. • Hip flexion contractures can be treated through an anterior approach, typically addressing the psoas, rectus femoris, TFL, sartorius and, rarely, the hip capsule. • Muscle rebalancing procedures (i.e. tendon transfers) are not indicated in MMC.

• Cast immobilization should be minimized and, ideally, avoided post-operatively, due to the risk of iatrogenic fracture. • Latex precautions must be practiced when providing surgical management in MMC.

Classic Papers Sharrard WJ.  The Segmental Innervation of the Lower Limb Muscles in Man. Ann R Coll Surg Engl, 1964. 35: p. 106–22. Sharrard’s classic article providing the basis of the commonly used neurosegmental classification in MMC. Frawley PA, Broughton NS, Menelaus, MB.  Anterior release for fixed flexion deformity of the hip in spina bifida. J Bone Joint Surg Br 1996;78-B:299–302. Case series of 38 children detailing a technique for anterior release for treating hip flexion contractures greater than 30° with functional impariment. No muscle transfers were performed and 43 of 57 hips had no recurrence. Broughton, N.S., et  al., The natural history of hip deformity in myelomeningocele. J Bone Joint Surg Br, 1993. 75(5): p. 760–3. This series of over 1000 children with MMC from Melbourne and Seattle helped confirm that the development of hip dislocation was not predicted by motor level and that its natural history was not as dire as previously thought.

Key Evidence Wright, J.G., Hip and spine surgery is of questionable value in spina bifida: an evidence-­ based review. Clin Orthop Relat Res, 2011. 469(5): p.  1258–64. This systematic review concluded that hip reduction surgery does not improve function in spina bifida (Grade B Recommendation) but suggested that a unilateral hip dislocation in children below L4 level should likely be reduced to treat limb length discrepancy

19  The Hip in Myelomeningocele

and to improve their Trendeleberg gait (Grade I Recommendation). Feiwell, E., D.  Sakai, and T.  Blatt, The effect of hip reduction on function in patients with myelomeningocele. Potential gains and hazards of surgical treatment. J Bone Joint Surg Am, 1978. 60(2): p.  169–73. This article assessed the impact of hip dislocation on ambulatory capacity and assessed recurrence rates following relocation. Approximaltely half of the children had surgical relocation and half were left dislocated. The redislocation rate was 40% in the operative group. In their series, the presence of hip dislocation was not related to ambulatory capacity. Alman BA, Bhandari M, Wright JG.  Function of dislocated hips in children with lower level spina bifida. J Bone Joint Surg Br. 1996;78:294–298. This study from the Hospital for Sick Children in Toronto reviewed the results of 52 children with dislocated hips, 30 of whom had surgery. For the hips that were reduced, a 30% redislocation rate was found. There were no functional differences (including ambulatory capacity) between the surgical and non-surgical groups. The surgical group had an increased rate of arthrofibrosis and limb length discrepancy.

Take Home Messages

• Hip deformities in MMC are common— including both soft tissue contractures and dislocation—and typically painless. • The prevalence of hip dislocation in MMC is not reliably predicted by neurosegmental level. • Functional limitations associated with dislocated hips are not improved with surgery. • Surgical management of functionally limiting soft tissue contractures by muscle releases, rather than relocation of dislocated hips, can help in gait improvement, standing, and orthotic use (e.g. RGO).

547

• The unilateral dislocated hip in MMC below L4 may benefit from surgical reduction. • Muscle rebalancing procedures are no longer indicated in MMC due to poor outcomes and failure to prevent re-contracture or recurrence of dislocation. • Cast immobilization should be avoided where possible to prevent iatrogenic fracture due to exacerbation of preexisting osteopenia in higher level MMC.

References 1. Broughton NS, Menelaus MB, Cole WG, Shurtleff DB. The natural history of hip deformity in myelomeningocele. J Bone Joint Surg Br. 1993;75(5):760–3. 2. Wright JG. Hip and spine surgery is of questionable value in spina bifida: an evidence-based review. Clin Orthop Relat Res. 2011;469(5):1258–64. https://doi. org/10.1007/s11999-010-1595-y. 3. Tulpius N.  Observationes medicae. Amsterdam: Elzevirium; 1641. 4. Asher M, Olson J.  Factors affecting the ambulatory status of patients with spina bifida cystica. J Bone Joint Surg Am. 1983;65(3):350–6. 5. Sharrard WJ.  The segmental innervation of the lower limb muscles in man. Ann R Coll Surg Engl. 1964;35:106–22. 6. Duckworth T, Yamashita T, Franks CI, Brown BH.  Somatosensory evoked cortical responses in children with spina bifida. Dev Med Child Neurol. 1976;18(1):19–24. 7. Stark Gordon D, Baker Geoffrey CW. The neurological involvement of the lower limbs in myelomeningocele. Dev Med Child Neurol. 1967;9(6):732–44. https://doi.org/10.1111/j.1469-8749.1967.tb02355.x. 8. Shurtleff DB.  International Myelodysplasia Study Group Database Coordination. Seattle, WA: Department of Pediatrics, University of Washington; 1993. 9. Charney EB, Melchionni JB, Smith DR. Community ambulation by children with myelomeningocele and high-level paralysis. J Pediatr Orthop. 1991;11(5):579–82. 10. Mazur JM, Kyle S. Efficacy of bracing the lower limbs and ambulation training in children with myelomeningocele. Dev Med Child Neurol. 2004;46(5):352–6. 11. Stillwell A, Menelaus MB.  Walking ability in mature patients with spina bifida. J Pediatr Orthop. 1983;3(2):184–90.

548 12. Swank M, Dias L. Myelomeningocele: a review of the orthopaedic aspects of 206 patients treated from birth with no selection criteria. Dev Med Child Neurol. 1992;34(12):1047–52. 13. De Souza LJ, Carroll N.  Ambulation of the braced myelomeningocele patient. J Bone Joint Surg Am. 1976;58(8):1112–8. 14. McDonald CM, Jaffe KM, Mosca VS, Shurtleff DB. Ambulatory outcome of children with myelomeningocele: effect of lower-extremity muscle strength. Dev Med Child Neurol. 1991;33(6):482–90. 15. Samuelsson L, Skoog M. Ambulation in patients with myelomeningocele: a multivariate statistical analysis. J Pediatr Orthop. 1988;8(5):569–75. 16. Swank M, Dias LS.  Walking ability in spina bifida patients: a model for predicting future ambulatory status based on sitting balance and motor level. J Pediatr Orthop. 1994;14(6):715–8. 17. Selber P, Dias L.  Sacral-level myelomeningocele: long-term outcome in adults. J Pediatr Orthop. 1998;18(4):423–7. 18. Brinker MR, Rosenfeld SR, Feiwell E, Granger SP, Mitchell DC, Rice JC.  Myelomeningocele at the sacral level. Long-term outcomes in adults. J Bone Joint Surg Am. 1994;76(9):1293–300. 19. Khoury MJ, Erickson JD, James LM.  Etiologic heterogeneity of neural tube defects. II. Clues from family studies. Am J Hum Genet. 1982;34(6):980–7. 20. Seller MJ. Recurrence risks for neural tube defects in a genetic counseling clinic population. J Med Genet. 1981;18(4):245–8. 21. Toriello HV, Higgins JV.  Occurrence of neural tube defects among first-, second-, and third-degree relatives of probands: results of a United States study. Am J Med Genet. 1983;15(4):601–6. https://doi. org/10.1002/ajmg.1320150409. 22. Yang Y, Chen J, Wang B, Ding C, Liu H. Association between MTHFR C677T polymorphism and neural tube defect risks: a comprehensive evaluation in three groups of NTD patients, mothers, and fathers. Birth Defects Res A Clin Mol Teratol. 2015;103(6):488– 500. https://doi.org/10.1002/bdra.23361. 23. Benedum CM, Yazdy MM, Mitchell AA, Werler MM.  Impact of periconceptional use of nitrosatable drugs on the risk of neural tube defects. Am J Epidemiol. 2015;182(8):675–84. https://doi. org/10.1093/aje/kwv126. 24. Loeken MR. Current perspectives on the causes of neural tube defects resulting from diabetic pregnancy. Am J Med Genet C Semin Med Genet. 2005;135C(1):77– 87. https://doi.org/10.1002/ajmg.c.30056. 25. Makelarski JA, Romitti PA, Rocheleau CM, Burns TL, Stewart PA, Waters MA, et al. Maternal periconceptional occupational pesticide exposure and neural tube defects. Birth Defects Res A Clin Mol Teratol. 2014;100(11):877–86. https://doi.org/10.1002/ bdra.23293. 26. Reefhuis J, Honein MA, Schieve LA, Rasmussen SA, National Birth Defects Prevention Study. Use of clomiphene citrate and birth defects, National Birth

E. Morakis et al. Defects Prevention Study, 1997–2005. Hum Reprod. 2011;26(2):451–7. https://doi.org/10.1093/humrep/ deq313. 27. Dreier JW, Andersen AM, Berg-Beckhoff G.  Systematic review and meta-analyses: fever in pregnancy and health impacts in the offspring. Pediatrics. 2014;133(3):e674–88. https://doi. org/10.1542/peds.2013-3205. 28. Hibbard E, Smithells RW. Folic acid metabolism and human embryopathy. Lancet. 1965;285(7398):1254. https://doi.org/10.1016/S0140-6736(65)91895-7. 29. Smithells RW, Sheppard S, Schorah CJ.  Vitamin deficiencies and neural tube defects. Arch Dis Child. 1976;51(12):944–50. 30. Wald NJ, Hackshaw AD, Stone R, Sourial NA. Blood folic acid and vitamin B12 in relation to neural tube defects. Br J Obstet Gynaecol. 1996;103(4):319–24. 31. Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA.  Folic acid antagonists during pregnancy and the risk of birth defects. N Engl J Med. 2000;343(22):1608–14. https://doi.org/10.1056/ NEJM200011303432204. 32. Laurence KM, James N, Miller MH, Tennant GB, Campbell H.  Double-blind randomised controlled trial of folate treatment before conception to prevent recurrence of neural-tube defects. Br Med J (Clin Res Ed). 1981;282(6275):1509–11. 33. MRC Vitamin Study Research Group. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet. 1991;338(8760):131–7. 34. Centers for Disease Control Prevention (CDC). Spina bifida and anencephaly before and after folic acid mandate--United States, 1995–1996 and 1999–2000. MMWR Morb Mortal Wkly Rep. 2004;53(17): 362–5. 35. Shurtleff DB, Menelaus MB, Staheli LT, Chew DE, Lamers JY, Stillwell A, et al. Natural history of flexion deformity of the hip in myelodysplasia. J Pediatr Orthop. 1986;6(6):666–73. 36. Breed AL, Healy PM.  The midlumbar myelomeningocele hip: mechanism of dislocation and treatment. J Pediatr Orthop. 1982;2(1):15–24. 37. Lindseth RE, Dias LS, Drennan JC. Myelomeningocele. Instr Course Lect. 1991;40:271–91. 38. Menelaus MB, Broughton NS. The orthopaedic management of spina bifida cystica, Current problems in orthopaedics. 3rd ed. Edinburgh: WB Saunders; 1998. 39. Buckley SL, Sponseller PD, Magid D. The acetabulum in congenital and neuromuscular hip instability. J Pediatr Orthop. 1991;11(4):498–501. 40. Gugenheim JJ, Gerson LP, Sadler C, Tullos HS. Pathologic morphology of the acetabulum in paralytic and congenital hip instability. J Pediatr Orthop. 1982;2(4):397–400. 41. Blundell Jones G.  Paralytic dislocation of the hip. J Bone Joint Surg. 1954;36(3):375–84. 42. Somerville EW.  Paralytic dislocation of the hip. J Bone Joint Surg Br. 1959;41-B(2):279–88.

19  The Hip in Myelomeningocele 43. Brookes M, Wardle E.  Muscle action and the shape of the femur. J Bone Joint Surg Br. 1962;44-B(2):398–411. 44. Dias LS, Locher FG, Mazur JM, Olysav D. Femoral neck abnormalities in spina bifida. Clin Orthop Relat Res. 1984;(184):164–8. 45. Dias LS, Hill JA. Evaluation of treatment of hip subluxation in myelomeningocele by intertrochanteric varus derotation femoral osteotomy. Orthop Clin North Am. 1980;11(1):31–7. 46. Feiwell E, Sakai D, Blatt T. The effect of hip reduction on function in patients with myelomeningocele. Potential gains and hazards of surgical treatment. J Bone Joint Surg Am. 1978;60(2):169–73. 47. Keggi J, Banta J, Walton C. The myelodysplastic hip and scoliosis. Eur J Pediatr Surg. 1991;1(Suppl 1):47. 48. Fraser RK, Bourke HM, Broughton NS, Menelaus MB.  Unilateral dislocation of the hip in spina bifida. A long-term follow-up. J Bone Joint Surg Br. 1995;77(4):615–9. 49. Lorber J. Results of treatment of myelomeningocele. An analysis of 524 unselected cases, with special reference to possible selection for treatment. Dev Med Child Neurol. 1971;13(3):279–303. 50. Davis BE, Daley CM, Shurtleff DB, Duguay S, Seidel K, Loeser JD, et  al. Long-term survival of individuals with myelomeningocele. Pediatr Neurosurg. 2005;41(4):186–91. https://doi.org/ 10.1159/000086559. 51. Oakeshott P, Hunt GM, Poulton A, Reid F. Expectation of life and unexpected death in open spina bifida: a 40-year complete, non-selective, longitudinal cohort study. Dev Med Child Neurol. 2010;52(8):749–53. https://doi.org/10.1111/j.1469-8749.2009.03543.x. 52. Wong LY, Paulozzi LJ. Survival of infants with spina bifida: a population study, 1979–94. Paediatr Perinat Epidemiol. 2001;15(4):374–8. 53. McDonnell GV, McCann JP.  Why do adults with spina bifida and hydrocephalus die? A clinic-based study. Eur J Pediatr Surg. 2000;10(S1):31–2. https:// doi.org/10.1055/s-2008-1072411. 54. Murray CB, Holmbeck GN, Ros AM, Flores DM, Mir SA, Varni JW. A longitudinal examination of healthrelated quality of life in children and adolescents with spina bifida. J Pediatr Psychol. 2015;40(4):419–30. https://doi.org/10.1093/jpepsy/jsu098. 55. Rocque BG, Bishop ER, Scogin MA, Hopson BD, Arynchyna AA, Boddiford CJ, et al. Assessing healthrelated quality of life in children with spina bifida. J Neurosurg Pediatr. 2015;15(2):144–9. https://doi. org/10.3171/2014.10.peds1441. 56. Rofail D, Maguire L, Kissner M, Colligs A, Abetz-­ Webb L.  Health-related quality of life is compromised in individuals with spina bifida: results from qualitative and quantitative studies. Eur J Obstet Gynecol Reprod Biol. 2014;181:214–22. https://doi. org/10.1016/j.ejogrb.2014.07.048. 57. Bellin MH, Zabel TA, Dicianno BE, Levey E, Garver K, Linroth R, et  al. Correlates of depressive and anxiety symptoms in young adults with spina bifida.

549 J Pediatr Psychol. 2010;35(7):778–89. https://doi. org/10.1093/jpepsy/jsp094. 58. Arnett J, Lynn Tanner J, Gibbons J, Ashdown BK.  Emerging adults in America: coming of age in the 21st century. Washington, DC: American Psychological Association; 2006. 59. Roach JW, Short BF, Saltzman HM.  Adult con sequences of spina bifida: a cohort study. Clin Orthop Relat Res. 2011;469(5):1246–52. https://doi. org/10.1007/s11999-010-1594-z. 60. Hoffman HJ, Hendrick EB, Humphreys RP.  The tethered spinal cord: its protean manifestations, diagnosis and surgical correction. Childs Brain. 1976;2(3):145–55. 61. Mazur JM, Shurtleff D, Menelaus M, Colliver J. Orthopaedic management of high-level spina bifida. Early walking compared with early use of a wheelchair. J Bone Joint Surg Am. 1989;71(1):56–61. 62. Wallace SJ.  The effect of upper-limb function on mobility of children with myelomeningocele. Dev Med Child Neurol Suppl. 1973;(Suppl 29):84–91. 63. Danielsson AJ, Bartonek A, Levey E, McHale K, Sponseller P, Saraste H. Associations between orthopaedic findings, ambulation and health-related quality of life in children with myelomeningocele. J Child Orthop. 2008;2(1):45–54. https://doi.org/10.1007/ s11832-007-0069-6. 64. Hoffer MM, Feiwell E, Perry R, Perry J, Bonnett C.  Functional ambulation in patients with myelomeningocele. J Bone Joint Surg Am. 1973;55(1): 137–48. 65. Huff CW, Ramsey PL. Myelodysplasia. The influence of the quadriceps and hip abductor muscles on ambulatory function and stability of the hip. J Bone Joint Surg Am. 1978;60(4):432–43. 66. Schoenmakers MA, Uiterwaal CS, Gulmans VA, Gooskens RH, Helders PJ.  Determinants of functional independence and quality of life in children with spina bifida. Clin Rehabil. 2005;19(6):677–85. https://doi.org/10.1191/0269215505cr865oa. 67. Bartonek A, Saraste H.  Factors influencing ambulation in myelomeningocele: a cross-sectional study. Dev Med Child Neurol. 2001;43(4):253–60. 68. Schopler SA, Menelaus MB.  Significance of the strength of the quadriceps muscles in children with myelomeningocele. J Pediatr Orthop. 1987;7(5):507–12. 69. Seitzberg A, Lind M, Biering-Sorensen F. Ambulation in adults with myelomeningocele. Is it possible to predict the level of ambulation in early life? Childs Nerv Syst. 2008;24(2):231–7. https://doi.org/10.1007/ s00381-007-0450-2. 70. Gabrieli AP, Vankoski SJ, Dias LS, Milani C, Lourenco A, Filho JL, et al. Gait analysis in low lumbar myelomeningocele patients with unilateral hip dislocation or subluxation. J Pediatr Orthop. 2003;23(3): 330–4. 71. Alman BA, et al. Function of dislocated hips in children with lower level spina bifida. J Bone Joint Surg Br. 1996;78(2):294–8.

550 72. Swaroop VT, Dias L.  Orthopedic management of spina bifida. Part I: hip, knee, and rotational deformities. J Child Orthop. 2009;3(6):441–9. https://doi. org/10.1007/s11832-009-0214-5. 73. Beuriat PA, Szathmari A, Hameury F, Poirot I, Massoud M, Massardier J et al. [Changes in the epidemiology of Spina Bifida in France in the last 30 years]. Neurochirurgie. 2017;63(2):109–111. doi:https://doi. org/10.1016/j.neuchi.2017.01.003. 74. Khoshnood B, Loane M, de Walle H, Arriola L, Addor MC, Barisic I, et al. Long term trends in prevalence of neural tube defects in Europe: population based study. BMJ. 2015;351:h5949. https://doi.org/10.1136/bmj. h5949. 75. North T, Cheong A, Steinbok P, Radic JA.  Trends in incidence and long-term outcomes of myelomeningocele in British Columbia. Childs Nerv Syst. 2018;34(4):717–24. https://doi.org/10.1007/ s00381-017-3685-6. 76. Orioli IM, Lima do Nascimento R, Lopez-Camelo JS, Castilla EE.  Effects of folic acid fortification on spina bifida prevalence in Brazil. Birth Defects Res A Clin Mol Teratol. 2011;91(9):831–5. https://doi. org/10.1002/bdra.20830. 77. Canfield MA, Marengo L, Ramadhani TA, Suarez L, Brender JD, Scheuerle A. The prevalence and predictors of anencephaly and spina bifida in Texas. Paediatr Perinat Epidemiol. 2009;23(1):41–50. https://doi. org/10.1111/j.1365-3016.2008.00975.x. 78. Centers for Disease Control Prevention (CDC). Racial/ethnic differences in the birth prevalence of spina bifida  - United States, 1995–2005. MMWR Morb Mortal Wkly Rep. 2009;57(53):1409–13. 79. Deak KL, Siegel DG, George TM, Gregory S, AshleyKoch A, Speer MC, et  al. Further evidence for a maternal genetic effect and a sex-influenced effect contributing to risk for human neural tube defects. Birth Defects Res A Clin Mol Teratol. 2008;82(10):662–9. https://doi.org/10.1002/bdra.20511. 80. Matson MA, Mahone EM, Zabel TA.  Serial neu ropsychological assessment and evidence of shunt malfunction in spina bifida: a longitudinal case study. Child Neuropsychol. 2005;11(4):315–32. https://doi. org/10.1080/09297040490916910. 81. McLone DG, Czyzewski D, Raimondi AJ, Sommers RC.  Central nervous system infections as a limiting factor in the intelligence of children with myelomeningocele. Pediatrics. 1982;70(3):338. 82. de Jong TP, Chrzan R, Klijn AJ, Dik P.  Treatment of the neurogenic bladder in spina bifida. Pediatr Nephrol. 2008;23(6):889–96. https://doi.org/10.1007/ s00467-008-0780-7. 83. Muller T, Arbeiter K, Aufricht C.  Renal function in meningomyelocele: risk factors, chronic renal failure, renal replacement therapy and transplantation. Curr Opin Urol. 2002;12(6):479–84. https://doi. org/10.1097/01.mou.0000039446.39928.32. 84. Smith K, Neville-Jan A, Freeman KA, Adams E, Mizokawa S, Dudgeon BJ, et  al. The effectiveness of bowel and bladder interventions in children with

E. Morakis et al. spina bifida. Dev Med Child Neurol. 2016;58(9):979– 88. https://doi.org/10.1111/dmcn.13095. 85. Sawin KJ, Liu T, Ward E, Thibadeau J, Schechter MS, Soe MM, et al. The National Spina Bifida Patient Registry: profile of a large cohort of participants from the first 10 clinics. J Pediatr. 2015;166(2):444–50.e1. https://doi.org/10.1016/j.jpeds.2014.09.039. 86. Ottolini K, Harris AB, Amling JK, Kennelly AM, Phillips LA, Tosi LL. Wound care challenges in children and adults with spina bifida: an open-cohort study. J Pediatr Rehabil Med. 2013;6(1):1–10. https:// doi.org/10.3233/PRM-130231. 87. Alliaume A. [Fractures of the long bones in myelomeningocele]. Arch Fr Pediatr. 1950;7(3):294–5. 88. Lock TR, Aronson DD.  Fractures in patients who have myelomeningocele. J Bone Joint Surg Am. 1989;71(8):1153–7. 89. Gabriel K. Natural history of hip deformity in spina bifida. In: Sarwark JF, Lubicky JP, Shriners Hospitals for Children, editors. Caring for the child with spina bifida: Shriners Hospitals for Children, Symposium, Oak Brook, Illinois, April 14–16, 2000. 1st ed. Rosemont, IL: American Academy of Orthopaedic Surgeon; 2001. p. 89. 90. Carroll NC, Sharrard WJ.  Long-term follow-up of posterior iliopsoas transplantation for paralytic dislocation of the hip. J Bone Joint Surg Am. 1972;54(3):551–60. 91. Samuelsson L, Eklof O. Hip instability in myelomeningocele. 158 patients followed for 15 years. Acta Orthop Scand. 1990;61(1):3–6. 92. Barden GA, Meyer LC, Stelling FH III. Myelodysplastics--fate of those followed for twenty years or more. J Bone Joint Surg Am. 1975;57(5): 643–7. 93. Park TS, Cail WS, Maggio WM, Mitchell DC.  Progressive spasticity and scoliosis in children with myelomeningocele. Radiological investigation and surgical treatment. J Neurosurg. 1985;62(3):367– 75. https://doi.org/10.3171/jns.1985.62.3.0367. 94. Sarwark JF, Weber DT, Gabrieli AP, McLone DG, Dias L.  Tethered cord syndrome in low motor level children with myelomeningocele. Pediatr Neurosurg. 1996;25(6):295–301. https://doi. org/10.1159/000121143. 95. Menelaus MB. Dislocation and deformity of the hip in children with spina bifida cystica. J Bone Joint Surg Br. 1969;51(2):238–51. 96. Sharrard WJ. Paralytic deformity in the lower limb. J Bone Joint Surg Br. 1967;49(4):731–47. 97. Vankoski SJ, Sarwark JF, Moore C, Dias L.  Characteristic pelvic, hip, and knee kinematic patterns in children with lumbosacral myelomeningocele. Gait Posture. 1995;3(1):51–7. https://doi. org/10.1016/0966-6362(95)90809-7. 98. Sarwark JF, Lubicky JP, Shriners Hospitals for Children. Caring for the child with spina bifida: Shriners Hospitals for Children, Symposium, Oak Brook, Illinois, April 14–16, 2000. 1st ed. Rosemont, IL: American Academy of Orthopaedic Surgeon; 2001.

19  The Hip in Myelomeningocele 99. Duffy CM, Hill AE, Cosgrove AP, Corry IS, Mollan RA, Graham HK. Three-dimensional gait analysis in spina bifida. J Pediatr Orthop. 1996;16(6):786–91. 100. Drummond DS, Moreau M, Cruess RL.  Post-­ operative neuropathic fractures in patients with myelomeningocele. Dev Med Child Neurol. 1981;23(2):147–50. 101. James CC.  Fractures of the lower limbs in spina bifida cystica: a survey of 44 fractures in 122 children. Dev Med Child Neurol Suppl. 1970;22(Suppl 22):88+. 102. Kumar SJ, Cowell HR, Townsend P.  Physeal, metaphyseal, and diaphyseal injuries of the lower extremities in children with myelomeningocele. J Pediatr Orthop. 1984;4(1):25–7. 103. Boytim MJ, Davidson RS, Charney E, Melchionni JB.  Neonatal fractures in myelomeningocele patients. J Pediatr Orthop. 1991;11(1):28–30. 104. Martinelli V, Dell’Atti C, Ausili E, Federici E, Magarelli N, Leone A, et  al. Risk of fracture prevention in spina bifida patients: correlation between bone mineral density, vitamin D, and electrolyte values. Childs Nerv Syst. 2015;31(8):1361–5. https:// doi.org/10.1007/s00381-015-2726-2. 105. Szalay EA, Cheema A.  Children with spina bifida are at risk for low bone density. Clin Orthop Relat Res. 2011;469(5):1253–7. https://doi.org/10.1007/ s11999-010-1634-8. 106. Valtonen KM, Goksor LA, Jonsson O, Mellstrom D, Alaranta HT, Viikari-Juntura ER.  Osteoporosis in adults with meningomyelocele: an unrecognized problem at rehabilitation clinics. Arch Phys Med Rehabil. 2006;87(3):376–82. https://doi. org/10.1016/j.apmr.2005.11.004. 107. Kaufman BA, Terbrock A, Winters N, Ito J, Klosterman A, Park TS.  Disbanding a multidisciplinary clinic: effects on the health care of myelomeningocele patients. Pediatr Neurosurg. 1994;21(1):36–44. https://doi.org/10.1159/000120812. 108. Broughton NS. The hip. In: Broughton NS, Menelaus MB, editors. Menelaus’ orthopaedic management of spina bifida cystica. 3rd ed. London: W.B. Saunders; 1998. p. 135–44. 109. Morakis E, Wright J.  Evidence-based treatment of spina bifida. In: Alshryda S, Huntley J,

551 Banaszkiewicz PA, editors. Evidence-based paediatric orthopaedics: the best answers to clinical questions. 1st ed. New York, NY: Springer; 2016. 110. Frawley PA, Broughton NS, Menelaus MB. Anterior release for fixed flexion deformity of the hip in spina bifida. J Bone Joint Surg Br. 1996;78(2):299–302. 111. Mazur JM.  Orthopedic complications of myelomeningocele. In: Epps CH, Bowen JR, editors. Complications in pediatric orthopaedic surgery. Philadelphia, PA: Lippincott; 1995. p. 545–64. 112. Buisson JS, Hamblen DL.  Electromyographic assessment of the transplanted ilio-psoas muscle in spina bifida cystica. Dev Med Child Neurol Suppl. 1972;27:29–33. 113. Rothe MJ.  A cluster of anaphylactic reactions in children with spina bifida during general anesthesia: epidemiological features, risk factors, and latex hypersensitivity. Dermatitis. 1995;6(3):53–61. 114. Meeropol E, Frost J, Pugh L, Roberts J, Ogden JA.  Latex allergy in children with myelodysplasia: a survey of Shriners hospitals. J Pediatr Orthop. 1993;13(1):1–4. 115. Pearson Michele L, Cole Jeane S, Jarvis William R.  How common is latex allergy? A survey of children with myelodysplasia. Dev Med Child Neurol. 1994;36(1):64–9. https://doi. org/10.1111/j.1469-8749.1994.tb11767.x. 116. Tosi LL, Slater JE, Shaer C, Mostello LA.  Latex allergy in spina bifida patients: prevalence and surgical implications. J Pediatr Orthop. 1993;13(6):709–12. 117. Nieto A, Estornell F, Mazón A, Reig C, Nieto A, García-Ibarra F.  Allergy to latex in spina bifida: a multivariate study of associated factors in 100 consecutive patients. J Allergy Clin Immunol. 1996;98(3):501–7. https://doi.org/10.1016/ S0091-6749(96)70082-9. 118. Yassin MS, Sanyurah S, Lierl MB, Fischer TJ, Oppenheimer S, Cross J, et  al. Evaluation of latex allergy in patients with meningomyelocele. Ann Allergy. 1992;69(3):207–11. 119. Slater JE.  Latex allergy. J Allergy Clin Immunol. 1994;94(2):139–49. https://doi.org/10.1053/ai.1994. v94.a55437.

The Hip in Spinal Muscular Atrophy

20

Jill E. Larson and Brian Snyder

Introduction

good sitting balance and maintenance of pelvic alignment [5]. This chapter will focus on the musculoskeletal While there is still a dearth of data to supburden of spinal muscular atrophy (SMA) spe- port hip reconstruction, recent unpublished data cifically involving the hip, recognizing that the at our institution suggests that hip reconstruction intrinsic relationship between hip and spine in the non-ambulatory SMA type I and II patient deformity prevalent in non-ambulatory SMA population results in excellent radiographic outchildren often requires co-management of these comes with acceptable rates of complication and skeletal abnormalities. Traditionally, the symp- subluxation/dislocation recurrence [6]. With the toms associated with hip dislocations in SMA introduction of Nusinersen, an intra-thecal based have been reported to be relatively benign, thus medication to treat SMA, the prognosis of funcobservation rather than surgical intervention has tionality and hip stability in the non-ambulatory been recommended out of concern that the sur- population is changing. Thus, orthopedic surgeons gical risks in these medically complex patients should consider the changing paradigm in the outweigh the potential benefits associated with treatment of hip instability in the SMA population surgical hip reconstruction [1–3]. While most towards operative reconstruction; given demonagree that hip reconstruction is necessary in the stration of improved strength, functionality and ambulatory SMA type III populations, there is ambulatory status in SMA patients treated with controversy in the literature regarding the out- Nusinersen [7]. This chapter summarizes both comes of surgical hip reconstruction in the non-­ the non-operative and operative treatments for hip ambulatory SMA population. Some authors have instability and spinal deformity in the SMA popureported high complication rates, with persis- lation with a focus on how to treat concomitant hip tent hip dislocations and low patient satisfaction instability and neuromuscular scoliosis. scores [1, 4] where others have advocated for hip reconstruction on the basis of providing comfort,

Pathophysiology

J. E. Larson (*) Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, IL, USA e-mail: [email protected] B. Snyder Boston Children’s Hospital, Boston, MA, USA © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_20

Musculoskeletal Burden of Disease Spinal muscular atrophy (SMA) consists of a group of neuromuscular disorders characterized by degeneration of the cell body within 553

554

the alpha motor neurons located in the anterior horn of the spinal cord, leading to a lower motor neuron-type syndrome. This degeneration occurs due to insufficient production of survival motor neuron (SMN) protein, induced by mutations or deletions in SMN1 gene (exon 7 and/or 8) on chromosome 5q, inherited in an autosomal recessive manner. The extent of clinical involvement depends on the number of copies of the SMN2 “rescue” gene, which produces mostly non-­functional SMN protein (≤2 copies  =  more severe disease; 3–4 copies = milder disease). The resulting clinical phenotype and musculoskeletal manifestations stem from an associated progressive muscle atrophy, weakness and paralysis [2, 8, 9]. SMA phenotypes are classified based on age of onset and maximum motor function achieved: Type I patients are very weak infants unable to sit unsupported; Type II patients are non-ambulatory children who can sit independently; Type III patients ambulate as children, but may lose functional gait in adulthood, whereas Type IV remain ambulatory as adults [2].

Hip Instability Hip instability progressing to dislocation is common in patients with SMA due to the asymmetric weakness of hip gluteal muscles relative to the psoas and adductors which contribute to the development of coxa valga and persistent femoral anteversion. The imbalance of muscle forces around the hip is further exacerbated by a decrease in the normal weight-bearing force of a non-ambulatory patient with SMA.  A combination of these asymmetric forces causes the center of rotation to migrate from the center of the hip joint to the lateral margin of the acetabulum, where articulation of the femoral head compresses the cartilaginous apophysis. By the HueterVolkmann principle, this compression inhibits growth and ossification at the lateral margin of the acetabulum, resulting in posterolateral

J. E. Larson and B. Snyder

acetabular deficiency. Over time, hip flexion and adduction contractures develop as a consequence of a persistent sitting posture that imparts mechanical forces equivalent to a Barlow maneuver, encouraging postero-lateral hip subluxation (Fig. 20.1). “Hip instability progressing to dislocation is common in patients with SMA, due to the asymmetric weakness of hip gluteal muscles relative to the psoas and adductors which contribute to the development of coxa valga and persistent femoral anteversion”.

Etiology of Hip Pain in SMA The etiology of pain, and specifically pain associated with a hip dislocation, in SMA is unclear. Typically, the pain generator from a dislocated hip thought to be from the femoral head cartilage abrading and degrading against a non-­ cartilagenous surface, such as the adjacent ilium. If the femoral head cartilage is nestled within the soft tissues of the abductor musculature, there is theoretically less pain, but other issues arise including stiffness and contractures that can further precipitate pain. In a subluxated hip, the pain is generated from asymmetric forces on the femoral head within the acetabulum creating early cartilage wear and degeneration. The force of gravity can exacerbate the pain of a dislocated or subluxated hip, so ambulatory patients often experience more pain than non-ambulatory patients. However, non-ambulatory patients still report pain in the hip region in situations of hip subluxation/dislocation [4]. It should be noted that hip pain in SMA is less than in other neuromuscular conditions such as cerebral palsy where increased spasticity generates more muscle tension and increased contractures around the hip joint [10]. However, despite the low tone and lower rates of stiffness in the SMA population, patients still get hip contractures of the muscles and capsule around the joint, which further exacerbate hip pain.

20  The Hip in Spinal Muscular Atrophy

555

Coxa Valgus: ↑ Neck Shaft Angle

Excessive Femoral Anteversion

Posterolateral Migration of Hip

Acetabular Dysplasia: Deficient posterolateral coverage of hip

Fig. 20.1  Pathophysiology of hip instability in SMA

Natural History From a musculoskeletal perspective, SMA patients experience substantial morbidity. In a recent survey of Cure SMA (a not-for-profit organization dedicated to the treatment and cure of SMA) (CURESMA) (Table  20.1), 60–70% of patient members reported pain at the back, hip, groin or feet some or all the time.

Table 20.1 CURESMA member survey results of reported presence of pain Pain All the time Sometime Rarely Never

Type 1 (%) 6.5 52 19.5 22.1

Type 2 (%) 18.2 43.6 22.8 15.4

Type 3 (%) 18.4 50.5 16.8 14.2

Traditionally, the symptoms associated with hip dislocations have been reported to be relatively benign, thus observation rather than surgical intervention has been recommended out of concern that the surgical risks in these medically complex patients outweigh the potential benefits associated with surgical hip reconstruction [4]. Some have argued that surgical hip reconstruction should only be considered in ambulatory patients with good muscle strength [11]. Sporer and colleagues [4] retrospectively evaluated 24 patients with SMA type II and observed hip dislocation in 12 patients (15 hips) and hip subluxation in 8 patients (15 hips). However, only 2 of the 24 patients reported hip pain and none had problems with perineal care. Of 17 patients with SMA type III, only 2 patients had a dislocated hip (2 hips) and 5 patients had a subluxated hip (5 hips). No patients reported pain, difficulty with perineal

J. E. Larson and B. Snyder

556

hygiene or decubitus ulcers. Outcomes of SMA type II and III patients who underwent surgical hip reconstruction have been inconsistent. Zenios and colleagues reported on long-­ term follow-up (15.8 years) of nine SMA patients who underwent surgical reconstruction of the hip (eight bilateral and one unilateral). Three patients experienced post-operative complications and four patients had recurrent hip instability. Only one patient reported satisfaction after hip surgery [12]. In a retrospective review of SMA patients with hip instability treated operatively at our institution—using more contemporary surgical techniques—demonstrated more encouraging results. We evaluated SMA patients with hip instability who underwent proximal femoral varus derotational osteotomy  ±  open hip reduction and acetabuloplasty. Nineteen hips in 10 patients underwent surgical intervention between 2010 and 2016. Average follow-up time was 2.96  years (range 1.23–7.31). There was notable radiographic improvement in hip position and alignment after reconstructive hip surgery: the average pre-operative migration index for the cohort was 54.7% (±25.9%) with an average improvement of 43.4% (±29.7%); the average pre-operative acetabular index of those hips that underwent acetabuloplasty was 23.3° (±4.36°) with an average improvement of 10.3° (±5.5°). Of the 15 hips that had discontinuity in Shenton’s arc pre-operatively, 12 (80%) had complete restoration. Two patients had improvement in their ambulatory status (one received Nusinersen). Although 8 of 10 patients reported some hip pain at final follow-up, often the pain was related to the retained hip plates. Complications in the cohort included a slipped capital femoral epiphysis related to osteoporosis that required revision surgery and two hips had recurrent subluxation or dislocation (16%) related to profound hypotonia [6].

Epidemiology The incidence of SMA is approximately 1 in 11,000 live births [9]. Of those, approximately 58% have SMA type I, 29% have SMA Type II, and 13% have SMA Type III.  SMA type IV is rarely observed [13]. Using life table analysis of rate of birth preva-

lence and estimated survival for each type of SMA, the prevalence of symptomatic SMA cases in the United States in 2016 is 9429 [14].

Hip Instability and SMA Hip instability is common in SMA patients, and the incidence corresponds to SMA type with increased frequency in the non-ambulatory population. Hip subluxation occurs in approximately 30–40% of SMA type II patients and 10–30% of SMA type III patients. Hip dislocation occurs in approximately 30% of SMA type II patients and 20–30% in SMA type III patients [4, 12]. Many SMA patients demonstrate concomitant hip and spine deformity (Fig. 20.2). A radiographic study of 49 SMA patients revealed that 26 patients (53%) had both hip and spine pathology. Of those patients, 58% had concurrent spine and hip abnormalities; 38% developed a spinal deformity prior to hip instability, and 4% developed hip instability prior to spinal deformity [15]. Thus, orthopedic surgeons must evaluate the entire axial skeleton and consider how surgical interventions will impact overall sitting and standing posture, position and function of the hip, pelvis and spine. “Hip instability is common in SMA patients, and the incidence corresponds to SMA type with increased frequency in the non-ambulatory population”.

Scoliosis and SMA Given high prevalence of concomitant hip and spine deformity for patients with SMA, associated pelvic obliquity may arise from an infra- or suprapelvic cause. Hence, the pediatric hip surgeon should have knowledge of the diagnosis and treatment of scoliosis when considering the treatment of hip pathology in patients with SMA. Scoliosis is highly prevalent in children with SMA type I and type II, with incidence of 60–90% and initial presentation in early childhood [2, 16]. The onset of scoliosis in SMA is earlier than idiopathic scoliosis, presenting at an average age of 8.8 years [17]. The curve progresses with age and is associated with decreased pulmonary function

20  The Hip in Spinal Muscular Atrophy

a

557

b

32 month old SMA II

d

c

5 year old

e

10 Year old

7 year old

f

11 year old

Fig. 20.2  Associated hip and spine pathology in SMA: (a) Bilateral symmetric coxa valgus with 50% migration index, mild acetabular dysplasia, level pelvis, no scoliosis; (b) Progressive, symmetric posterolateral hip subluxation, mild acetabular dysplasia, level pelvis, no scoliosis; (c, d) Neuromuscular scoliosis contributes to associated

14 year old

pelvic obliquity and “windswept posture” (adducted right hip, abducted left hip); failure to control pelvic obliquity with initial spinal instrumentation contributes to vertical migration of the right hemipelvis and relative acetabular dysplasia. (e, f) Progressive uncovering of the right femoral head to near dislocation of the right hip

J. E. Larson and B. Snyder

558

[17, 18]. For SMA type I and II patients, scoliosis with Cobb angle >20° should be monitored every 6  months until skeletally maturity and yearly after skeletally maturity [19].

Clinical Presentation The clinical findings of SMA include progressive, symmetric, proximal greater than distal muscle weakness affecting the legs more than the arms, sparing the facial muscles, but involving bulbar muscles. Tongue fasciculation and absent muscle stretch reflexes are hallmark features of SMA Types I and II. Sensation and cognition are typically preserved for all Types. Semi-annual examinations should be performed to assess strength, joint range of motion, motor function, and gait (where appropriate) [20–24] (see Essential Clinical Tests). The primary orthopedic surgical issues involve the axial skeleton (i.e. pelvis and spine). Besides affecting the musculoskeletal system, these patients suffer from respiratory, nutritional and metabolic dysfunction that contribute to morbidity and mortality. In Types I and II, weakness of the intercostal muscles, with relative sparing of the diaphragm, is associated with development of a “bell-shaped” thorax that influences pulmonary function and the need for supplemental respiratory support [25]. Therefore, these patients require comprehensive care involving the participation of multiple disciplines to manage the medical comorbidities, especially during the peri-operative phase. Patients should be evaluated on a regular basis by pulmonologists, neurologists, gastroenterologists, feeding specialists, physiatrists and orthopedists. Pulmonary function testing is essential to assess fitness for surgery, where indicated. As these patients are mostly non-ambulatory, they are susceptible to osteopenia and insufficiency fractures depending on disease severity and weight-bearing capacity. As such, an evaluation of their ‘bone health’ should also be performed regularly. This evaluation typically includes 25OH vitamin D levels and DEXA scanning; this is particularly helpful for pre-­operative assessment of bone quality and for cases where internal fixation devices may be used.

Essential Clinical Tests

• Assess strength and range of motion testing for all joints of the lower extremities (especially noting contractures of hip flexors/adductors, hamstrings and gastrocnemius-soleus complex that affect sitting/standing posture and stability) • Detailed exam of the hip to evaluate for pain and/or instability (Galeazzi, Barlow, Ortolani) • Detailed exam of the spine to evaluate for scoliosis, kyphosis, pelvic obliquity and sitting balance • Assess the ability to sit with or without assistance (including “tri-pod” arm support) • Assess the ability to stand and transfer in/out of wheel chair with or without assistance • Perform an observational gait analysis with attention to the following areas: –– Limp  =  leg length difference, hip instability or pain –– Hyperlordosis = weak gluteus maximus and/or psoas contracture –– Abductor lurch  =  weak gluteus medius –– Crouched posture  =  weak quadriceps, gluteus maximus, contractures psoas and/or hamstrings –– Toe walking  =  contractures of gastroc-soleus complex and weak tibialis anterior –– Genu recurvatum = weak quadriceps, contracture gastrocnemius

Imaging A formal radiographic surveillance program to monitor for hip instability or spinal deformity in SMA has not been established in the literature. However, if a patient is greater than 6 months of age upon initial evaluation, an AP pelvis should be obtained to document baseline the hip status. Subsequent radiographs should be obtained if

20  The Hip in Spinal Muscular Atrophy

there is any concerning change in examination, such as decreased hip range of motion, increasing contractures or positive Galeazzi, Ortoloni and/or Barlow findings. This orthopedic evaluation should occur routinely at 6  month intervals with XRs of the pelvis obtained yearly [3]. Similarly, if the clinical exam is concerning for increasing spinal curvature or asymmetric sitting balance, a sitting PA and lateral spine XR should be obtained. A patient reporting new hip, back or thigh pain should prompt further radiographic evaluation of the hip and spine. An AP and lateral Pelvis is typically obtained in the supine position in young or non-­ambulatory patients. If there is a concern for hip instability in an ambulatory patient, then a standing AP pelvis should be obtained to document the hip position with weight-bearing forces. DEXA scans should be obtained on all SMA patients to assess for osteopenia and guide the medical management. DEXA scans should also be routinely evaluated pre-operatively to assess for bone quality and to help guide surgeons on instrumentation choices in the setting of poor bone quality. Finally, surgical planning for hip reconstruction should include a CT scan of the pelvis to assess for acetabular dysplasia and identify the areas of poor femoral head coverage to guide surgical decision making on the need for and type of pelvic osteotomy.

Essential Imaging Tests

• X-ray (typically supine) AP and lateral Pelvis—assess femoral neck shaft angle for coxa valga, migration percentage (MP) and acetabular index (AI)/acetabular angle (angle from a horizontal line drawn from the inferior margin of the tear drops to the lateral edge of the acetabular roof; used when the triradiate cartilage is closed) at least yearly (similar to CP surveillance) (Fig. 20.3) • Standing AP hips in capable patients to provoke hip instability with weight bearing

559

• X-ray PA/lateral Sitting/Standing spine—assess for scoliosis, associated Cobb angle and pelvic obliquity • DEXA scan to assess for osteopenia and need for medical management • Pre-operative CT scanning for assessment of acetabular dysplasia

Non-operative Management Medical Management Traditionally, the treatment of SMA has been supportive, focusing on maximizing pulmonary and musculoskeletal function with use of bracing and therapy services. Recently however, the US Food and Drug Administration (FDA) approved Nusinersen, the first disease modifying drug treatment for SMA.  Nusinersen is an antisense oligonucleotide drug that modifies pre-­messenger RNA splicing of the SMN2 gene, thereby promoting increased production of a biologically effective, full-length SMN protein [7]. It is administered to the cerebrospinal fluid via intrathecal injection to deliver the drug directly to the spinal motor neurons which degenerate in patients with SMA [26]. Predicated on the positive results of two phase 3, randomized placebo controlled studies, Nusinersen was officially released for the treatment of all forms of SMA in December 2016 [19, 27]. While long-term follow-up studies of the impact of Nusinersen are pending, initial results of infants treated with intrathecal Nusinersen have improved motor function and increased survival without the permanent use of assisted ventilation compared to those undergoing a sham procedure (ENDEAR study) [7]. These infants continue to be followed in an open-label extension study (SHINE) [28], to assess the effects of longer treatment duration on motor function and quality of life. The results of a clinical trial investigating the use of Nusinersen for older SMA type II and III children (2–12  years of age at enrollment) is forthcoming (CHERISH trial) [5].

560

J. E. Larson and B. Snyder

Fig. 20.3  Monitoring of the hip in SMA

Nusinersen and other drugs in the development pipeline have the potential to change the natural history of SMA, its clinical symptoms and management approaches. Even with the introduction of disease modifying treatments, however a multidisciplinary approach is paramount to the comprehensive management of SMA. “Nusinersen and other drugs in the development pipeline have the potential to change the natural history of SMA, its clinical symptoms and management approaches”.

Use of Physical Therapy and Orthoses Physical therapy alone has not been demonstrated to maintain hip stability or to prevent hip instability. Use of hip abduction splints to concentrically relocate an unstable hip may help prevent hip adduction contractures and maintain functional range of motion for toileting, hygiene and activities of daily living, but will not prevent

progressive hip instability, especially if there is associated pelvic obliquity and scoliosis. The objectives for musculoskeletal management in SMA type II sitters are to restore or promote function and mobility by preventing joint contractures and scoliosis. Modalities for stretching include techniques that can be achieved manually with orthoses, splints, active-assistive stretching and serial casting to maintain functional range of motion about the shoulder, elbow, wrist, hip, knee and ankles in SMA type II patients. Fujak and colleagues recommended utilization of a selfpropelled or power assisted wheelchair in nonambulatory children as early as age 3 to facilitate functional independence. Additionally, a walking apparatus (e.g., swivel walker or gait trainer) or passive support for standing (e.g. standing frame) should be initiated as early as age 2 for patients with appropriate head and trunk control, or with the support of a soft thoracolumbar orthosis for those lacking sufficient trunk control. Aquatic based therapies are useful for strengthening of core and hip girdle musculature and simulated walking.

20  The Hip in Spinal Muscular Atrophy

In patients with SMA type III, musculoskeletal therapy should focus on improving core, hip girdle and quadriceps muscle strength, balance and functional mobility. Both closed and open chain strengthening using dynamic and static forms should be incorporated into an exercise program to maintain functional ambulation. Many SMA patients become overweight and obese after reaching puberty, making it difficult for already weak muscles to support the trunk during walking. For example, the gluteus medius must generate nearly three times the force of body weight to support the trunk during single legged stance, so that for every pound in body mass over ideal body weight, means that the gluteus medius must generate an additional 3 lb of force. Thus, maintaining a healthy weight is important for the longevity of bone and joint health in patients with SMA

Essentials of Non-operative Management of the Hip

• Daily stretching of Psoas, adductors, hamstrings • Mobilize into stander or gait trainer at least 2  h/day with hips moderately abducted • Sleep in hip abduction brace • Strengthening of hip girdle musculature (land and water based physical therapies)

Non-operative Management of Scoliosis As sitting balance and pelvic obliquity can significantly impact the stability of a hip in a non-­ ambulatory SMA patient, any substantial associated spinal curvature also needs to be addressed. While bracing in neuromuscular scoliosis has not been shown to slow curve progression, bracing with a thoracolumbar-sacral orthoses (TLSO) may be used to augment truncal support in patients unable to sit or stand independently. Bracing is particularly critical for SMA type I and type II patients to promote postural

561

stabilization and to maximize function. There is no consensus on the type of brace to be used, but a circumferential, anterior opening, soft spinal thoracolumbar orthosis with a large anterior window to allow for abdominal expansion and to facilitate diaphragmatic breathing is typically recommended [19]. Cervical bracing can be used for head support for safety and transportation. Thoracic bracing is not typically used during walking in SMA type III, as it may adversely affect ambulatory ability.

Essentials of Non-operative Management of Scoliosis

• TLSO should be considered to provide trunk support in sitting and standing and to partially correct scoliotic curves >20° (use thoracic and/or lumbar pads at curve apex, cut out window opposite pad to allow translation of trunk) • Use to control pelvic obliquity and maintain stable posture, and position the head over a leveled pelvis • Monitor curves for progression every 6  months with radiographic evaluation (sitting AP and lateral view in and out of TLSO) • Circumferential, soft spinal thoracolumbar orthosis with a large anterior window to allow for abdominal expansion and to facilitate diaphragmatic breathing

Operative Management  perative Management of Hip O Instability As Nusinersen is changing the trajectory of SMA, hip reconstructive surgery should be considered as an SMA patient’s ambulatory status may improve, and the surgical complication rate is similar to reconstructive hip surgery in other neuromuscular conditions such as Cerebral Palsy [6, 29]. The consensus opinion of orthopedic surgeons participating in a recent European Neuromuscular Centre

562

(ENMC) International Workshop suggested that unilateral and bilateral hip instability should be surgically managed in patients with significant pain or impaired function [19]. Consideration should be given to the radiographic appearance of the hip including the migration index or percent coverage of the femoral head by the acetabulum and the depth of the acetabulum indicated by the acetabular index when contemplating surgical intervention (Fig. 20.3).

Indications for Surgery Concerning surgery of the hip in SMA, there is little controversy for children who can or may walk: reconstruct the hip if it is unstable (i.e. subluxated or dislocated). For non-ambulatory children undergoing Nusinersen therapy, surgical intervention may be indicated for the maintenance of pelvic alignment, to treat or prevent the onset of painful arthritis, to address pelvic obliquity associated with scoliosis, and to improve sitting balance. Surgical reconstruction of coxa valga with an associated dysplastic acetabulum is less complicated in children less than 10  years of age when the tri-radiate cartilage is open. However, Nusinersen is not curative and therefore persistent muscle weakness and asymmetry can still lead to recurrent coxa valga and hip subluxation in younger patients (less than 6  years of age), requiring revision surgery at a later stage. “For non-ambulatory SMA patients undergoing Nusinersen therapy, surgical hip reconstruction may be indicated for the maintenance of pelvic alignment, to treat or prevent the onset of painful arthritis, to address pelvic obliquity associated with scoliosis, and to improve sitting balance”. In the young adult patient with a chronically dislocated hip, leave the hip dislocated if painless. However, if the patient has pain related to coxa arthrosis, then a salvage procedure may be indicated, such as a McHale procedure (femoral head resection arthroplasty plus proximal femoral valgus osteotomy; Chap. 18, Fig. 18.26) [30].

Surgical Technique Before addressing the osseous component of hip instability, orthopedic surgeons must be aware of

J. E. Larson and B. Snyder

associated joint contractures that often need to be surgically released to facilitate range of motion and upright posture in stander/gait trainer. Contractures of lower extremities develop early and increase progressively with age in non-­ ambulatory patients with SMA.  Typically, contractures involve muscles crossing two joints including: (1) the hamstrings (causing knee flexion contractures), (2) the psoas, adductor longus and gracilis (contributing to flexion and adduction contractures), and (3) the gastrocnemius (causing ankle equinus) [18]. Joint contractures with functional limitations affecting gait are less common in SMA type III but should be monitored nonetheless [31]. Judicious fractional lengthening of musculotendinous contractures should be considered to improve functional joint range of motion and to improve sitting/standing posture, recognizing that lengthening a weak muscle weakens it further. Specifically for the hip, there should be at least 45° of abduction to perform a proximal varus derotational osteotomy (VDRO). Thus, an open adductor tenotomy/myotomy may be required in concert with the VDRO to achieve appropriate post operative hip range of motion. “Before addressing the osseous component of hip instability, orthopedic surgeons must be aware of associated joint contractures that often need to be surgically released to facilitate range of motion and upright posture in stander/gait trainer”. A proximal varus derotational osteotomy (VDRO) is often combined with acetabuloplasty if there is significant posterior/lateral subluxation of the femoral head with associated dysplasia of the acetabulum (Fig. 20.4). A preoperative CT scan of the pelvis is useful for surgical planning of the acetabuloplasty, although not universally indicated. An intra-operative hip arthrogram can also help facilitate the decisionmaking process on whether or not to perform an acetabuloplasty, noting that if the capital epiphysis is stably contained by the cartilaginous apophysis and labrum outlined by arthrogram during provocative maneuvers, the need for an acetabuloplasty may be obviated given the remodeling potential of the hip joint in young children. Occasionally, an open hip reduction

20  The Hip in Spinal Muscular Atrophy

563

a

b

c

Fig. 20.4  Surgical procedure Varus Derotation Osteotomy: (a) Initial placement of guidewires for a 90° fixed angle blade plate; (b) chisel and derotational k-wire placement prior to subtrochanteric osteotomy with

improved varus alignment of the femoral head; (c) final instrumentation with blade plate noting medialization of the femoral shaft and improved but still deficient femoral head coverage suggesting need for acetabuloplasty

564

J. E. Larson and B. Snyder

a

b

Fig. 20.5  Surgical procedure peri-acetabular osteotomies: (a) open tri-radiate cartilage-Pemberton-Degas-San Diego type reshaping acetabuloplasty with hinging through the triradiate cartilage and autograft; (b) Pre-op images show that after pathological fracture treated elsewhere, this patient

is required for severely dislocated hips that do not reduce on pre-operative frog-leg lateral images. If the tri-radiate cartilage is open, then a Pemberton-Dega-San Diego-type reshaping acetabuloplasty is performed (Fig. 20.5a); however if the tri-radiate cartilage is closed and the patient has good bone quality (and ambulatory potential), we prefer to perform a Chiari osteotomy (Fig. 20.5b).

was fixed in valgus and developed hip subluxation related to underlying acetabular dysplasia. A revision VDRO was performed a 100% displacement Chiari osteotomy and Latarjet like bone block anteriorly (small screw in tri-cortical iliac crest graft) to address multi-directional instability

Essential Surgical Techniques

• Must have sufficient (>45°) abduction prior to VDRO. Perform an open adductor tenotomy if insufficient abduction • For VDRO, perform a sub-vastus lateral approach to proximal femur • Salter (ilio-inguinal) approach to anterior hip joint for open reduction (where needed) and for acetabuloplasty

20  The Hip in Spinal Muscular Atrophy

• Correct valgus femoral neck-shaft angle to 100° for non-walkers, 120° for walkers • During VDRO, shorten the femur to relax hamstring and rectus femoris contractures • Use resected femoral ring from the VDRO as structural auto-graft for interposition into the iliac osteotomy during acetabuloplasty • During VDRO, derotate the femur to achieve ~15° anteversion • Use rigid hip plate for fixation to allow early range of motion and allow patient to sit up for pulmonary toilet • Perform concomitant acetabuloplasty if: –– The acetabulum is shallow (AI >25°) –– The hip does not concentrically reduce into the socket after VDRO (check with intra-operative hip arthrogram) • The choice of acetabular osteotomy depends on the status of the tri-radiate cartilage (TRC), potential to walk, and extent of degenerative changes in the hip joint. In particular: –– If the TRC is open, perform an incomplete peri-acetabular osteotomy above the sourcil, extending the osteotomy posteriorly, splitting the posterior column, and hinging on the TRC to correct posterolateral acetabular deficiency –– If the TRC is closed and the patient has good bone quality and ambulatory potential, then perform a Chiari osteotomy • AVOID SPICA cast immobilization following hip reconstruction to avoid osteopenia and post-operative insufficiency fractures. Instead, immobilize with a hip abduction brace that allows flexion at hip joint

Peri-operative Management

• Epidural anesthesia  ×  3  days post-­ operatively + anti-inflammatories (minimize narcotic use to avoid respiratory depression)

565

• Diazepam if muscle spasms • A-frame leg casts or Hip Abduction Brace for 3 weeks post-operatively • Be vigilant for skin breakdown and pressure ulcers. Frequent changes in position, air mattress are required • Early mobilization into sitting posture for pulmonary toilet • Chest physiotherapy, incentive spirometry for pulmonary toilet • Deep vein thrombosis (DVT) prophylaxis: pneumatic compression boots + low molecular weight heparin for adolescents >16 years of age • Active assisted range of motion at knee and ankle post-op day 2 • Aquatherapy starting at 3 weeks after surgery to strengthen hip girdle, trunk, quadriceps, and allow for simulated walking • Weight bearing/standing ± gait training (for SMA Type III) at 6–8  weeks after surgery if radiographic evidence of sufficient bone healing at osteotomies

Operative Pitfalls

• Do both hips, even if only one hip subluxated, since unilateral VDRO may result in pelvic obliquity and contralateral hip is at high risk for later instability • Poor bone quality related to osteopenia increases risk of failed fixation –– Pre-treat with vitamin D  ± Bisphosphonates to improve bone density at least 6 months before proposed surgery –– Use higher angle hip plate (120°) with blade or fixation screws oriented along femoral neck axis to minimize cantilever moment and risk of failure at bone-implant interface –– Augment healing of osteotomy with demineralized bone matrix –– Supplement stability of pelvic osteotomy with interposed structural ring allograft bone (from the VDRO or fibular strut allograft)

566

• If there is coexistent hip and spine deformity, consider correcting the hip deformity first (especially if painful) since iliac screw fixation to stabilize the pelvis as part of spinal instrumentation may interfere with the acetabuloplasty. However if there is marked pelvic obliquity associated with the spinal deformity, the spine may need to be addressed first to establish level pelvic orientation. • Counsel families on the risk of recurrent hip subluxation/dislocation (Fig. 20.6) –– Associated pelvic obliquity related to spinal deformity may increase the risk for recurrent hip instability— treat spinal deformity with instrumentation to pelvis

Fig. 20.6  Risk of recurrent hip subluxation/dislocation: 6-year-old Male with SMA type I who underwent bilateral adductor tenotomies, bilateral VDRO with extension osteotomies, open reduction and acetabuloplasty of left

J. E. Larson and B. Snyder

Operative Management of Scoliosis As over half of SMA patients have both hip and spine pathology, a component of the hip treatment algorithm may include posterior spinal fusion to level the pelvis and correct pelvic obliquity [15]. The treatment of scoliosis in patients with SMA is complicated because a limited time-­frame exists in which these patients have sufficient lung capacity to successfully undergo spinal surgery. If spinal surgery is not performed within this narrow window, the opportunity to maximize medical management is lost and patients are less likely to survive spinal surgery [32]. The goal of surgical intervention is to maximize physical function and pulmonary function. While pulmonary function in SMA types II and III continues to decline after scoliosis surgery, the rate of decline is less marked and

hip. Two-year follow-up demonstrates concentrically reduced hip on the Left with good femoral head coverage. However, the right-side demonstrates a dislocated hip

20  The Hip in Spinal Muscular Atrophy

prolonged survival of patients justifies the aggressive management of scoliosis to prevent deformity progression and improve sitting comfort [17, 33]. “As over half of SMA patients have both hip and spine pathology, a component of the hip treatment algorithm may include posterior spinal fusion to level the pelvis and correct pelvic obliquity” Surgical intervention (Fig. 20.7) should be considered according to curve magnitude (i.e. major curve Cobb angle ≥50°) and rate of progression (≥10° per year) [19]. Other factors, such as decreasing respiratory function, parasol rib deformity,

567

hyperkyphosis and adverse effects on functional mobility, pelvic obliquity, and trunk imbalance should also be considered. Mesfin and colleagues proposed a surgical algorithm for selecting spinal construct type and levels [11]: (1) growing rod anchored to pelvis with skeletal age less than 10 years and curve magnitude greater than 70°, and (2) posterior spinal fusion with pedicle screws at most levels T2 to pelvis for patients with open triradiate cartilage and curve magnitude greater than 70°, skeletal age greater than 10 years. Recent literature has supported the use of growing rods constructs in the treatment of scoliosis in the skeletally immature SMA patient population

Fig. 20.7  Surgical indication for spinal instrumentation growing rod instrumentation with improved pelvic obliqin SMA: 9-year-old male with SMA type II with pre-­ uity and spinal alignment operative right thoracolumbar curve of 58° underwent

568

[34, 35]. McElroy and colleagues retrospectively reviewed 15 patients with SMA with an average of 54-month follow-up and compared them to 80 growing rod patients with early onset idiopathic scoliosis. They found an improvement in trunk height in (average increase of 8.7 ± 3.2 cm) and space available for lung ratio (increased from 0.86  ±  0.15 to 0.94  ±  0.21) in the SMA cohort while also controlling curve magnitude (Cobb angle decreased from 89 ± 19° to 55 ± 17°) and pelvic obliquity. Although they noted a longer hospital stay in the SMA cohort, there was a lower complication rate than the age matched growing rod patients with idiopathic scoliosis [34].

Classic Papers Current Concepts Review. The Diagnosis and Orthopaedic Treatment of Childhood Spinal Muscular Atrophy, Peripheral Neuropathy, Friedreich Ataxia, and Arthrogryposis. Shapiro, Frederic; Specht, Linda. JBJS: November 1993 Volume 75 - Issue 11 - p 1699–1714. This classic review from Shapiro and Specht discusses the need for operative treatment of displaced hips, particularly for Type II SMA. They discussed the use of a varus osteotomy of the proximal femur ± capsulorrhaphy to prevent hip dislocation but also stressed that the risk of recurrent coxa valga and repeat subluxation was high, often necessitating repeat surgery. Spinal muscular atrophy. Lunn, M.R. and C.H. Wang,. Lancet, 2008. 371(9630): p. 2120– 33. Comprehensive review of the clincal presentation, molecular pathogenesis, diagnosis, treatment and research directions concerning SMA.

Key Evidence Finkel, R.S., et  al., Nusinersen versus Sham Control in Infantile-Onset Spinal Muscular Atrophy. N Engl J Med, 2017. 377(18): p.  1723–1732. A randomized, double-blind, sham-controlled, phase 3 efficacy and safety trial of Nusinersen found a significantly higher

J. E. Larson and B. Snyder

percentage of infants in the Nusinersen group than in the control group had a motor-milestone response in interim analysis (21 of 51 infants [41%] vs. 0 of 27 [0%], P  72,400 specimens. Eur J is changing the trajectory of functionalHum Genet. 2012;20(1):27–32. 10. Abresch RT, et  al. Assessment of pain and health-­ ity and ambulation in SMA patients. related quality of life in slowly progressive neuThus, hip reconstructive surgery should romuscular disease. Am J Hosp Palliat Care. be considered in a clinically or radio2002;19(1):39–48. graphically unstable hip as the surgical 11. Mesfin A, Sponseller PD, Leet AI.  Spinal muscular atrophy: manifestations and management. J Am Acad complication rate is similar to reconOrthop Surg. 2012;20(6):393–401. structive hip surgery in other neuromus 12. Zenios M, et al. Operative treatment for hip subluxcular conditions. ation in spinal muscular atrophy. J Bone Joint Surg Br. • More than 50% of SMA patients dem2005;87(11):1541–4. 13. Ogino S, Wilson RB.  Spinal muscular atrophy: onstrate concomitant hip and spine molecular genetics and diagnostics. Expert Rev Mol deformity; therefore, a clinician must Diagn. 2004;4(1):15–29. evaluate the entire axial skeleton and 14. Lally C, et  al. Indirect estimation of the preva consider how surgical interventions will lence of spinal muscular atrophy Type I, II, and III in the United States. Orphanet J Rare Dis. 2017; impact overall sitting/standing posture, 12(1):175. position and function of the hip, pelvis 15. Patel J, Shapiro F. Simultaneous progression patterns and spine of scoliosis, pelvic obliquity, and hip subluxation/dislocation in non-ambulatory neuromuscular patients: an approach to deformity documentation. J Child Orthop. 2015;9(5):345–56. 16. Mercuri E, Bertini E, Iannaccone ST. Childhood spinal muscular atrophy: controversies and challenges. References Lancet Neurol. 2012;11(5):443–52. 17. Phillips DP, et al. Surgical treatment of scoliosis in a 1. Canavese F, Sussman MD. Strategies of hip managespinal muscular atrophy population. Spine (Phila Pa ment in neuromuscular disorders: Duchenne Muscular 1976). 1990;15(9):942–5. Dystrophy, Spinal Muscular Atrophy, Charcot-­ 18. Fujak A, et al. Contractures of the lower extremities Marie-­ Tooth Disease and Arthrogryposis Multiplex in spinal muscular atrophy type II. Descriptive cliniCongenita. Hip Int. 2009;19(Suppl 6):S46–52. cal study with retrospective data collection. Ortop 2. Lunn MR, Wang CH. Spinal muscular atrophy. Lancet. Traumatol Rehabil. 2011;13(1):27–36. 2008;371(9630):2120–33. 19. Finkel RS, Sejersen T, Mercuri E. 218th ENMC 3. Prior TW, Finanger E.  Spinal muscular atrophy. In: International Workshop, in Revisiting the Consensus on Adam MP, et al., editors. GeneReviews((R)). Seattle, Standards of Care in SMA. Naarden, The Netherlands: WA: University of Washington; 1993. ENMC SMA Workshop Study Group; 2016. 4. Sporer SM, Smith BG.  Hip dislocation in patients 20. Glanzman AM, et  al. The Children’s Hospital of with spinal muscular atrophy. J Pediatr Orthop. Philadelphia Infant Test of Neuromuscular Disorders 2003;23(1):10–4.

Take Home Messages

570 (CHOP INTEND): test development and reliability. Neuromuscul Disord. 2010;20(3):155–61. 21. Mazzone E, et al. Six minute walk test in type III spinal muscular atrophy: a 12 month longitudinal study. Neuromuscul Disord. 2013;23(8):624–8. 22. Mazzone E, et  al. Assessing upper limb function in nonambulant SMA patients: development of a new module. Neuromuscul Disord. 2011;21(6):406–12. 23. Vuillerot C, et al. Responsiveness of the motor function measure in patients with spinal muscular atrophy. Arch Phys Med Rehabil. 2013;94(8):1555–61. 24. Montes J, et  al. Six-Minute Walk Test demonstrates motor fatigue in spinal muscular atrophy. Neurology. 2010;74(10):833–8. 25. Livingston K, et al. Parasol rib deformity in hypotonic neuromuscular scoliosis: a new radiographic definition, and a comparison of short-term treatment outcomes with VEPTR and growing rods. Spine (Phila Pa 1976). 2015;40(13):E780–6. 26. Evers MM, Toonen LJ, van Roon-Mom WM.  Antisense oligonucleotides in therapy for neurodegenerative disorders. Adv Drug Deliv Rev. 2015;87:90–103. 27. Mendell JR, et  al. Single-dose gene-replacement therapy for spinal muscular atrophy. N Engl J Med. 2017;377(18):1713–22. 28. A study for participants with spinal muscular atrophy (SMA) who previously participated in nusin-

J. E. Larson and B. Snyder ersen (ISIS 396443) investigational studies. (SHINE). https://clinicaltrials.gov/ct2/show/NCT02594124. (Identification No. NCT02594124). 29. Ruzbarsky JJ, et al. Risk factors and complications in hip reconstruction for nonambulatory patients with cerebral palsy. J Child Orthop. 2013;7(6):487–500. 30. McHale KA, Bagg M, Nason SS.  Treatment of the chronically dislocated hip in adolescents with cerebral palsy with femoral head resection and subtrochanteric valgus osteotomy. J Pediatr Orthop. 1990;10(4):504–9. 31. Wang HY, et al. Joint range of motion limitations in children and young adults with spinal muscular atrophy. Arch Phys Med Rehabil. 2004;85(10):1689–93. 32. Robinson D, et  al. Scoliosis and lung function in spinal muscular atrophy. Eur Spine J. 1995;4(5): 268–73. 33. Chng SY, et al. Pulmonary function and scoliosis in children with spinal muscular atrophy types II and III. J Paediatr Child Health. 2003;39(9):673–6. 34. McElroy MJ, et  al. Growing rods for scoliosis in spinal muscular atrophy: structural effects, complications, and hospital stays. Spine (Phila Pa 1976). 2011;36(16):1305–11. 35. Chandran S, et  al. Early treatment of scoliosis with growing rods in children with severe spinal muscular atrophy: a preliminary report. J Pediatr Orthop. 2011;31(4):450–4.

The Hip in Muscular Dystrophy

21

Deborah M. Eastwood

Introduction The muscular dystrophies (MD) are a group of inherited conditions characterized by gradual, progressive muscle degeneration with weakening, increasing disability and premature death [1, 2, 4]. Duchenne Muscular Dystrophy (DMD) is the most common and severe type. Despite substantial advances in understanding the pathophysiology, a treatment that significantly alters the natural history of the muscle pathology in this and other MDs remains out of reach although the use of oral steroid medication has delayed deterioration and perhaps truly improved some aspects of the condition. When applying surgical principles to the management of these dystrophies, their progressive nature and increasing muscle weakness must be taken into account. Although in many neuromuscular conditions, hip subluxation and dislocation occur commonly, this association is not seen frequently in patients with MDs such as DMD.

D. M. Eastwood (*) Great Ormond St. Hospital for Children, London, UK The Catterall Unit, The Royal National Orthopaedic Hospital, Stanmore, Middlesex, UK © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_21

Pathophysiology Dystrophin, a large protein found mainly in skeletal and cardiac muscle, is part of a protein complex that acts as an anchor connecting the cytoskeleton of each muscle cell with the lattice of proteins and other molecules in the extracellular matrix. Dystrophin is thus an important element in the interactions of the muscle membrane and the extracellular environment and is essential in maintaining muscle integrity [1, 4, 6]. The dystrophin gene is located on the short arm of Chromosome X (near the p21 locus). Mutations in this gene are responsible for both the Duchenne and Becker forms of MD and they occur almost exclusively in boys. Muscle cells without enough functional dystrophin become damaged as muscles contract and relax with use. Abnormalities in dystrophin production and/ or function thus lead to muscle dysfunction. In the absence of dystrophin, there is progressive leakage of intracellular contents through the disrupted muscle membrane, which accounts for the high serum levels of Creatine Phosphokinase (CK). The CK itself excites an inflammatory response within the muscle tissue via mast cells, which leads to muscle fibrosis and further loss of muscle function. Dystrophin may also play a role in intracellular signaling; for example, it participates in the regulation of nitric oxide synthesis [4]. Overall, mechanical destabilization rather 571

D. M. Eastwood

572

than signaling dysfunction is the primary cause of muscle fibre necrosis [6]. Over time, the damaged muscle fibres undergo necrosis leading to the weakness and cardiomyopathy characteristic of Duchenne and Becker muscular dystrophies [1, 4, 6]. Mutations that prevent the production of any functional dystrophin (less than 5% of normal levels) cause DMD, whilst the in-frame mutation linked to Becker MD results in the production of an abnormal version of dystrophin (which retains some function). In Becker MD, overall levels of dystrophin are 30–80% of normal. Both DMD and Becker MD are dystrophinopathies [7]. Molecular-based therapies including gene replacement using adenovirus vectors have been trialled but considerable challenges remain in translating this work to clinical practice [4, 8]. The differing mutations account for the phenotypic differences between DMD and Becker MD and also within each MD itself. Certain aspects of the phenotype, such as early wheelchair dependency or delayed onset of dilated cardiomyopathy, appear to be related to specific exon deletions. Classically, in DMD, there is proximal to distal progression of the muscle weakness with pseudohypertrophy of the calf complex [2]. Other forms of MD are caused by alterations in the dystrophin-associated-glycoproteins encoded by genes outside the X-chromosome [1, 4]. With respect to the hip in DMD, the hip extensors (primarily gluteus maximus) are first to be Fig. 21.1  Timeline of the effects of DMD on patient function

0

5

affected, resulting in an increased anterior pelvic tilt. The tightness in tensor fascia latae may accentuate the lumbar lordosis which directs the body weight posterior to the hip joints and anterior to the knee joint. Hip flexion contractures subsequently develop secondary to this muscle imbalance and associated anterior pelvic tilt. The hip abductors (primarily gluteus medius) tend to follow the hip extensors; their progressive weakness being responsible for the waddling gait and excessive pelvic motion seen in patients with DMD. In analogy to other neuromuscular disorders, weakness and ambulatory impairment may be responsible for the proximal femoral dysplasia and hip subluxation sometimes seen.

Natural History DMD is a life-limiting condition that affects boys with signs and symptoms identifiable in early childhood. Usually patients live into only their third to fourth decade, despite the advances in medical therapy [1, 2, 4, 7] (Fig. 21.1). Affected children may have delayed motor skills, such as sitting, standing and walking and usually present to their doctors from the age of 2–3 years onwards. The mean age at diagnosis is 4 years [2]. In DMD, the muscle weakness worsens rapidly and children are usually wheelchair-­ dependent by adolescence [1, 2, 4]. Once wheelchair-bound, patients with MD tend to develop worsening contractures and a rap10

15

20

25

Walking problems Wheel chair - skeletal deformity Very limited use of arms Ventilation at night Ventilation 24hrs Death DMD patient age

30

21  The Hip in Muscular Dystrophy

idly progressive scoliosis. The increasing scoliosis often occurs coincidently with an incremental reduction in forced vital capacity (FVC) that exacerbates their poor cardiopulmonary function [5]. Bone mineral density is reduced with an associated increased fracture risk, and this is exacerbated by the steroids used as part of the medical management. Twenty to sixty percent of cases have lowenergy extremity fractures, principally involving tibia, fibula and distal femur and up to 30% have symptomatic vertebral fractures [5]. These can lead to premature loss of walking ability. The clinical manifestations in Becker MD start later, progress more slowly and with less severity, with a correspondingly increased survival into the 5th–6th decades [4]. In both DMD and Becker MD, the cardiomyopathy begins in adolescence. Later, there is enlargement of the heart muscle with onset of the signs and symptoms of a dilated cardiomyopathy (arrhythmias, shortness of breath and fatigue). Once established, the cardiac problems worsen rapidly and become life-threatening in most cases [4, 5]. Whilst hip contractures in DMD are common, subluxation and dislocation are seen less ­frequently, although a 2001 report did identify 19 of 54 DMD patients (35%) with hip problems (15 unilateral subluxation, one bilateral subluxation and three with unilateral dislocation) [9]. Only one of the 19 hips was painful and none of the hips underwent surgery. Migration percentage increased in response to increasing pelvic obliquity and hence subluxation was manifested late in the clinical course. Many other reports do not make any specific reference to an incidence of hip subluxation or dislocation [5, 7, 10]. “Subluxation and dislocation occur in the later stages of DMD and in patients who have developed pelvic obliquity [9]”. Compared to other neuromuscular conditions, the contractures in DMD take the hip into flexion and abduction (with external rotation), a position which may be protective against increasing subluxation [9]. There is a theoretical risk of anterior hip dislocation in a patient who spends more time in a semi supine position with hips flexed, abducted

573

and externally rotated (a combination of contractures, weakness and gravity). This position may result in stretching of the anterior hip capsule and ligaments [9] but this mechanism is not mentioned in other reports [5, 7, 10]. As deformity, contractures and cardiorespiratory function worsen once the patient becomes wheelchair-based, the goal of orthopedic management, be it conservative or surgical, is to prolong patients’ ambulatory status as much as possible. DMD is primarily a muscle disorder and therefore sensation is preserved which minimizes some of the complications of a wheelchair-­ based life such as pressure sores [7].

Epidemiology Overall, the incidence of the muscular dystrophies varies with the specific type and there are also some national variations with Ryder et  al. [2] reporting an estimated point prevalence for DMD (the most common dystrophy) ranging from 1.9 (USA), 2.2 (UK), 6.1 (Canada) to 10.9 (France) per 100,000 males. Other reports quote similar numbers [1, 4, 7]. The most common is DMD with 1 case per 3500 live male births (Table  21.1). One third of cases occur as a result of spontaneous new mutations. Becker MD is the second most common form, with an incidence of 1 case per 30,000 live male births. The other types of MD are all rare [4, 7].

Clinical Presentation Boys with DMD may be a little late starting to walk and classically the waddling (Trendelenburg), wide-based toddler gait persists with hyperlorTable 21.1  Types of muscular dystrophy (MD) Inheritance patterns Sex-linked Autosomal dominant Autosomal recessive

Examples Duchenne, Becker and Emery-Dreifuss Facioscapulohumeral, distal, ocular, oculopharyngeal Limb girdle type

D. M. Eastwood

574

dosis of the lumbar spine and a tendency to toe walk [11]. Much of this is due to muscle weakness (most noticeably in gluteus maximus and medius) and hence an inability to support a single leg stance. The first signs are manifest by the age of 3–5  years, but there is frequently a delay in diagnosis [2, 7]. During the early stages of the disease, subtle changes to muscle length and joint range are apparent but may not be identified if the clinical examination is not performed in a precise manner. In such instances, an opportunity to make an early diagnosis can be missed. A positive Gowers test was initially considered to be pathognomonic of DMD but it is more accurately simply a measure of proximal muscle weakness [12]. Subtle abnormalities can be missed if the test is not done carefully [12, 13]. Over time, relative weakness leads to contractures in the antagonist muscles. Hip flexion and abduction contractures not only increase the lumbar lordosis but limit the ability to stand or lie supine comfortably. Patients may present with low back pain. The Ober test can be used to identify abduction contractures associated with iliotibial band tightness. For this test, the patient lies in the decubitus position with the hip placed in abduction and extension. The extent of abduction contracture is measured by the loss of adduction to neutral. The Thomas test should be used to measure the extent of hip flexion contracture, typically due to tightness of the iliopsoas, sartorius, rectus femoris, and tensor fascia latae. The hip adductors are not typically tight in DMD. The natural history is that by the age of 10, most boys with DMD are unable to walk and once they become wheelchair-based, their hip and knee contractures worsen rapidly and significant foot deformities develop. The deformity and flexibility at each joint level should be documented regularly, in addition to their maximal functional level [1, 5, 7]. Better documentation of the inter-related problems should allow earlier and more appropriate treatment in selected cases. [1, 2, 5]. Classically, the scoliosis deteriorates once the boys become wheelchair dependent. With modern medical management, there has been an

improvement in the natural history but the evidence is not wholly convincing: certainly, life expectancy has improved and walking ability has been prolonged into early adolescence—but issues relating to quality of life remain [5, 7, 14, 15]. With Becker MD, the clinical presentation is milder and more heterogenous [4]. The use of functional tests such the North Star Ambulatory Assessment (NSAA) and the 6MWT and other TFT (Timed Function Tests) are encouraged and these act as a monitor of deteriorating function and/or the effect of treatment [3, 16].

Essential Clinical/Diagnostic Tests

• Complete neuromuscular examination –– Gowers test considered pathognomonic for DMD and/or proximal muscle weakness ∘∘ Test should be performed supine and lying down ∘∘ Test prone crawling position –– Specific examination of the hip range of movements, muscle strength and degree of contracture –– Ober test to identify abduction contracture –– Thomas test to identify hip flexion contracture • Assessment of associated cardiorespiratory problems • Consultation with a geneticist –– Genetic analysis • Serum creatine kinase • Muscle biopsy and staining for dystrophin antibody –– Less necessary with advent of easier access to genetic testing

Imaging Ultrasound (US) and magnetic resonance imaging (MRI) are valuable in quantifying the extent and the severity of muscle involvement with the

21  The Hip in Muscular Dystrophy

MDs particularly for DMD and Becker MD [17]. Even in the early stages of MD, ultrasonography shows increased echogenicity in the affected muscles, with a corresponding reduction in the underlying bone echo. On MRI, the appearance of skeletal muscle correlates strongly with motor function. In Becker MD, the T1w MRI appearances act as a biomarker not only for disease severity but also for disease progression in terms of functional deterioration over a 12  month period. It may therefore help stratify patients and aid evaluation of treatment effects [17]. Changes may also be specific to certain genotypes. “Severity of muscle fatty substitution was significantly correlated with specific DMD mutations [17]”. Regular assessment of the spinal curve is essential once walking has stopped but there are no accepted guidelines relating to the necessity and/or frequency of pelvic radiographs to identify hip displacement. When reviewing the radiographs, it is important to look at the triad of thoracolumbar scoliosis, pelvic obliquity and femoral head subluxation/dislocation. The close inter-relationship of these three factors is often under-estimated [18]. In Patel and Shapiro’s study of 211 neuromuscular patients, in the best documented patients, hip deformity developed more rapidly than the spinal deformity in only 16% (10/63) whereas the reverse was true in 44% (28/63) and in the remainder, the scoliosis and the hip displacement happened concurrently [18]. Fifty two percent of this cohort had DMD and in these patients, spinal deformity progressed more rapidly than hip subluxation [18]. Standard radiographic measurements used in other neuromuscular disorders (e.g. cerebral palsy) can be utilized when assessing hip X-rays in DMD or BMD.  These include Reimers migration percentage (MP) to quantity the extent of hip subluxation, the acetabular index (AI) to evaluate for acetabular dysplasia and neck shaft angle (NSA) to quantify the morphology of the proximal femur (typically coxa valga).

575

Essential Imaging

• US • MRI • Plain Radiographs –– AP Pelvis • MP (migration percentage) • NSA (neck shaft angle) • AI (acetabular index) –– AP/Lateral Spine • DEXA scan for bone mineral density

Non-operative Management The ‘Gold Standard’ treatment for DMD now includes the routine use of corticosteroid therapy before boys start to lose significant muscle function, and certainly before they have stopped walking [3]. Steroid (prednisolone or deflazacort) is taken on a daily basis and the evidence suggests that it slows the progression of muscle weakness and allows the child to retain functional mobility for additional time (up to 3  years) [1–3, 5, 14, 16]. Studies looking at continuous versus intermittent steroid therapy are underway to define the optimum regimen [19]. Overall, the alterations in disease progression secondary to steroid therapy are poorly described making it difficult to judge the added benefit of some of the ‘new’ treatment modalities [3]. Steroid therapy is associated with retention of muscle strength and more importantly motor skills (such as stair climbing and timed motor tests (TMT)) although it requires specific outcome measures to quantify this accurately [16]. Overall, despite the use of steroids, significant muscle weakness in still seen in young children with DMD.  Unfortunately, there are no studies available regarding the impact of steroid treatment on the risk of hip subluxation or dislocation in DMD. Whilst steroid therapy has proven benefit in terms of maintaining walking ability, it is associated with the development of osteoporosis and an increased risk of fragility fractures particularly in the vertebral bodies [1, 5].

D. M. Eastwood

576

Maintaining good bone health is therefore an important aspect of patient care and treatments to increase bone mineral density (BMD) include vitamin D and calcium supplements as well as intravenous/oral bisphosphonates [1, 4, 5]. Whole body vibration has also been trialled [20]. One study [21] reported that Risedronate was well tolerated by most patients, appeared to maintain lumbar spine adjusted BMD Z-scores and may reduce fracture rates. Bisphosphonate treatment is usually reserved for those with fractures and/or bone pain with a low BMD on DXA [22]. Gordon et  al. [23] reported an increased survival rate in patients treated with bisphosphonates in addition to their steroid therapy, whilst Sbrocchi et al. [24] suggested that intravenous bisphosphonate use was associated with improvements in back pain and stabilization/ improvement in vertebral height ratios in vertebral bodies that had previously been fractured. They were unclear as to whether or not the incidence of new fractures had been influenced significantly by the addition of bisphosphonate therapy. Houston et al. [25] stated that in DMD boys with low bone mineral density, the Z score at the hip trended down without alendronate treatment and trended upward/stabilized with treatment. However, a recent Cochrane review [20] did not identify any high-quality evidence to guide the use of treatment to prevent or to treat such steroid-related complications. As with all neuromuscular conditions, in the MDs, hip subluxation is a potential problem. The possible relationship between subluxation and pelvic tilt calls for better control of sitting posture to prevent or at least minimize the pelvic tilt [15, 18, 26]. It might suggest that spinal stabilization should be carried out at an early age when any scoliosis and pelvic tilt are still mild, so that progressive subluxation of the hip may be delayed or prevented, in addition to maintaining sitting balance and comfort. With the current medical management, scoliosis progression is not as definite as previous and thus the indications for—and timing of—spinal surgery are still a little unclear. The majority of papers on spinal management in DMD make no mention of hip problems [5, 15, 27].

Physiotherapy Historically, exercise with daily standing periods of 2 h or more has been shown to slow the progression of the spinal curve and the associated impaired respiration [27]. Physiotherapy is more effective in delaying the development of contractures than treating them, so stretching and exercise programmes should start early [1, 4]. For the hip, the focus should be on stretching the hip flexors and abductors. More severe abduction contractures can interfere with comfortable wheelchair seating and as such their prevention should be encouraged. Excessive hip flexion contractures can exacerbate anterior pelvic tilt and lumbar lordosis leading to pain. Attempting to prevent or delay progression of these contractures with physiotherapy may be of benefit. As the patient’s weakness progresses the use of orthotics and/or walking aids will be required to maintain independence and function, particularly for the activities of daily living. In ambulant patients, orthotics such as drop-lock KAFOs to support weak quads against an evolving knee fixed flexion deformity need to be light weight to facilitate continued walking and to prevent immobility. It is important to teach the patient and family techniques for energy conservation, minimization of overuse fatigue and joint protection. Pre-operative respiratory exercises have been shown to increase forced vital capacity (FVC) and limit post-operative respiratory complications [15].

Essential Non-operative Treatments

• Steroid medication • Maintain mobility • Physiotherapy programmes –– Stretching, splinting, positioning • Wheelchair seat wedge to induce lumbar lordosis • Maintenance of bone health • General advice to reduce fracture risk • Regular cardiorespiratory review –– Angiotensin converting enzyme (ACE inhibitors) –– Angiotensin II receptor blockers (ARBs) –– BiPAP

21  The Hip in Muscular Dystrophy

tures may still be indicated to facilitate use of sitto-stand chairs and aid in standing transfers.

Non-operative Pitfalls

• Osteoporosis • Fragility fractures • Weight gain

“Will surgical treatment really improve the patient’s QoL?”.

Operative Management One approach to operative management in this group of patients relates the timing of surgical intervention to the functional status of the patient (Table 21.2). The concept of ‘prophylactic’ or early surgery was popularized by Yves Rideau in the 1980s [29–31]. Early surgery should take place before boys lose their ability to walk as rehabilitation is significantly quicker at this stage and considered to be more reliable. [30, 31] The aim is to conserve the child’s ability to stand/walk for as long as possible. For those who are already wheelchair-­ bound, surgery to release contrac-

Table 21.2  Timing of and indications for surgery in DMD Timing For patients WHO are able to walk or Within 1 month of losing this ability As patient becomes wheelchair based Rehabilitative Wheelchair user Therapeutic use of a standing frame Palliative Wheelchair user Ambulatory

577

Comments Early aggressive soft tissue releases can prolong walking for up to 3.5 years [28–30] Such surgery is now infrequently performed at hip level (see text) Early spinal stabilization for scoliosis that develops as walking ceases Surgery is used only to correct deformities that may limit physical therapy and orthosis wear Surgery is used only to treat problems of immediate concern for the patient’s comfort, such as difficulty with shoe wearing, ulcerations, and seating position/comfort in wheelchairs

Before embarking on surgery in children with DMD it is important to address the question as to whether or not the planned surgical intervention will improve quality of life and advance the individual’s psycho-social situation. The family and patient may wish to be ‘spared’ such interventions from both a physical and psychological point of view. “When knowledge of the clinical course became clear, therapeutic decisions would come more easily” attributed to Prof A Cournand”. Rideau himself acknowledged that it is only an understanding of the clinical course in an individual patient that allows you to make a decision regarding therapeutic interventions. In DMD, the potential relationship between hip subluxation and pelvic tilt necessitates good control of sitting posture to prevent or at least minimize the pelvic tilt. It might suggest that spinal stabilization should be carried out at an early age when any scoliosis and pelvic tilt are still mild, so that progressive subluxation of the hip may be delayed or prevented, in addition to maintaining sitting balance and comfort. This is, however, unproven. Early appropriate treatment of the spine is considered the single most important surgical treatment: the surgical ‘burden’ is less if surgery is performed earlier on a generally fitter child. The overall age at which spinal surgery takes place has increased slightly over time which may be attributable to the prolonged walking associated with recent steroid management. In a recent 30  year review, it was noted that 92% of cases were fused to the pelvis to control pelvic tilt and pelvic obliquity [27]. Archer et  al. commented that both the upper and lower limits of the fusion remain controversial [15]. A Cochrane database review in 2015 [33], looked for evidence to support the use of spinal surgery in DMD but found no studies that met their inclusion criteria. They therefore felt unable to provide an evidence-

D. M. Eastwood

578

based conclusion to guide clinical practice and recommended that patients should be informed of the uncertain nature of the benefits versus risks of such surgery [33]. Others feel that despite the lack of RCTs, spinal fusion can be recommended as there is sufficient evidence from prospective cohort studies that s­urgery improves function, sitting balance and tolerance, pain and quality of life [5, 15, 34].

Soft Tissue Releases If a recognized goal of treatment is to prolong standing and walking, this goal can be achieved by releasing muscle contractures. Various tenotomies have been described and discussed in the literature: most papers discuss release of the ankle, knee and hip joints [7, 28, 29]. Rideau and the Forsts have promoted early hip abductor release as part of the strategy for improving contractures at all levels of the lower limbs and prolonging ambulation (Table 21.3). In thin patients, percutaneous tenotomies have been promoted for knee and ankle contractures [31]. There is concern that in patients treated with steroids that percutaneous surgery might be associated with wound healing problems but this has not been substantiated [30]. Forst and Forst [30] stated that joint restriction was often apparent as early as age 4–6 years, and if the other indications for surgery were present (quadriceps power MRC grade 3 or more and an ability to stand from a supine position in children less than 5  years of age) surgery should be considered. The hip releases popularized by Rideau [29] and described by Forst and Forst [30] include release of tensor fascia latae, sartorius and rectus femoris. Knee and ankle contractures were addressed simultaneously and no plaster casts are used. An epidural catheter provides pain relief and anaesthetic agents that risk hyperthermia and/or Table 21.3  Muscle resections that have been described for use in Duchenne muscular dystrophy [7, 29–31] Name Souter-Strathclyde Yount

Description Proximal ITB resection Distal ITB resection Complete ITB resection

hepatic dysfunction are avoided. Patients regain their ability to stand 2 days postoperatively and are usually walking by 4–6 days [7, 30]. Surgery should be performed early for maximal effect and when hip and knee contractures have progressed past 30°, soft tissue releases are rarely worthwhile [7, 30, 35]. The need for this type of surgery has reduced since the routine use of deflazacort and the authors [30] note no increase in perioperative complications when they have operated on patients who have received steroid therapy. In 1993, Smith et al. noted that only 49% of releases were ‘successful’ at hip level in terms of regaining full extension and adduction [31]. Overall, however, treated patients did continue to walk for longer, perhaps because even a limited correction at hip joint level was beneficial when combined with a full correction at knee and ankle levels. The recent report by the DMD Care Considerations working party states that hip and knee releases are no longer recommended for patients who are still in the ambulatory phase [5, 10].

Other Surgical Procedures Whilst there are reports in the literature of hip subluxation and dislocation, these are invariably in patients with DMD (Fig.  21.2) rather than the other MDs and there are no specific guidelines for their management. The surgical principles that are often applied to hip migration in neuromuscular patients must be adapted to the particular circumstances identified in boys with DMD.  Steroid use and inactivity are associated with obesity and osteoporosis in this patient group and the additional co-morbidities represented by their cardiomyopathy mean that this patient group is at high risk. The lifetime fracture risk in children with DMD is twice normal [15]. Whilst children with DMD are still walking, both lower and upper limb fractures should be treated aggressively to maintain weight bearing and normal activity and reduce the effects of immobility osteoporosis

579

21  The Hip in Muscular Dystrophy

Operative Pitfalls

• Malignant hyperthermia –– More common in patients with muscle disease –– Risk diminished with nitrous oxide, intravenous narcotics, sedatives, and non-depolarizing muscle relaxants • Intra-operative steroid induced osteoporotic fractures • Death due to cardiac disease

Fig. 21.2  Anteroposterior radiograph of a 20  year-old man with DMD. He has had his spine fused to the pelvis, is on assisted ventilation and shows signs of osteoporosis. His subluxated left hip is not painful and although he does have marked muscle contractures these are not interfering with his comfort or personal care

being additive to those induced by steroid therapy [5]. Femoral neck fractures do occur and pose a difficult problem particularly if the child is nonambulatory with osteoporotic bone. Conservative treatment with pain relief and appropriate physiotherapy to maintain functional status can be successful with a low associated risk of complications [1, 5, 7].

Essential Surgical Procedures • Spine stabilisation Possible Surgical Procedures • Soft Tissue Releases –– Sartorius, Tensor fascia latae, Rectus femoris

Surgical interventions are accompanied by an increased risk of both morbidity and mortality [1–5, 7, 27]. In part this can be related to the associated cardiomyopathy but compared to patients with other neuromuscular disorders, such as cerebral palsy, patients with DMD undergoing scoliosis surgery had a significantly higher complication rate (38.5% vs. 16.7%) with higher rates of deep wound infection and hepatotoxicity [36]. The latter was unique to DMD patients and associated with increased intra-operative blood loss [35]. High complication rates were also noted by Scannell et al. and these had remained constant over a 30 year period [27]. It is also important to recognize that inhalational anaesthetic agents and succinyl choline can result in potentially fatal hyperkalaemia and rhabdomyolysis highlighting the need for a multidisciplinary assessment pre-operatively and co-­ordinated care in the perioperative period [1–5, 7, 15].

Classic Papers Sussman MD. 2002. Duchenne Muscular Dystrophy. J Am Acad Orthop Surgeons 10:138–51. This review article predates some of the most recent developments at cellular and at surgical level in the field of MD but it provides a clinical and historical perspective of how care has changed this century. Rideau YM. 2012. Requiem: Memories by a Myologist Acta Myologica 31:48–60. This article describes one man’s dedication to improving the

D. M. Eastwood

580

quality of life in patients with DMD: a holistic approach to care is highlighted. In never considering neuromuscular disease to be untreatable, Yves Rideau has found ways to ameliorate every aspect of these conditions. His work has resulted in immeasurably enhancing the quality of life of his patients.

Key Evidence Smith SE, Green NE, Cole RJ, Robison JD, Fenichel GM 1993 Prolongation of ambulation in children with Duchenne muscular dystrophy by subcutaneous lower limb tenotomy. J Pediatr Orthop 13:336–40. A cohort study that compared boys who accepted subcutaneous lower limb tenotomy with controls who also showed indications for surgery but who refused intervention. The treated group showed complete contracture correction at the ankle, 58% correction at the knee and 49% at the hip. They also walked for longer (in braces) and stood for longer compared to the control group. Birnkrant DJ, Bushby K, Bann CM et  al for the DMD Care Considerations Working Group. 2018. Lancet series. These three papers report the current practice guidelines for the management of all aspects of care of patients with Duchenne muscular dystrophy.

Take Home Messages

• MDs are a group of rare inherited conditions characterised by progressive muscle weakness and a shortened life span. • In DMD, the use of steroid therapy prolongs walking and life although the associated problems of obesity and osteoporosis are significant. • Early spinal fusion once walking ability is lost, reduces the deterioration in pulmonary function. • The role of soft tissue release for hip joint contracture is unclear. • Subluxation/dislocation of the hip joint is rare compared to other neuromuscular conditions.

References 1. Birnkrant DJ, Bushby K, Bann CM et al for the DMD Care Considerations Working Group. Diagnosis and management of Duchenne muscular dystrophy part 1: diagnosis and neuromuscular rehabilitation, endocrine and gastrointestinal and nutritional management. Lancet. 2018;17:251–67. 2. Ryder S, Leadley RM, Armstrong N, Westwood M, de Kock S, Butt T, Jain M, Kleijnen J. The burden, epidemiology, costs and treatment for Duchenne muscular dystrophy: an evidence review. Orphanet J Rare Dis. 2017;112:79–100. 3. Matthews E, Brassington R, Kuntzer T, Jichi F, Mansur AY.  Corticosteroids for the treatment of Duchenne muscular dystrophy. Cochrane Database Syst Rev. 2016;5(5):CD003725. 4. Dowling JJ, Gonorazky HD, Cohn RD, Campbell C.  Treating pediatric neuromuscular disorders: the future is now. Am J Med Genet. 2018;176:804–41. 5. Birnkrant DJ, Bushby K, Bann CM et al for the DMD Care Considerations Working Group. Diagnosis and management of Duchenne muscular dystrophy part 2: respiratory, cardiac, bone health and orthopaedic management. Lancet. 2018;17:347–61. 6. Judge LM, Haraguchi M, Chamberlain JS. Dissecting the signalling and mechanical functions of the dystrophin-glycoprotein complex. J Cell Sci. 2006;119:1537–46. 7. Sussman MD.  Duchenne muscular dystrophy. J Am Acad Orthop Surg. 2002;10:138–51. 8. Rodino-Klapac LR, Mendell JR, Sahenk Z. Update on the treatment of Duchenne muscular dystrophy. Curr Neurol Neurosci Rep. 2013;13:332. 9. Chan KG, Galasko CS, Delaney C.  Hip subluxation and dislocation in Duchenne muscular dystrophy. J Pediatr Orthop B. 2001;10:219–25. 10. Birnkrant DJ, Bushby K, Bann CM et al for the DMD Care Considerations Working Group. Diagnosis and management of Duchenne muscular dystrophy part 3: primary care, emergency management, psychosocial care and transitions of care across the lifespan. Lancet. 2018;17:1–11. 11. Heberer K, Fowler E, Staudt L, Sienko S, Buckon CE, Bagley A, Sison-Williamson M, McDonald CM, Sussman M.  Hip kinetics during gait are clinically meaningful outcomes in young boys with Duchenne muscular dystrophy. Gait Posture. 2016;48: 159–64. 12. Wallace GB, Newton RW.  Gower’s sign revisited. Arch Dis Child. 1989;64:1317–9. 13. Chang RF, Mubarak SJ. Pathomechanics of Gowers’ sign. A video analysis of a spectrum of Gowers’ maneuvers. Clin Orthop Relat Res. 2012;470:1987–91. 14. Lebel DE, Corston JA, McAdam LC, Biggar WD, Alman BA. Glucocorticoid treatment for the prevention of scoliosis in children with Duchenne muscular dystrophy: long term follow-up. J Bone Joint Surg Am. 2013;95:1057–61.

21  The Hip in Muscular Dystrophy 15. Archer JE, Gardner AC, Roper HP, Chikermane AA, Tatman AJ. Duchenne muscular dystrophy: the management of scoliosis. J Spine Surg. 2016;2:185–94. 16. Buckon C, Sienko S, Bagley A, Sison-Williamson M, Fowler E, Staudt L, Heberer K, McDonald CM, Sussman M.  Can quantitative muscle strength and functional motor ability differentiate the influence of age and corticosteroids in ambulatory boys with Duchenne muscular dystrophy? PLoS Curr. 2016;8:8. 17. Barp A, Bello L, Caumo L, Campadello P, Semplicini C, Lazzarotto A, Sorarù G, Calore C, Rampado A, Motta R, Stramare R, Pegoraro E.  Muscle MRI and functional outcome measures in Becker muscular dystrophy. Sci Rep. 2017;7:16060. 18. Patel J, Shapiro F. Simultaneous progression patterns of scoliosis, pelvic obliquity, and hip subluxation/dislocation in non-ambulatory neuromuscular patients: an approach to deformity documentation. J Child Orthop. 2015;9:345–56. 19. Griggs R.. Finding the optimum regiment for Duchenne muscular dystrophy (FOR DMD). https:// clinicaltrials.gove/ct2/show/NCT01603407. Accessed 20 Apr 2018. 20. Bell JM, Shields MD, Watters J, Hamilton A, Beringer T, Elliott M, Quinlivan R, Tirupathi S, Blackwood B.  Interventions to prevent and treat corticosteroid-­ induced osteoporosis and prevent osteoporotic fractures in Duchenne muscular dystrophy. Cochrane Database Syst Rev. 2017;24(1):CD010899. 21. Srinivasan R, Rawlings D, Wood CL, Cheetham T, Moreno AC, Mayhew A, Eagle M, Guglieri M, Starub V, Owen C, Bushby K, Sarkozy A. Prophylactic oral bisphosphonate therapy in Duchenne muscular dystrophy. Muscle Nerve. 2016;54:79–85. 22. Buckner JL, Bowden SA, Mahan JD.  Optimising bone health in Duchenne muscular dystrophy. Int J Endocinol. 2015;2015:928385. 23. Gordon KE, Dooley JM, Sheppard KM, MacSween J, Esser MJ.  Impact of bisphosphonates on survival for patients with Duchenne muscular dystrophy. Pediatrics. 2011;127:e353–8. 24. Sbrocchi AM, Rauch F, Jacob P, McCormick A, McMillan HJ, Matzinger MA, Ward LM. The use of intravenous bisphosphonate therapy to treat vertebral fractures due to osteoporosis among boys with Duchenne muscular dystrophy. Osteoporosis Int. 2012;23:2803–11.

581 25. Houston C, Matthews K, Shibli-Rahhal A.  Bone density and alendronate effects in Duchenne muscular dystrophy patients. Muscle Nerve. 2014;49: 506–11. 26. Frischhut B, Krismer M, Stoeckl B, Landauer F, Auckenthaler T.  Pelvic tilt in neuromuscular disorders. J Pediatr Orthop B. 2000;9:221–8. 27. Scannell BP, Yaszay B, Bartley CE, Newton PO, Mubarak SJ.  Surgical correction of scoliosis in patients with Duchenne muscular dystrophy: 30-year experience. J Pediatr Orthop. 2017;37:464–9. 28. Galasko CS, Williamson JB, Delaney CM. Lung function in Duchenne muscular dystrophy. Eur Spine J. 1995;5:263–7. 29. Rideau Y, Duport G, Delaubier A, Guillou C, Renardel-­ Irani A, Bach JR. Early treatment to preserve quality of locomotion for children with Duchenne muscular dystrophy. Semin Neurol. 1995;15:9–1. 30. Forst J, Forst R. Surgical treatment of Duchenne muscular dystrophy patients in Germany: the present situation. Acta Myol. 2012;31:21–3. 31. Smith SE, Green NE, Cole RJ, Robison JD, Fenichel GM.  Prolongation of ambulation in children with Duchenne muscular dystrophy by subcutaneous lower limb tenotomy. J Pediatr Orthop. 1993;13:336–40. 32. Rideau YM. Requiem: memories by a myologist. Acta Myol. 2012;31:48–60. 33. Cheuk DK, Wong V, Wriage E, Baxter P, Cole A.  Surgery for scoliosis in Duchenne muscular dystrophy. Cochrane Database Syst Rev. 2015;1(10):CD005375. 34. Suk KS, Lee BH, Moon SH, Choi YC, Shin DE, Ha JW, Song KM, Kim HS.  Functional outcomes in Duchenne muscular dystrophy scoliosis: comparison of the differences between surgical and nonsurgical management. J Bone Joint Surg Am. 2014;96: 409–15. 35. Canavese F, Sussman MD. Strategies of hip management in neuromuscular disorders: Duchenne muscular dystrophy, spinal muscular atrophy, charcot-marie-­ tooth disease and arthrogryposis multiplex congenita. Hip Int. 2009;19(Suppl 6):S46–52. 36. Duckworth AD, Mitchell MJ, Tsirikos AI. Incidence and risk factors for postoperative complications after scoliosis surgery in patients with Duchenne muscular dystrophy: a comparison with other neuromuscular conditions. Bone Joint J. 2014;96:943–9.

The Hip in Charcot-Marie-Tooth Disease

22

Neil Saran

Introduction Charcot-Marie-Tooth Disease (CMT) is the most common form of hereditary neuropathy and is characterized by a slowly progressive motor and sensory polyneuropathy. There are many genetic loci associated with CMT including autosomal dominant, autosomal recessive and X-linked inheritance patterns, as well as sporadic forms. CMT is typically divided into various phenotypes each of which is associated with multiple genes. The two most common phenotypes are CMT 1 and CMT 2. CMT 1 is the most common and is caused by peripheral nerve demyelination resulting in myelinopathy. CMT 2 is caused by Wallerian axonal degeneration resulting in axonopathy. Intermediate forms that affect both myelin and axons also exist. CMT manifests at various ages, from infantile to late adulthood. In keeping, there is considerable variability in the severity of the disease. Initial presentation is usually marked by lower leg weakness and foot deformities. Classically, patients present with a cavus foot deformity that

N. Saran (*) Department of Paediatric Surgery, McGill University, Montreal, QC, Canada Shriners Hospital for Children, Montreal, QC, Canada Montreal Children’s Hospital, Montreal, QC, Canada e-mail: [email protected] © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_22

develops secondary to weakness of tibialis anterior and peroneal muscles and an imbalance between the intrinsic and extrinsic muscles. Other orthopaedic manifestations include scoliosis, hip dysplasia, and upper extremity weakness (typically presenting much later than lower extremity weakness). In 1985, Kumar et al. highlighted the association between CMT and asymptomatic or mildly symptomatic severe hip dysplasia [7]. They also suggested that the onset of the dysplasia was more insidious than with congenital hip dysplasia; furthermore, they recommended screening for hip dysplasia in CMT patients. Subsequently it has been estimated that the incidence of hip dysplasia in patients with CMT is 6–9% [11, 19].

Pathophysiology Hips in patients with CMT are usually normal at birth and become dysplastic only once muscular imbalance is present, suggesting that the origin of the dysplasia is neuromuscular [3, 18]. The progressive axonal and/or demyelinating disease of peripheral nerves seen in CMT eventually results in muscle weakness which typically manifests in late childhood or during adolescence [1, 3]. In addition to the classic presentation of distal lower leg weakness due to peripheral neuropathy, proximal muscle weakness is also seen although it is typically less profound than distally [2]. Some 583

N. Saran

584

patients exhibit signs of proximal muscle weakness with a wide based or Trendelenburg gait at presentation [6, 11]. A proposed hypothesis on the pathophysiology of hip dysplasia in CMT is that proximal muscle weakness, predominantly involving the abductors and hip extensors, induces secondary bony changes resulting in a shallow acetabulum and a valgus, anteverted femoral neck [3]. While this hypothesis would suggest a posterolateral subluxation pattern as seen in other neuromuscular conditions such as cerebral palsy, cases of profound hip flexion weakness [5, 12] and excessively anteverted acetabuli [10] are also seen which would suggest anterior subluxation patterns. While the pathophysiology is not clearly understood, there is a clear neuromuscular component involved that predisposes patients with CMT to develop hip dysplasia.

Natural History Hip dysplasia in CMT is typically diagnosed in adolescence or early adulthood, although childhood onset is sometimes seen. Age at onset decreases with increasing severity of neurologic involvement. Hip dysplasia is either identified on screening radiographs in patients known to have CMT, or after a radiograph is taken for signs or symptoms suggestive of hip pathology. Hip dysplasia is often silent in CMT patients often present late with severe dysplasia. “Hip dysplasia is often silent in CMT and patients often present late with severe dysplasia”. While true natural history studies are lacking, the dysplasia appears to be progressive. It is unknown if the progression stops after skeletal maturity. Some hips go on to develop clinical instability and dislocations. Osteoarthritis of the hip is not an uncommon finding [10, 15]. Novais et  al. found that out of 19 patients with CMT hip dysplasia, three had moderate (Tönnis grade 2) and one had severe (Tönnis grade 3) osteoarthritis at an average age of 23 ± 8 years (average ± standard deviation) suggesting that the long term natural history of hip dysplasia in CMT may

include severe osteoarthritis [10]. That being said, whether or not they develop painful osteoarthritis is unknown. Outcomes data of surgical intervention in these patients are also limited. In 27 young adult patients with unilateral CMT hip dysplasia, Novais and colleagues [9] showed improved modified Harris Hip Scores at an average of 5.2 years after a reconstruction with a periacetabular osteotomy. Two patients in this group were converted to a total hip arthroplasty during the follow-up period. In 14 adolescent patients (19 hips), Stover and colleagues noted that despite an overall improvement in radiologic parameters, 37% of hips had progression of osteoarthritis at an average follow-up of 3.4 years after a reconstructive periacetabular osteotomy. The findings of these studies suggest that despite improvement in radiologic hip morphometric parameters and patient-reported outcome measures, reconstructed CMT hips continue to undergo radiologic degenerative changes in short term follow-up. One possible explanation for this is that the diagnosis is often made late because these patients typically remain minimally symptomatic despite severe hip dysplasia. It is well known that pre-existing osteoarthritis is a risk factor for failure of treatment of acetabular dysplasia in developmental dysplasia of the hip (DDH) [8]. In fact, a Bernese periacetabular osteotomy is not recommended for treatment of DDH in adolescent or young adults once Tönnis grade 2 changes are seen [8]. Earlier diagnosis and treatment prior to the development of severe dysplasia may improve outcomes in these patients. As such, a surveillance program is recommended. “Earlier diagnosis and treatment prior to the development of severe dysplasia may improve outcomes in these patients. As such, a surveillance program is recommended”.

Epidemiology Charcot-Marie-Tooth disease affects approximately 1 in 2500 individuals with an equal distribution of males and females [14]. Hip dysplasia

22  The Hip in Charcot-Marie-Tooth Disease

is seen in 6–9% of patients with CMT although this may under-represent the true prevalence. In many instances, the diagnosis of hip dysplasia and CMT is not made until adulthood—as evidenced by several case series reporting the association of late onset, severe dysplasia with this diagnosis [4, 5, 17]. Walker and colleagues found evidence of hip dysplasia in 18% of females compared to only 2% of males, suggesting a higher prevalence in females with CMT [19]. Interestingly, when evaluating CMT patients who underwent periacetabular osteotomies, Novais and colleagues [9] found that 52% were female; this suggested a similar prevalence of hip dysplasia by gender for those patients requiring surgical reconstruction. Involvement can be unilateral or bilateral and age at onset varies with severity of the disease. “Hip dysplasia is seen in 6–9% of patients with CMT”

Clinical Presentation Patients with CMT may complain of clumsiness, difficulty running, frequent tripping, or ankle sprains and are more likely to present with lower extremity weakness, cavus foot or claw toes than with hip dysplasia. However, in some cases, patients present with groin, abductor, or peritrochanteric pain or symptoms related to muscle weakness. Pain related to hip subluxation is variable although it is typically a late finding in CMT hip dysplasia. “Pain related to hip subluxation is variable although it is typically a late finding in CMT hip dysplasia”. Signs and symptoms related to hip pathology in CMT include abductor fatigue, Trendelenburg gait, abductor lurch, or a wide-based gait. Examination of the hip may uncover hip instability when the hip is put into extreme positions of flexion, adduction and internal rotation which can result in posterior dislocation of the hip and a palpable clunk. Similarly, extension and external rotation of the hip may produce anterior disloca-

585

tion and a palpable clunk. The more typical findings on physical examination of these patients include weakness of peroneal, tibialis anterior (extensor substitution on active ankle dorsiflexion), hip abductor, hip flexor, and hand muscles, atrophy of the foot intrinsics, calf, and hand intrinsics, decreased light touch, pain, proprioception and vibratory sense, and the classic cavus or cavovarus foot and claw toe deformities.

Essential Clinical Tests

• Gait assessment looking for abductor weakness or ataxia (Trendelenburg, Romberg) • Assess strength of hip muscles (abductors, extensors, flexors) • Examine for anterior and posterior instability of the hip • Assess for atrophy of the calves, intrinsic muscles (hand) and for hyporeflexia to help suggest the diagnosis • Assess for foot deformities (most commonly: cavus foot; ankle equinus; extensor substitution on active ankle dorsiflexion) • Perform distal lower and upper extremity sensory and motor exam

Imaging The mainstay of imaging consists of a standing anteroposterior (AP) pelvis X-ray, an abduction internal rotation (AIR) view, and in patients with a closed tri-radiate cartilage, a false profile view. The AP pelvis X-ray must be done with the patellae pointing forward and enables quantification of various hip dysplasia parameters including: subluxation, lateral femoral head coverage and acetabular dysplasia (Fig.  22.1). In adolescent and young adult patients, the AP pelvis X-ray can also help assess acetabular version by looking for a crossover sign, posterior wall sign and ischial spine sign (see Chap. 9). The abduction internal rotation (Von Rosen) view is performed when subluxation is seen on the AP pelvis X-ray and is used to assess reduc-

586

Fig. 22.1  Anteroposterior pelvic radiograph in a 10 year-­ old male was performed for a complaint of bilateral groin pain worsening over the last year. He had a bilateral Trendelenburg gait and mild equinus contractures in both feet. His hips rotated internally 80° and externally 40°. Both hips are subluxated as evidenced by a break in Shenton’s line. The migration percentage is over 50% on the right and approximately 50% on the left. The center edge angles are negative. The medial space is widened. A work up was initiated and he was diagnosed with CMT Type 1A

Fig. 22.2  Abduction internal rotation pelvic radiograph showing both femoral heads centering very nicely with restoration of Shenton’s line and improved femoral head coverage. Slight widening of the medial joint space on the right side is persistent and could be evaluated with an arthrogram to determine the true congruency of the reduction although the femoral head and acetabular sourcil articulate with excellent congruency

ibility of the subluxation (Fig.  22.2). This is important to determine the feasibility of a redirectional osteotomy (e.g. PAO) (see section “Operative Management”). In addition, it can provide a better estimate of the true femoral neck

N. Saran

Fig. 22.3  False profile radiograph of the right hip reveals anterior subluxation and an anteverted acetabulum. The right hip is almost dislocated

shaft angle than the AP pelvis, when there is excessive femoral anteversion. A false profile view can provide valuable information regarding anterior coverage of the femoral head—as measured by the anterior center-­edge angle. In some instances, a false profile view may disclose previously unrecognized anterior subluxation (Fig. 22.3). Patients with CMT have been shown to have severely altered hip morphometry in comparison to those with developmental dysplasia of the hip, including more severe acetabular dysplasia and subluxation, decreased acetabular volume, decreased femoral head coverage, increased acetabular anteversion, increased femoral anteversion and coxa valga [10]. While many of these osseous abnormalities are best quantified using CT, plain radiographs are generally adequate for preoperative planning (Fig. 22.4). A specific instance where CT scans can be useful includes the revision setting where bony landmarks may be obscured by previous surgery. Magnetic resonance imaging (MRI) can be obtained to assess the labrum in young adult patients with dysplasia to determine if a simultaneous hip arthroscopy will be needed for labral repair prior to, or concomitant with, a reconstruction procedure. As

587

22  The Hip in Charcot-Marie-Tooth Disease Fig. 22.4 Trans equatorial CT cuts of both hips show anterior dislocation of the left hip and increased bilateral acetabular anteversion (30° on the left and 35° on the right)

radiographs may underestimate femoral head coverage due to unossified peripheral acetabular cartilage in younger patients, theoretically, MRI could be helpful for patients with an open triradiate cartilage to determine whether lateral or anterolateral coverage is required; however, to date, this has not been described for CMT hip dysplasia.

Essential Imaging Tests and Measurements

AP Pelvis • Shenton’s line • Migration percentage • Acetabular index • Center edge angle Abduction internal rotation (Von Rosen) view • Assesses reducibility for a subluxated hip False Profile view • Anterior subluxation • Anterior center edge angle

Non-operative Management There is no proven benefit of physical therapy or orthoses in patients with CMT hip dysplasia. Ramdharry and colleagues conducted a pilot study on a hip flexor muscle resistance training program to improve walking endurance related to hip flexor fatigue; they found that although there was a mea-

surable improvement in muscle strength, there was no improvement in walking parameters [13]. Even though, theoretically, a global hip strengthening program in CMT patients may delay or prevent onset of hip dysplasia, there is no clinical evidence to support this hypothesis. For the asymptomatic hip with mild dysplasia and no subluxation, there is no consensus as to whether prophylactic surgery should be performed. If non-operative management is chosen, annual radiographs should be obtained until skeletal maturity to look for progression of dysplasia, or onset of symptoms. “If non-operative management is chosen, annual radiographs should be obtained until skeletal maturity to look for progression of dysplasia, or onset of symptoms”.

Operative Management Infantile hip dysplasia in a patient with CMT is most likely DDH given that CMT hips are typically normal at birth (see section “Pathophysiology”). These are usually two separate entities that can coexist. In such a case, the infantile hip dislocation should be treated as in DDH (Fig. 22.5a, b). Due to the underlying neurological disease, these hips are more likely to re-subluxate during childhood or adolescence (Fig. 22.5c–g); therefore, annual radiographic follow-up until skeletal maturity is strongly r­ecommended. On a similar note, if

N. Saran

588

a

b

c

d

e

f

Fig. 22.5 (a) Ultrasound of left hip in a 6 week-old with bilateral Barlow positive hips showing decreased coverage and a dysplastic acetabulum. The patient was treated in a Pavlik Harness for 4 months at which point the ultrasound normalized. (b) AP pelvic radiograph at 6 months of age shows that the hips are reduced and the acetabular indices are approximately 30° bilaterally. (c) AP pelvic X-ray at 3 years of age shows bilateral hip subluxation which reduces on abduction internal rotation views (d). (e) After a work up that reveals the patient to have CMT type 1A due to a chromosome 17 duplication, she undergoes a varus derotation osteotomy of both femora and

bilateral San Diego pelvic osteotomies and is placed in an A-frame cast for 6 weeks. (f) AP pelvis radiograph at 6 months postoperatively shows that the hips appear well reduced although the acetabular index on the left remains elevated. (g) AP pelvic X-ray at 8 years of age and 5 years postoperatively show an increasing coxa valga of both hips, a break in Shenton’s line, center edge angle of 0°, and a migration index of 50% on the left side. At this point she was scheduled for bilateral varus derotation osteotomies and a San Diego osteotomy of the left hip. (Images used with permission from T. Benaroch, Montreal Children’s Hospital)

22  The Hip in Charcot-Marie-Tooth Disease

589

g

Fig. 22.5 (continued)

unexplained resubluxation occurs in previously treated hip dysplasia presumed to be DDH, the diagnosis of CMT should be considered. “…if unexplained resubluxation occurs in previously treated hip dysplasia presumed to be DDH, the diagnosis of CMT should be considered”. The majority of CMT hip dysplasia cases that require surgical treatment can be divided into two groups based on whether the triradiate cartilage is open or closed. The young patient with a dysplastic acetabulum (open triradiate cartilage) is treated with an acetabuloplasty such as the San Diego or Pemberton depending on whether global or anterolateral coverage is desired and will usually require a femoral varus derotation osteotomy in order to normalize femoral anteversion and coxa valga (Fig.  22.6). If it is felt that the acetabulum is not capacious and that a redirection rather than a reshaping is required, a triple pelvic osteotomy may be considered. The goals of surgery, in young patients, are to improve coverage of the femoral head and to normalize the acetabular index, femoral anteversion, and coxa valga in order to achieve stability and minimize recurrence. While there is no recommended value for neck shaft angle, we generally decide on the amount of varus and derotation required based on an intraoperative fluoroscopic abduction internal rotation view. The amount of abduction and inter-

Fig. 22.6  AP pelvis radiograph at 2 months postop after removal of spica cast for the patient featured in Fig. 22.1 and  22.2. Bilateral femoral derotation osteotomies and pelvic San Diego osteotomies have been performed. A capsulorrhaphy was not performed. The right femur was fixed with a sliding hip screw device while the left femur was fixed with an offset locking plate. Both femoral heads are well covered and the medial space has normalized. The acetabular dysplasia has been largely corrected

nal rotation required to yield a congruent and concentric joint decides the amount of varus and derotation that is performed. If there is no subluxation but only acetabular and femoral dysplasia, we tend to aim for a neck shaft angle of 130°, in order to minimize effects of over-varusization while attempting to decrease the risk of recurrence—which is high in those requiring surgery prior to triradiate cartilage closure (Fig.  22.7). Post operatively, the patient remains in a spica cast for 6  weeks at which point, physiotherapy and weight-bearing is initiated. “The goals of surgery, in young patients, are to improve coverage of the femoral head and to normalize the acetabular index, femoral anteversion, and coxa valga in order to achieve stability and minimize recurrence”. A capsulorrhaphy is performed when an open reduction is required. An open reduction is typically performed in younger patients when the hip does not fully reduce in the abduction internal rotation position under general anaesthetic. Also, an open reduction should be considered for older children or adolescent patients with a recent onset fixed dislocation in

N. Saran

590

a

b

c

d

e

Fig. 22.7 AP pelvis radiograph at 3 months for the patient featured in Fig.  22.1,   22.2 and   22.6 (a) shows maintenance of alignment. However, the medial clear space appears slightly wider bilaterally, although more obviously on the left. At 2 years postop (b), there is obvious recurrence of hip dysplasia. Both femoral heads are

subluxated with the left hip almost dislocated and both acetabula are dysplastic. (c) At 2 years postop, the false profile view of the left hip shows an anterior dislocation and anteverted acetabulum. Clinical photos of the left hip going from a dislocated position (d) in extension to relocated (e) in flexion

which femoral head sphericity is maintained. While there is no clinical evidence on the efficacy of capsulorrhaphy in CMT, it is recommended in cases where the hip fully reduces under anaesthetic in the abduction internal rotation position but instability persists after osteotomies have been performed. A capsulorrhaphy can also be advocated for in patients with a very high migration percentage as these patients are more likely to have persistent instability post operatively.

The adolescent or young adult patient with acetabular dysplasia is treated with a PAO so long as there is minimal osteoarthritis (Tönnis grade 0 or 1) and there is good to excellent congruency of the hip joint as per the Yasunaga classification of congruency [20] on an abduction internal rotation radiograph (Fig.  22.8). Congruency is considered excellent when there is near identical configuration of the acetabulum and the femoral head and good when the subchondral surfaces are not identical but the joint space is uniformly well

22  The Hip in Charcot-Marie-Tooth Disease

preserved [20]. The goals of surgery in the adolescent or young adult patient are to improve femoral head coverage, acetabular index and achieve hip stability. If a PAO is not feasible due to the size of the posterior column, or if the triradiate cartilage is still open, a triple pelvic osteotomy may be performed. A femoral osteotomy is added only if there is excessive valgus or femoral

591

anteversion, or for persistant hip subluxation after a PAO or triple, to help achieve a concentric hip reduction (Figs. 22.9 and 22.10). “The goals of surgery in the adolescent or young adult patient are to improve femoral head coverage, acetabular index and achieve hip stability”.

a

b

c

d

Fig. 22.8 (a) Abduction internal rotation pelvic radiograph for the patient featured in Figs. 22.1, 22.2, 22.6 and 22.7 shows that the femoral heads center. Staged bilateral PAO is planned along with a derotation osteotomy of the left femur. (b) At the time of surgery, the PAO was performed first and the femoral head appeared stable on an oblique fluoroscopic view with the hip in extension. During hardware removal from the femur purulent fluid was noted and therefore the femoral osteotomy was not

e

performed. (c) AP pelvis shows adequate lateral coverage of the femoral head. Interestingly, the patient’s pelvic retroversion has increased dramatically which will in turn increase the risk of an anterior subluxation. Left (d) and right (e) false profile X-rays show improvement in anterior coverage and both hips are no longer subluxated anteriorly although anterior coverage is still not optimal. This is likely secondary to the increase in pelvic retroversion since surgery

N. Saran

592

a

c

d

b

Fig. 22.9 (a) False profile view of the left hip two years after revision surgery for the patient featured in Figs. 22.1, 22.2, 22.6 and 22.8 shows that it is re-subluxated; the patient is booked for a femoral derotation osteotomy and another PAO to further improve anterior coverage. (b) The latest AP pelvic radiograph done 2 years after his last surgery show well centered hips with lateral over-coverage. False profile X-rays of the right hip (c) show that the hip remains well

aligned and the left hip (d) is not subluxated and is adequately covered anteriorly (although additional anterior coverage would be optimal). His pelvis remains retroverted and a hip flexor strengthening physiotherapy program was initiated in the hope that his pelvis will antevert and improve anterior femoral head coverage. The patient is ambulatory and has no hip pain although he is limited due to pain in his cavovarus feet (reconstructed 1 year earlier)

When performing redirectional pelvic osteotomies in patients with CMT, one must take care to not retract or tension the sciatic nerve, due to the increased risk of nerve palsy [3, 9, 15, 18]. Some authors recommend routine use of intraoperative neuromonitoring when performing hip reconstruction surgery in patients with CMT [3]. Post operatively, these patients are kept toe touch or non-weight bearing for 6  weeks until

substantial bony callus is observed, at which point physiotherapy and weight-bearing is initiated. “When performing redirectional pelvic osteotomies in patients with CMT, one must take care to not retract or tension the sciatic nerve, due to the increased risk of nerve palsy”.

22  The Hip in Charcot-Marie-Tooth Disease

a

593

b

c

d

e

f

Fig. 22.10 (a) AP pelvic X-ray of a 14 year-old girl with a known family history of CMT. She had a 1 year history of increasing left hip pain. Her AP pelvis X-ray shows a break in Shenton’s line, increased medial space, increased acetabular index, decreased center edge angle, and a peripheral rim fracture from overloading of the lateral acetabular edge. She also has coxa valga with a neck shaft angle of 145°. Her right hip is mildly dysplastic with no subluxation. (b) False profile view of the left hip highlights decreased ante-

rior coverage. (c) Abduction internal rotation view shows that the hip centers nicely. (d) Post-operative AP pelvic X-ray shows a well aligned hip after a PAO and 15°- varus 20°-derotation osteotomy. The center edge angle measures 40° and Shenton’s line is restored. (e) Post-operative false profile X-ray shows the hip to be well reduced and an anterior center edge angle of 40°. The hardware was removed after 2 years and at 3 year follow-up, the hip remains stable on AP pelvis X-ray (f) and the patient is pain free

N. Saran

594

An adolescent or young adult with a symptomatic subluxated hip, with minimal osteoarthritis (Tönnis grade 0 or 1) that cannot be made concentric or congruent, may be treated with salvage procedures, such as a Chiari osteotomy or shelf augmentation, with or without a femoral osteotomy. Complication rates of surgical intervention in this patient population are higher than in patients with DDH. There is a known increased risk of sciatic and peroneal nerve palsies associated with redirectional osteotomies [3, 9, 15, 18]. It is felt that due to their underlying peripheral neuropathy, the sciatic and peroneal nerves are extremely sensitive to traction or pressure. Stover and colleagues had one patient out of 14 patients that underwent a PAO with transient complete bilateral peroneal nerve palsies. Novais and colleagues had a severe sciatic nerve palsy that did not recover in one out of 27 patients that underwent a PAO.  They also described a 33% complication rate for sequelae requiring additional treatment, compared to 13% in DDH.  These complications included: nerve palsy, heterotopic ossification, infection, pubic nonunion, persistent pain, unplanned surgery, and conversion to total hip arthroplasty [9]. An increased risk of postoperative pubic and ischial fractures has also been described after PAO in this population [15, 16].

Essential Surgical Techniques

Pelvic Osteotomies • San Diego acetabuloplasty is performed in patients with an open triradiate cartilage when global acetabular hypoplasia is suspected and the acetabulum is capacious. • A Pemberton acetabuloplasty is optimal for anterolateral deficiency of a capacious acetabulum when the triradiate cartilage is still open • Triple pelvic osteotomy is performed in patients with an open triradiate cartilage

when the acetabulum is not capacious. It is also used when the triradiate cartilage is closed but the pelvis is too small for a Bernese periacetabular osteotomy. • Bernese periacetabular osteotomy is performed in patients with a closed triradiate cartilage for hips where a concentric reduction is either present or can be achieved, for example by varus proximal femoral osteotomy. • Chiari osteotomy or shelf augmentation can be considered in young adults for subluxated and symptomatic hips in which a concentric reduction cannot be achieved. Femoral Osteotomies • Varus osteotomy is performed for coxa valga in patients with an open triradiate cartilage. This will optimize centralization of the femoral head within the acetabulum in order to decrease chances of recurrence which are higher in younger patients. It is also used when the triradiate cartilage is closed if the hip is not centered after a pelvic osteotomy despite having a reducible hip or if the coxa valga is excessive (>145°). • The amount of varus is decided based on the amount of abduction required to center the femoral head in the acetabulum (i.e. restore Shenton’s line and increase the acetabular coverage of the femoral head). The neck shaft angle should not be reduced to less than 120° to prevent post-operative relative abductor insufficiency. If a varus osteotomy is performed for excessive coxa valga without instability, the neck shaft angle should be decreased to no less than 130° in order to limit the effect of abductor dysfunction related to varusization. • Femoral derotation is often performed when there is excessive anteversion con-

22  The Hip in Charcot-Marie-Tooth Disease

tributing to anterior hip instability, (typically identified on the false profile view; Fig.  22.7c) not solved by a redirectional acetabular osteotomy alone. Open Reduction • Although there is no literature available to guide treatment of a fixed hip dislocation in CMT patients, we suggest that the rules used to decide on whether to reduce a fixed dislocation in a spastic cerebral palsy patient can be extended to children and adolescent patients with CMT (see Chap. 20). So long as the fixed dislocation is not chronic and the femoral head has maintained its sphericity, an open reduction is recommended. Capsulorrhaphy • Capsulorrhaphy is performed –– whenever an open reduction is required to achieve a concentric reduction (i.e. when the femoral head does not fully reduce on abduction internal rotation X-rays, even under general anaesthesia). –– for patients with migration percentage close to or greater than 100%. –– if there is evidence of persistent intraoperative fluoroscopic instability after acetabular and femoral osteotomies have been completed. This can be seen by putting the hip in flexion, adduction and internal rotation for posterior instability and in extension and external rotation for anterior instability. Also, the leg should be adducted to assess for lateral instability. In these extreme positions, fluoroscopy should be used to assess for subluxation of the joint which if present, needs to be addressed.

595

Operative Pitfalls

• Instability should be assessed after pelvic and or femoral osteotomies both clinically and fluoroscopically by putting the hip into the provocative positions for anterior, lateral and posterior subluxation. Instability should be addressed with a capsulorrhaphy. • Varus osteotomies should be performed with an offset device to allow medialization of the shaft in order to realign the mechanical axis and prevent the varus bowing of the proximal femur that often o ccurs if the axis not considered. Some CMT patients will require a total hip replacement in the future and, therefore, maintaining proximal femoral anatomic alignment is important to allow for later passage of a femoral stem. • Acetabular anteversion can contribute to anterior instability. A preoperative evaluation of acetabular version with either an AP pelvic radiograph or a CT scan should be performed if there is a concern for anterior instability. In such a case, anterior coverage may be required (Figs. 22.7 and 22.8). • Attempts should be made to improve lateral and anterior femoral head coverage to the high end of normal values for center-edge angle. In adolescent and young adult patients, one should aim for center edge angles closer to 40° than to 25°. That said, one must be careful not to cause iatrogenic femoroacetabular impingement by over correction.

596

N. Saran

Orthop Relat Res. 2014 Feb;472(2):665–73. This is a retrospective comparative study between CMT and DDH patients that underwent Kumar SJ, Marks HG, Bowen JR, MacEwen PAO. Fourteen patients (19 hips) with CMT hip GD.  Hip dysplasia associated with Charcot-­ dysplasia set to undergo PAO were retrospecMarie-­ Tooth disease in the older child and tively evaluated using radiographs and CT scans. adolescent. J Pediatr Orthop. 1985 Sep-­ These were compared to a matched cohort of 45 Oct;5(5):511–4. An association between CMT patients (45 hips) with DDH set to undergo a and hip dysplasia is realized in this case series of PAO. The authors found that CMT hip dysplasia five patients. The authors noted that hip dysplasia is more severe than DDH. In addition, there were in this patient group is minimally symptomatic more arthritic hips in the CMT group. Both latand, in order to detect the problem early, recom- eral and anterior center edge angles and acetabumended routine and regular hip examinations in lar volumes were smaller. Also, anterior and patients with CMT.  They also described two posterior osseous support was lower, the area of cases of sciatic nerve palsies following redirec- the femoral head covered was less; hip subluxtional osteotomies of the pelvis and recom- ation, coxa valga, acetabular anteversion, and mended care in protecting the sciatic and peroneal Tönnis angles were greater. Novais EN, Kim YJ, Carry PM, Millis nerves. Finally, they highlighted the importance of getting these patients back on their feet as soon MB.  Periacetabular Osteotomy Redirects the as possible to limit the deconditioning that can Acetabulum and Improves Pain in Charcot-­ Tooth Hip Dysplasia With Higher occur with prolonged immobilization or succes- Marie-­ Complications Compared With Developmental sive surgeries over short intervals. Dysplasia of the Hip. J Pediatr Orthop. 2016 Dec;36(8):853–859. This is a retrospective Key Evidence review of 27 patients that underwent PAO for CMT hip dysplasia looking at complication rates van Erve RH, Driessen AP. Developmental hip and patient-reported outcome measures, comdysplasia in hereditary motor and sensory neu- pared to a matched cohort of 54 DDH patients. ropathy type 1. J Pediatr Orthop. 1999 Jan– The median modified Harris Hip Score went Feb;19(1):92–6. This report is a case series of from 63 preoperatively to 88 at final follow-up on three patients with CMT. Two of these patients had average 5.2  years after reconstruction, showing multiple normal radiographs at various time points an overall beneficial effect. Complication rates in prior to the development of hip dysplasia confirm- the CMT group were higher than in the DDH ing that these hips are normal initially. In addition, group; 33% of CMT compared to 13% of DDH the authors report one of the patients developing a patients had a complication that required medical peroneal nerve palsy in the postoperative period or surgical treatment. The complications in the after a triple pelvic osteotomy. The authors state CMT group included one heterotopic ossificathat the nerve palsy was related to direct compres- tion, one nonunion with minimal symptoms, one sion of the nerve on a traction mattress. infection, one pubic nonunion requiring bone Novais EN, Bixby SD, Rennick J, Carry grafting, two with persistent pain and unplanned PM, Kim YJ, Millis MB. Hip dysplasia is more surgery, two failures that were converted to total severe in Charcot-Marie-Tooth disease than in hip arthroplasty, and one severe sciatic nerve developmental dysplasia of the hip. Clin palsy.

Classic Papers

22  The Hip in Charcot-Marie-Tooth Disease

Take Home Messages

• Hip dysplasia in CMT is often silent and therefore annual radiographic surveillance is recommended with an AP pelvis radiograph. • Children presenting for the first time with hip dysplasia after age 8 with no known past medical history should be evaluated for CMT. • Recurrence of hip subluxation in a patient otherwise presumed to have DDH, should be evaluated for CMT along with other connective tissue and neuromuscular disorders. • Asymptomatic skeletally immature hips with mild dysplasia that are not subluxated can be monitored with regular radiographs. • Symptomatic hips and asymptomatic subluxated hips in adolescent or young adult patients should be treated with a pelvic osteotomy with or without a femoral varus derotation osteotomy. • Complication rates related to CMT hip dysplasia reconstructions is higher than in DDH. • Peripheral nerves are predisposed to injury in patients with CMT undergoing hip reconstruction. It is imperative to minimize traction or pressure to peripheral nerves during a surgical intervention as well as to carefully pad nerves at risk from injury related to positioning. • Intraoperative neuromonitoring of the sciatic nerve should be considered for pelvic redirectional osteotomies due to the higher risk of neural injury in CMT. • Recurrence of hip dysplasia after reconstruction is common. These hips should be followed closely, especially in the childhood and adolescent onset group.

597

References 1. Bamford NS, White KK, Robinett SA, Otto RK, Gospe SM Jr. Neuromuscular hip dysplasia in CharcotMarie-Tooth disease type 1A. Dev Med Child Neurol. 2009;51(5):408–11. 2. Carter GT, Abresch RT, Fowler WM Jr, Johnson ER, Kilmer DD, McDonald CM. Profiles of neuromuscular diseases. Hereditary motor and sensory neuropathy, types I and II. Am J Phys Med Rehabil. 1995;74(5 Suppl):S140–9. 3. Chan G, Bowen JR, Kumar SJ. Evaluation and treatment of hip dysplasia in Charcot-Marie-Tooth disease. Orthop Clin North Am. 2006;37(2):203–9. 4. Cucuzzella TR, Guille JT, MacEwen GD.  Charcot-­ Marie-­Tooth disease associated with hip dysplasia: a case report. Del Med J. 1996;68(6):305–7. 5. Fuller JE, DeLuca PA.  Acetabular dysplasia and Charcot-Marie-Tooth disease in a family. A report of four cases. J Bone Joint Surg Am. 1995;77(7):1087–91. 6. Hadianfard MJ, Ashraf A.  Hip dysplasia associated with a hereditary sensorimotor polyneuropathy mimics a myopathic process. Ann Indian Acad Neurol. 2012;15(3):211–3. 7. Kumar SJ, Marks HG, Bowen JR, MacEwen GD. Hip dysplasia associated with Charcot-Marie-Tooth disease in the older child and adolescent. J Pediatr Orthop. 1985;5(5):511–4. 8. Lerch TD, Steppacher SD, Liechti EF, Tannast M, Siebenrock KA. One-third of hips after periacetabular osteotomy survive 30 years with good clinical results, no progression of arthritis, or conversion to THA. Clin Orthop Relat Res. 2017;475(4):1154–68. 9. Novais EN, Kim YJ, Carry PM, Millis MB. Periacetabular osteotomy redirects the acetabulum and improves pain in Charcot-Marie-Tooth hip dysplasia with higher complications compared with developmental dysplasia of the hip. J Pediatr Orthop. 2016;36(8):853–9. 10. Novais EN, Bixby SD, Rennick J, Carry PM, Kim YJ, Millis MB.  Hip dysplasia is more severe in Charcot-Marie-Tooth disease than in developmental dysplasia of the hip. Clin Orthop Relat Res. 2014;472(2):665–73. 11. Pailthorpe CA, Benson MK. Hip dysplasia in hereditary motor and sensory neuropathies. J Bone Joint Surg Br. 1992;74(4):538–40. 12. Ramdharry GM, Day BL, Reilly MM, Marsden JF. Hip flexor fatigue limits walking in Charcot-Marie-­ Tooth disease. Muscle Nerve. 2009;40(1):103–11. 13. Ramdharry GM, Pollard A, Anderson C, Laurá M, Murphy SM, Dudziec M, Dewar EL, Hutton E, Grant

598 R, Reilly MM.  A pilot study of proximal strength training in Charcot-Marie-Tooth disease. J Peripher Nerv Syst. 2014;19(4):328–32. 14. Skre H.  Genetic and clinical aspects of Charcot-­ Marie-­Tooth’s disease. Clin Genet. 1974;6(2):98–118. 15. Stover MD, Podeszwa DA, De La Rocha A, Sucato DJ. Early results of the Bernese periacetabular osteotomy for symptomatic dysplasia in Charcot-Marie-­ Tooth disease. Hip Int. 2013;23(Suppl 9):S2–7. 16. Swann M, Sucato DJ, Romero J, Podeszwa DA.  Fracture at the ischio-pubic junction after periacetabular osteotomy in the adolescent population. J Pediatr Orthop. 2017;37(2):127–32.

N. Saran 17. Ushiyama T, Tanaka C, Kawasaski T, Matsusue Y. Hip dysplasia in Charcot-Marie-Tooth disease: report of a family. J Orthop Sci. 2003;8(4):610–2. 18. van Erve RH, Driessen AP.  Developmental hip dysplasia in hereditary motor and sensory neuropathy type 1. J Pediatr Orthop. 1999;19(1):92–6. 19. Walker JL, Nelson KR, Heavilon JA, Stevens DB, Lubicky JP, Ogden JA, VandenBrink KA. Hip abnormalities in children with Charcot-Marie-Tooth disease. J Pediatr Orthop. 1994;14(1):54–9. 20. Yasunaga Y, Ikuta Y, Kanazawa T, Takahashi K, Hisatome T.  The state of the articular cartilage at the time of surgery as an indication for rotational acetabular osteotomy. J Bone Joint Surg Br. 2001;83(7):1001–4.

The Hip in Poliomyelitis

23

Hugh G. Watts, Benjamin Joseph, and Sanjeev Sabharwal

Introduction Poliomyelitis is a viral disease involving the anterior horn cells of the spinal cord, resulting in paralytic muscle imbalance involving the extremities and trunk. Under the influence of musculoskeletal growth, this imbalance leads to the development of progressive soft tissue and bony deformities, including those involving the hip. Problems for children and adolescents involving the hip following poliomyelitis fall into three main spheres: (1) deformities resulting from contractures, (2) weakness or imbalance of muscles acting on the hip joint, and (3) hip joint instability. Hip contractures following acute poliomyelitis are extremely common; treatment is simple. Paralysis of the gluteal muscles results in an unattractive gait pattern that is very energy demanding; treatment is often unsatisfactory. Hip dislocation, fortunately, is very uncommon, but extraordinarily difficult to manage when it does occur. H. G. Watts (*) Shriners Hospitals for Children, Los Angeles, CA, USA e-mail: [email protected] B. Joseph Aster Medcity, Kochi, Kerala, India S. Sabharwal Department of Orthopedic Surgery, Benioff Children’s Hospital of Oakland, University of California, San Francisco, Oakland, CA, USA © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_23

The literature concerning treatment options in poliomyelitis is scant and outdated. The quality of the evidence, by current standards, is very low. Most published series are small. Treatments are not compared to other options; random assignment of patients to treatment options has never been reported. Now, fresh cases are too few and scattered worldwide, making the gathering of new, more reliable evidence nearly impossible. This must be kept in mind in reading the following material.

Pathophysiology Poliomyelitis is caused by an enterovirus (poliovirus) infection. The offending organism enters through the gastrointestinal (GI) tract and spreads in the serum. It enters the central nervous system through peripheral nerve roots. It migrates proximally to the anterior horn cells in the spinal cord where it causes motor neuron injury and death. This then causes paralysis of the muscle supplied by those neurons. Poliovirus is spread from one person to the next by the fecal/oral route through contamination of food and water. In those infected, the virus is shed from the GI tract for up to 1 month. There is a 7- to 14-day incubation period. Approximately 50% of those infected will show no clinical ­illness. Another 48% will have an abortive illness consisting of muscle aches or headaches characteristic of an irritation of the meningeal linings of the brain and spinal cord. 599

H. G. Watts et al.

600

Only about 1–5 in 1000 children who get polio progress to paralytic disease, in which the muscles become weak, floppy and poorly controlled, and, finally, completely paralyzed; a condition also known as acute flaccid paralysis. “Only about one to five in 1000 children who get polio progress to paralytic disease”. There are three types of poliovirus. It is the Type 1 virus which is the scourge. Types 2 and 3 have been virtually eliminated. Sharrard demonstrated that the motor neurons in the anterior horns of the spinal cord responsible for specific muscles are clustered in vertical columns within these anterior horns [44]. The spinal cord infection is a limited lesion affecting only a local area of these columns. Columns of cells innervating the upper lumbar region tend to be longer than those affecting the lower lumbar and sacral portion of the spinal cord (hence more cells survive unaffected). Consequently, the limited lesions caused by the infection have a greater impact on the muscles innervated by the lower lumbar and sacral regions. While lesions more commonly affect the proximal muscles of the lower extremities, it is the distal muscles that lose more function.

Weakness and Recovery Manual muscle testing can evaluate each major muscle group. The definition of muscle strength recovery is the improvement by one manual muscle grade. It is generally seen within a 6-week time frame. There are many differing factors affecting the prognosis for recovery. Children recover faster. One third of all affected individuals will have no recovery. A muscle demonstrating no function after 3 months from illness will have no recovery. Early improvement in even a Grade 1 muscle has a good prognosis for a full recovery. Most of the recovery occurs in the first 12 months; there is only 5% recovery thereafter. There is no additional recovery of muscle function 24 months after illness. Recovery occurs for two reasons. First there is resolution of intracellular edema in the cells that have not died. The resultant anterior horn cells can function but may be damaged leaving them

vulnerable to early degeneration later in the life. Second, with muscle fibers left orphaned by the death of their controlling anterior horn cell there is sprouting of axons from adjacent unaffected cells thus supplying new motor end plates. Later, the resulting giant axons may be vulnerable to early degeneration. During the convalescence following the acute polio infection, there develops a slowly progressive fibrous metaplasia of the paralyzed muscular fibers as well as the interstitial tissues, leading to contractures.

Natural History Recovery After Poliomyelitis The recovery of a child after developing paralytic poliomyelitis can be broken into three phases: acute, convalescent, and chronic.

Acute Phase This spans the time from the onset of the disease until the child has been afebrile for 48 h. Principally, the care in this phase will be in the hands of the pediatrician or internist. The early focus is on providing comfort for the child. Muscle spasms must be treated, and hot packs are the mainstay. Positioning to prevent contractures is important. Ideally, even during this early phase, the orthopedic surgeon and the rehabilitation personnel should be involved to establish a base line against which recovery will be recorded. Convalescent Phase This usually lasts 16–24 months after the disease onset. During this time, muscles that are going to recover will do so, mostly in the first 12 months. The objectives during this phase are to attain maximum recovery of involved muscles, ­maintain, or restore, normal joint range of motion by physiotherapy, and to prevent deformities by good positioning. Some corrective surgery may be required but usually surgical interventions are delayed until the chronic phase. Initially hot packs for spasms may be required. Passive range of motion exercises are helpful. Some use of orthoses may be needed.

23  The Hip in Poliomyelitis

Chronic Phase This phase extends from about 24 months after disease onset, after which no further recovery should be expected for the life-time of the child. This chapter deals primarily with children in the chronic phase, as it is during this period that the musculoskeletal manifestations of the disease have reached their steady state and functionally limiting contractures and/or deformities may be treated with or without surgery as needed.

Importance of Age at Onset Most commonly, polio strikes infants after the age of 6–12 months when the immunity of breast milk antibodies from the mother has waned. In countries where the disease has been absent for some time, young adults are also at risk. If the initial paralysis occurs in early childhood, muscle weakness may affect the growth in length, girth and shape of the bones. The proximal femur may show an increased neck-shaft valgus angle, and the acetabulum and lateral pelvic wall may not develop adequately. The earlier the child develops the infection, the greater the chances of developing subsequent bony deformity.

Impact of Hip Problems in Polio Contractures of the Hip Deformities commonly seen in the hip include flexion, abduction and external rotation contractures, often in combination (Fig.  23.1). Adduction and internal rotation deformities are much less frequent. Abduction or adduction deformities cause pelvic obliquity and a compensatory lumbar scoliosis, predisposing to hip dislocation, especially on the high side of the pelvis (Fig. 23.2). Scoliosis due to polio may be the primary cause of pelvic obliquity in some children (i.e. supra-pelvic cause) (Fig. 23.3). These deformities are due to soft tissue contractures to begin with, though in long-standing cases, adaptive bony changes may supervene. Careful muscle power testing prior to soft tissue release

601

Fig. 23.1  Deformities commonly seen in the hip are flexion, abduction and external rotation, often in combination

is of paramount importance when dealing with an abduction deformity. If the hip abductors are paralyzed, retaining a mild degree of a fixed abduction deformity can decrease a Trendelenburg gait (i.e. under-correction is desirable if there is hip abductor weakness).

Hip Dislocation A hip may become unstable mechanically, even to the point of dislocation due to several factors: a severe leg length discrepancy with shortening of the opposite side, inadequate lateral hip coverage due to insufficient lateral acetabular development, or pelvic obliquity from lumbar scoliosis or imbalance of adduction/abduction on the two sides. There may also be proximal femoral valgus due to growth disturbances resulting from abductor weakness, leading to uncovering of the femoral head and a loss of the stabilizing effect of the acetabular roof. The exact cause of hip dislocation after poliomyelitis is uncertain [28]. Elzinga, [11] after careful evaluation of 26 children stated that “no definite type of paralysis was responsible for the ­dislocation, but that dislocation was most frequent in hips with strong flexors, adductors, and internal rotators and weak extensors, abductors, and external rotators.” “No definite type of paralysis is responsible, but dislocation is most frequent in hips with strong flexors, adductors, and internal rotators and weak extensors, abductors, and external rotators”.

602

H. G. Watts et al.

Fig. 23.2  Hip contractures (abduction or adduction) or lumbar scoliosis can lead to pelvic obliquity with dislocation of adducted hip. Top: a right hip abduction contracture leading to a contralateral hip dislocation. Middle: a

left hip adduction contracture leading to ipsilateral hip dislocation. Bottom: a right lumbar scoliosis leading to adduction of the left hip with dislocation of the left hip

The age at which the child contracted polio, however, does appear to be an important feature. Parsons and Seddon [27] found that only 16 children had paralytic subluxation or dislocation and all but one contracted polio before age 18 months. Of the 11 children with a hip dislocation that Weissman [38] treated, seven had their polio before the age of 2 years. Lau [18] also reported that almost two thirds of his patients had polio before the age of 2 years.

(AFP) has markedly increased. After India was declared “polio-free”, 59,436, 53,421 and 53,383 AFP cases were reported annually between 2012 and 2014; none of whom tested positive for polio [6, 12, 25, 42]. In addition, there is a new cluster of children, mostly in the USA, designated as having Acute Flaccid Myelitis (AFM). As of June 30, 2017, there have been 285 children reported with AFM [7]. The etiology of this syndrome has not been established, though it is possible that this is caused by a polio-like enterovirus. Viruses D68 (EV-D68) and 71 have been implicated but not all children have had one or other of these viruses isolated. We need to think not only of “Polio” but also of these “Polio-Like” diseases as the consequences of paralysis may be similar.

Epidemiology  nderstanding Polio and Polio-like U Diseases The incidence of poliomyelitis worldwide has declined dramatically from an estimated 350,000  in 1988 to 74 reported cases in 2015. However, the incidence of polio-like diseases, collectively referred to as “Acute Flaccid Paralysis”

“We need to think not only of “Polio” but also of these “Polio-Like” diseases as the consequences of paralysis may be similar”.

23  The Hip in Poliomyelitis

603

Clinical Presentation Acute Stage Following the acute episode of paralytic poliomyelitis, the child is usually left lying supine with the hip in flexion, abduction and external rotation. During the early weeks, pain in the affected limbs may be marked, requiring nursing with hot packs and limiting the use of passive exercises.

Convalescent to Chronic Stages

Fig. 23.3  Scoliosis due to spinal involvement with poliomyelitis may be the primary cause of hip problems

Prevalence of Hip Problems in Polio Though contractures at the hip are common, hip dislocation is rare; Miller and Irwin [22] reported only 20 paralytic hip dislocations among 5400 patients with poliomyelitis.

Hip contractures are frequently seen in association with knee flexion and ankle plantar flexion contractures. With time, these contractures may lead to skeletal deformities including external femoral torsion and pelvic obliquity. An increase in lumbar lordosis, initially mild and flexible, may be seen as the weight of the thighs is pulled to the bed. Later, when the child tries to stand or walk and needs to have his thigh at right angles to the floor, this lordosis may become alarming. Subsequently, this spinal flexibility may give way to a fixed lordosis. The external femoral torsion, together with contracture of the tensor fascia lata, commonly leads to genu valgum and an external rotary subluxation of the knee (with the tibia rotated on the on the femur) (Fig. 23.4). Measurements of the contractures are helpful for following the progress of treatment for the contractures. While it is recognized that clinical measurement of the exact degree of contracture is unreliable, measurement by a goniometer (a measurement that is noted on the patient’s clinical record) is much preferable to an approximation that goes unrecorded and soon forgotten. Late-presenting contractures are usually accompanied by severe muscle wasting and bony changes in the shape of the joint.

604

H. G. Watts et al.

been developed but are expensive and frequently unavailable in clinical settings. In the future, other quantifying techniques, such as measurement of muscle cross-section by MRI may become useful as a proxy for strength. In 1961 Beasley [2] published an important study outlining what manual muscle testing grades really represent in polio patients, because of the long-term implications. All individuals lose muscle strength with increasing age, so that grade 3 quadriceps strength (representing 10% of true normal) means that a person may drop to a grade 2 with age—and not be able to function at their previous level. A similar deterioration is possible for grades 4 and 5. A person who might have been able to walk without bracing or crutches may subsequently require external supports to continue.

 linical Presentation of Specific C Deformities

Fig. 23.4  Flexion, abduction and external rotation contractures of the left hip in a boy following polio. The flexion deformity of the knee and the equinus deformity of the ankle are evident. External rotation of the right leg is also present

Muscle Strength Testing in Polio As the primary musculoskeletal deformities in polio—and their successful surgical management—are dependent on the presence of muscle imbalance and its correction (i.e. rebalancing, typically via tendon transfer; see “Operative Management” section), the determination of peri-articular muscle strength is paramount. Muscle strength determination in children with polio can be difficult and manual muscle testing for strength in children in general has its limitations. Isolating an individual muscle to test can be a challenge depending on the cooperation of the child, the presence of obesity, and the experience and skill of the tester. More reliable methods of strength testing such as dynamometry have

 ip Abductor Paralysis H Occasionally the hip can be flail, with paralysis of all the muscles acting on it. More frequently some muscles are spared. If the gluteus medius alone is paralyzed the child walks with a lean of the trunk to the affected side in stance phase, the so-called “abductor lurch” (also known as a “Duchenne lurch” or “Trendelenburg lurch) which is both unsightly and energy-inefficient.  ip Extensor Paralysis H Paralysis of the gluteal maximus is also disabling with a characteristic lurch where the lumbar spine arches backwards during the stance phase of gait to compensate for the loss of hip extension power. If the weakness is in the maximus alone, the truncal lean is backwards. If, in addition, there is weakness of the abductors the trunk leans laterally and posteriorly in stance. Hip Instability From a clinical standpoint, the subluxated or dislocated hip most commonly manifests as a gait disturbance secondary to a relative abductor

23  The Hip in Poliomyelitis

insufficiency, limb length discrepancy, or both. Thankfully, the unstable hip in polio is rarely painful. A coincident abduction contracture in the contralateral hip can lead to fixed pelvic obliquity and lumbar scoliosis. The converse can also occur, whereby a structural lumbar scoliosis providing a “supra-pelvic cause” of pelvic obliquity results in femoral head uncovering and hip instability. Hence, any clinical examination of the hip in polio should also include a detailed examination of the spine. If a scoliosis is detected, one must then determine whether it is compensatory (i.e. due to an “infra-pelvic cause”) or structural in nature. “Thankfully, the unstable hip in polio is rarely painful”.

Essential Clinical Tests

• Mandatory complete manual muscle strength and joint range of motion testing to include upper extremity and trunk and not just the hip. • Leg length evaluation including estimated discrepancy at maturity. • Scoliosis assessment.

Imaging Plain radiographs at this early stage are generally not helpful. During the late stage, however, X-rays may be helpful in determining the extent of bone and joint deformity. Assessment of femoral neck-shaft angle, the extent of acetabular dysplasia, and the degree of joint arthrosis, are primary considerations when considering operative management. The use of special imaging techniques such as 3D CT measures of acetabular deficiencies does have a place in the literature of poliomyelitis which long antedated these techniques. We have not found them to be of particular use to us currently. Occasionally the use of arthrography has been reported but its value has not been proven.

605

Non-operative Management I nterrelationship Between Hip Problems and Other Impairments Problems related to the hip cannot be dealt with in isolation of the body’s total balance of needs. For instance, a deficient opponens muscle to the thumb may make standard crutch use very difficult and may require an alteration in plans for hip surgery. A knee-ankle-foot orthosis (KAFO) necessary for stabilizing the knee in extension may make obvious a hip flexion contracture that makes it impossible for the child to stand up, let alone walk. Being aware of all such interrelationships in muscle groups forms the basis of the art of managing a child with poliomyelitis.

Acute and Convalescent Stages Preventive positioning of the extremities at the earliest possible time is invaluable. The parents and attendants must be made aware of the importance of aligning the limbs without abduction of the hips and any support from under the knees should be removed. Pillows placed along the sides of the thighs can help to reduce external rotation. Proper positioning of the other affected joints should not be ignored while focusing on the hips. Passive range of motion, within the limits tolerated during the very early—and typically painful—stage, should be encouraged. Bracing for the prevention of hip contractures is not practical during the early stage of the disease, but simple ankle-foot orthoses (AFOs) and KAFOs (made as polypropylene moulds without hinges) may be useful for the prevention of contractures in the knees and ankles during this period. Flexion or abduction contractures of the hip of less than 30° can usually be managed non-­ operatively with stretching exercises, traction or serial casting. Measurements and documentation of the contractures can be helpful for following the progress of treatment. Flexion and/or abduction

H. G. Watts et al.

606

contractures of the hip of less than 30° can usually be managed non-operatively. The difficulty in stabilizing the pelvis, once larger hip contractures have developed, makes stretching exercises, traction or casting unlikely to succeed.

Chronic Stage Pre-operative or post-operative skeletal traction may be of help when addressing contractures greater than 60°. Stabilizing the pelvis may be difficult, especially if both hips are similarly involved. In patients with unilateral deformity of the hip, flexion of the opposite limb helps to stabilize the pelvis.

Non-operative Management for Specific Deformities  ip Abductor Paralysis H The use of crutches (axillary or forearm)—if the muscles of the upper extremity allow it, particularly the triceps and the thumb opponens— can certainly help to reduce an abductor lurch. However, crutches are often considered a nuisance by the patient, particularly because they inhibit hand use. “The use of crutches can certainly help to reduce gait disturbances such as an abductor lurch, but are often considered a nuisance by the patient, particularly because they inhibit hand use”.

 ip Extensor Paralysis H As for weakness of the hip abductors, crutches (axillary or forearm) may be used if the muscles of the upper extremity allow it, particularly the triceps. However, the gluteus maximus weakness requires that one crutch be supporting at all times, or the patient falls forward. Hip Instability Most often, unstable hips in polio are not painful and as such, “benign neglect” may be the most prudent course of management. This avoids stiff-

ness due to operations especially in floor-sitting cultures and those who squat for their toilet needs. Braces and crutches may also be useful for gait stabilization. However, weakness in the upper extremities, especially the triceps and the opponens of the thumb, may limit the use of crutches, and crutches are often considered a nuisance by the patient. “Most often, unstable hips in polio are not painful and as such, “benign neglect” may be the most prudent course of management”.

Essential Non-operative Management Methods

• Preventive positioning of the extremities (acute/convalescent stages) • Stretching exercises for flexion or abduction contractures less than 30° • Use of crutches to help to reduce an abductor lurch or to support loss of hip extensor strength • Judicious use of bracing for gait stabilization where needed • Avoid iatrogenic stiffness by avoiding surgery in asymptomatic, dislocated hips • Use crutches if the power of the upper limb muscles permit

Operative Management Treatment of Flexion and Abduction Contractures Convalescent Stage During the early stages, a hip flexion or abduction contracture in excess of 30° can usually be managed by surgical release of the tensor fascia latae [5] as described by Yount [43]. This release also helps with the associated flexion contractures at the knee. Huckstep’s [16] enthusiastic emphasis on the use of multiple subcutaneous incisions stemmed

23  The Hip in Poliomyelitis

from the huge needs (and limited resources) that he faced in Africa in his day. While the subcutaneous approach can still be used, the availability of better operative facilities and better anesthesia has largely relegated this technique to history. Release of the tensor fascia latae as described by Yount is unlikely to have much effect on the hip during this stage, although its benefits to the knee may still be important. An extensive release of the tensor fascia proximally is more likely to be successful [20]. The operation ascribed to Soutter [33] is usually helpful but requires a large anterior ilio-femoral incision. Alternatively, the operation described by Ober [26], and modified by Barr [1] may be equally successful. We have found that the transfer of the iliac crest, as recommended by Campbell [4], is rarely, if ever, needed. The patient should be prepared for surgery with both lower extremities exposed so that the opposite hip can be fully flexed and adducted to control the pelvis, thereby putting the tissues holding the affected limb in abduction and flexion at maximum tension. The child is best managed post-operatively with a spica cast extending from the nipples to the knees (or to the toes if knee and ankle contractures are addressed at the same time) holding the hip in 0–10° of extension and approximately 10° of abduction for 3–4 weeks. Exercises are started soon after removal of the spica.

Chronic Stage An extensive release of the tensor fascia latae proximally is needed. Release of the sartorius rarely adds to the success. Release of the hip capsule has been rarely used in our hands for fear of developing joint instability. Residual tightness of the hip capsule can be treated by serial casting or traction. With severe deformities, femoral shortening may be needed to prevent excessive stretch of the nerves [41]. After-treatment is the same as that in the convalescent stage.

607

stiffness can be more disabling than the deformity. Other problems to be aware of are nerve stretching and concomitant malignant hypertension. Overzealous releases of the hip capsule may lead to disastrous instability of the hip joint. “Overzealous releases of the hip capsule may lead to disastrous instability of the hip joint”.

Principles of Tendon Transfer While we are not impressed by the results of tendon transfer procedures about the hip for children and adolescents, some discussion of the details of tendon transfers is important. Tendon transfer surgery may be performed in polio to improve function by augmenting muscle strength for a weakened or totally paralyzed muscle that is functionally important [9]. There are several important preoperative considerations to ponder. The goal is to restore or improve function or to prevent or decrease a deformity. Before a muscle-­ tendon unit is moved, however, there must be an adequate range of motion in the joint. A tendon transfer does not increase range of motion. “Before a muscle-tendon unit is moved, there must be an adequate range of motion in the joint. A tendon transfer does not increase range of motion”.

One must also consider how capable the transferred muscle is to replace the paralyzed muscle. The cross section of a muscle gives some idea of the stamina of muscle, i.e. not only an ability to perform the intended function, but also its ability to do so repetitively. The degree of excursion (travel) of the intended muscle for transfer may indicate whether it can achieve the desired movement or just stabilize the joint. For example, a transfer of the erector spinae to substitute for weakened gluteus maximus can only be expected to stabilize the hip in extension; rather than move Operative Pitfalls Problems of prolonged hospitalization, exten- the joint through the expected excursion, and do sive rehabilitation, in the face of inadequate post-­ so over prolonged walking. In addition to its excursion, the muscle to hospital support services must be considered before embarking on major surgery. Post-­operatively joint be transferred must be strong enough as it will

608

typically lose one grade of strength after transfer. Therefore, for a transferred muscle to be functionally useful, a grade of at least 4 (before transfer) is necessary. However, a grade 3 muscle, while unable to provide a useful function after transfer, may be effective in preventing deformity by better balancing forces about the hip, though based on our review, there is no proof of this concept. The transferred tendon must be properly positioned so that its line of pull is as straight as possible. If this is not done, and there is a large angle over which it operates, the soft tissues holding the tendon may stretch and assume a straight line of pull over time, subsequently decreasing its tension. Just how tightly the tendon is stretched to the new attachment point may weaken the transferred muscle by shifting its location on the Blix curve. The strength of a muscle depends on how much that muscle is stretched or relaxed. A muscle is strongest when it is contracting from its normal resting length. As the muscle is stretched beyond that point, the muscle gets weaker. The same is true when the muscle is allowed to shorten to less than its resting length. During a muscle-transfer operation, a surgeon may stretch the transferred tendon inordinately to avoid fixing the muscle too loosely. This may well weaken the transfer’s ability to function. The phase of a muscles activity during gait, whether during swing or stance, must also be considered. For example, the external oblique muscle is active during the swing phase and not during stance, so it cannot be expected to actively affect a Trendelenburg limp if transferred to augment weak hip abductors. There is little information in the journal articles concerning the reliability of measurements in muscle transfers about the hip, which makes assessment of the surgical results difficult.

 perative Management for Specific O Deformities  ip Abductor Paralysis H The operative procedures most commonly resorted to are: (1) transfer of the external oblique, (2) Mustard’s iliopsoas transfer, (3) free gracilis transfer.

H. G. Watts et al.

External Oblique Transfer For the treatment of hip abductor paralysis, transfer of the external oblique muscle to the greater trochanter has been reported to improve gait parameters including a reduction of abductor lurch and improved endurance [3, 29]. The virtue of this operation is that the nerve supply of the external oblique is from a different spinal segment than that of the gluteus medius and minimus, so it is less likely to be weakened or paralyzed. Manual muscle testing of the external oblique is not always easy and depends greatly on the experience of the tester. The operation, as described by Thomas [36], is straightforward but, unfortunately, has limited effectiveness. Thomas himself stated that “the abductor limp has not been completely eliminated, in any instance.” While in their hands they had “excellent results” in 3 of 19 patients, he hastened to add “it should be mentioned that each of these had slight abductor power before operation”. Electromyographic studies [3] of the external oblique muscle before and after transfer showed that the muscle functioned only during the swing phase, i.e. an “out-of-phase transfer”. One has to assume that any benefit from an external oblique transfer is thus due to tenodesis, and its efficacy will be enhanced by inserting the tendon into the greater trochanter under maximum tension. Cabaud and colleagues [3] reported an improvement in gait with decreased limp and swaying; increased ability to walk long distances was seen in 17 of 19 patients after external oblique transfer. Nine patients became free of their assistive ambulatory devices. Subjective improvement was observed in 16 patients with improved appearance being the main element of satisfaction for most patients. Similarly, Shahcheraghi [29] reported the main reasons for satisfaction was improved appearance with decreased body sway and limp while walking, sense of stability, and decreased fatigability. Hammesfahr and colleagues [15] reported preliminary results on 37 patients who were operated on between the ages of 10 years 8 months to 25 years 6 months, but followed up for only 9 months. Thirteen patients became brace-free, 23 continued to wear a KAFO and one was unchanged, wearing a hip-knee-

23  The Hip in Poliomyelitis

ankle-foot orthosis (HKAFO) postoperatively. No patients experienced Trendelenburg gait elimination, though the degree of sway decreased considerably. Cabaud and colleagues reported 14 complications in 98 external oblique transfer patients: one avascular necrosis of the femoral head, three deep and three superficial infections, three femoral fractures, two pull outs of the transfer, one fascial hernia and one transfer requiring reefing. “Following external oblique transfer, a Trendelenburg gait may not be completely eliminated, but the degree of truncal sway can be expected to decrease considerably”. Mustard’s Transfer of Iliopsoas The muscle transfer described by Mustard [23, 24] has been shown to work well to reduce the hip abductor (Trendelenburg) lurch, but requires skill and patience. Here the psoas muscle is transferred via a notch in the anterior ilium and the tendon is attached to the greater trochanter of the femur to work as a hip abductor. It is important that this is followed by a period of intense physical therapy. Later, Sharrard [31] introduced a variation of the Mustard procedure where the psoas muscle was placed through a hole in the posterior ilium and then attached more posteriorly into the greater trochanter of the femur. The idea was that the transfer would work not only as a hip abductor, but also as a hip extensor. This variant was used for a while, particularly in children with spina bifida with subluxating hips, but has been given up as it was found that it often did not work due to fibrosis of the transferred muscle, as it went through the bony hole in the ilium. In our opinion, this procedure has no place in the current management of children with polio paralysis. The Mustard procedure may work satisfactorily if the psoas muscle is strong enough. Manual muscle testing of psoas muscle is straightforward, although none of the published reports give the preoperative psoas muscle strength. Technically the Mustard’s operation is more challenging than the external oblique transfer, but with the current widespread enthusiasm for open

609

hip operations for developmental dysplasia of the hip (DDH) and slipped capital femoral epiphysis (SCFE), pediatric orthopedic surgeons are more familiar with the anatomy than they were during Mustard’s time. Mustard [24] reported that of 49 patients with polio treated with iliopsoas transfer, the youngest patient was 3 1/2 years old. The operations were done from 1 to 54 years after the acute poliomyelitis event. The age at operation had no obvious effect on outcome. A simple AP X-ray of the hips, preferably done standing, should be obtained pre-­operatively to assure that the hip is stable in the joint before tendon transfer. The ultimate post-operative results of the procedure appear to be moderate at best [40]. We do not feel that the operation is as valuable as many have considered it in the past, especially where experienced post-operative physical therapy is not readily available. The ideal patient is one in whom the hip abductors are weak but gluteus maximus, sartorius, iliopsoas, quadriceps and abdominal muscle are “Good” (i.e. 4/5). In a 1959 report of 50 of such transfers (which included 27 patients with a gluteus maximus strength of less than four out of five), only 5 (11%) had no lurch at final follow-­up. Another 13 (30%) were classified as “moderate”, or “fair” and the remaining 26 (60%) were classed as having a lurch but less than what was present pre-operatively. Lau [17] reported on 39 patients with hip instability of whom 15 had Mustard’s operation; no patient had abolition of the Trendelenburg gait. Chabaud’s comprehensive study of 24 cases of iliopsoas transfer reported femoral nerve injury (three cases), femoral head osteonecrosis (one case), and marked abduction contracture requiring z-lengthening of the transferred iliopsoas tendon (one case). One patient required reefing or tightening of an iliopsoas transfer 2 years after the initial transfer operation. Free Gracilis Transfer In 1985, Dao and colleagues [10] provided a preliminary report describing the use of a free gracilis muscle transfer for the treatment of paralysis of the hip abductors. Details of patient selection

610

as to the age at operation and ambulatory status are absent. He described 12 cases, ten of whom had been followed 1–3 years. Two illustrative cases, a male of age 27 years at operation, and a male at age 22 years were reported with a result that the grafted gluteus medius showed obvious contractibility at grade 5 strength and electromyography showed normal muscle potential. At 1-year follow-up, the second patient showed improved hip abductor strength and demonstrated a stable gait. All the other patients had satisfactory cosmetic appearance, improved gait, increase in hip stability, and the ability to walk longer distances. Trendelenburg tests yielded negative results in eight cases out of ten cases. The patients could bear weight for longer duration than preoperatively. Muscle contraction could be detected when the affected limb was abducted or when the Trendelenburg test was performed. Two cases had muscle contractility at Grade 5, six cases at Grade 4, and two cases at Grade 2. Electromyograms demonstrated normal muscle potential in eight cases. No patient had complaints about the donor site and there was no functional impairment. Since Dao’s preliminary report, there have been a number of studies published concerning the use of a free gracilis muscle transfer into the upper extremity to improve weakness of elbow flexion due to poliomyelitis, but none reporting further on its use in the lower extremity. Until more data appear, the use of free muscle transfer for the treatment of treatment of weak hip abductors remains intriguing but not for general use.

 ip Extensor Paralysis H Erector spinae transfer to the greater trochanter can minimize the lurch associated with paralysis of the hip extensors. The technique as described by Ober [26] and modified by Barr [1], is straightforward. Treatment is frequently less successful in the eyes of the patient than the surgeon. Patient selection is the major stumbling block. Too high an expectation on the part of the patient or the surgeon is frequently disappointed. An AP X-ray of the hips, preferably done standing, should be obtained to assure that the hip is stable in the joint before embarking on tendon transfer for hip extensor paralysis.

H. G. Watts et al.

Outcomes Following Tendon Transfers for Hip Abductor and Extensor Paralysis The only gait laboratory study focused on outcomes of tendon transfers for hip abductor and extensor paralysis in polio reported no decrease in energy consumption following these ­operations highlighted above. In several other studies [3, 15, 29], however, patients claimed that their endurance did improve. We remain unenthusiastic about the value of these tendon transfers and have limited them to very occasional, carefully selected, cases. “Although some studies claim an improvement in gait and endurance, we remain unenthusiastic about the value of these tendon transfers and have limited them to very occasional, carefully selected, cases”.

 ip Joint Instability H As in myelodysplasia, a dislocated hip may not be a significant impairment to function, but multiple surgical attempts to regain muscle balance resulting in hip stiffness can be the cause of much greater disability. Hip stiffness is much more of a problem in cultures where floor sitting is an important activity of daily living—the very cultures where polio is more common. Although there has been much concern about regaining hip joint stability after paralysis about the hip following polio, fortunately, the need for operative treatment is uncommon. It is important to clarify in one’s mind the reasons that give concern when managing a child with hip instability secondary to polio, and not to confuse them when assessing the need for treatment. Of the reasons to treat the unstable hip in polio, the major one is pain. Fortunately, this is very infrequent. Parsons and Sneddon noted that no child presented with pain [27]. While it is mentioned frequently as a concern, few of the articles concerning hip instability present evidence of pain; let alone assess their results in terms of the ability to manage it when present. A much more common issue is that of hip instability (due to subluxation or dislocation) causing an unsightly and tiring walking pattern. While limping is very commonly associ-

23  The Hip in Poliomyelitis

ated, the surgical correction of hip instability is often unsuccessful in improving function since it is very dependent on weaknesses involving the other joints of the ipsilateral limb as well as the contralateral lower limb, upper limbs, and spine. The least important issue is that of radiographic evidence of inadequate acetabular coverage. While moderate improvements in radiographic measurements have been attained, these have been in the face of significant complications and morbidity with little functional benefit.

611

Pre-operative plain radiographs of the hips taken in the standing position (where possible) are necessary and the most useful for planning purposes. Hip arthrograms were used by Weismann [38] before undertaking his Colonna arthroplasties, but are otherwise not essential. Modern 3-D reconstruction techniques may help, but the data from the literature preceded these imaging techniques. There are many options for operative management of the unstable hip in polio including: (1) the capsular arthroplasty of Colonna, (2) varus “The surgical correction of hip instability proximal femoral osteotomy, (3) pelvic supis often unsuccessful in improving funcport osteotomy, (4) salvage procedures such as tion since it is very dependent on muscle the shelf augmentation and Chiari osteotomies, weakness involving of the ipsilateral and (5) Salter innominate osteotomy, (6) Pemberton the contralateral lower limbs, upper limbs, osteotomy, (7) Steel [34] triple pelvic osteotomy, and spine, and not just the hip”. (8) Bernese periacetabular osteotomy, and (9) The treatment options include surgery to cor- arthrodesis. rect the anatomy of the bones which include: (1) Trying to sort out the efficacy of each option proximal femoral surgery [37] to alter the neck/ is extremely difficult. First, because the patients shaft angle, (2) pelvic surgery to redirect the ace- vary enormously in the variety of affected body tabulum, (3) a pelvic augmentation surgery (such parts and degree of weakness of the multiple as a shelf or Chiari procedure) to cover a non-­ muscles involved, and secondly, most of the concentric femoral head, or (4) hip arthrodesis. patients have already undergone an array of operHips that are unstable dynamically present ative combinations at many anatomic levels, so a bigger problem, the major cause being the the individual contribution of each procedure is imbalance of muscles around the hip. Altering muddied. the anatomy of the bones of hip joint alone without also altering the muscle imbalance cannot be Colonna Capsular Arthroplasty expected to help and will lead to a recurrence of In 1959, Weissman [38] reported on the results deformity. of six children, operated upon by Colonna’s [8] The report of Parsons and Seddon [27] pro- capsular arthroplasty. In this procedure, capsular vides a good overview of the problems when flaps are interposed between the femoral head they state that “the records of treatment of these and acetabulum to treat the dislocated hip. In patients are, in main, a catalogue of failure”. 1999, Weissman [39] reported on an additional The interrelationships that need to be consid- three children treated with this technique. All ered with respect to the treatment of hip instabil- nine children had poliomyelitis as infants under ity include: the age 3 years. All underwent the Colonna procedure between the ages of 3 and 8 years of age. • abduction contracture of the contralateral hip They were followed from 10 to 14  years post-­ • a severe leg length discrepancy with shorten- operatively. The technique was that described by ing of the opposite side Colonna, including the supracondylar osteotomy • inadequate lateral acetabular development performed 6 weeks later to rotate the limb back • pelvic obliquity from lumbar scoliosis result- into normal alignment. An adductor tenotomy was ing in inadequate lateral hip coverage performed concomitantly with the arthroplasty in • proximal femoral valgus due to growth distur- those patients who had any power in these musbances resulting from abductor weakness cles. It was not performed when the hip was flail.

612

No preliminary traction was used. Three children required a subsequent inter-­trochanteric femoral osteotomy after developing hip adduction contractures. The presence or absence of pain preor post-operatively was not commented upon. The results were gauged on improvement in gait. Out of the nine children, four had no improvement. The other five were reported to have a full range of motion, and their gait was considered improved. Their hip radiographs showed normal joint space with a contained head in four patients, and one other had a slightly enlarged acetabulum. There are no other data for the use of this procedure in children with dislocations of the hip due to poliomyelitis. We feel that the use of the Colonna procedure has little if any role in treating clinically relevant chronic dislocations of the hip in children with poliomyelitis. Varus Proximal Femoral Osteotomy Parsons and Seddon [27] reported 13 varus proximal femoral osteotomy operations which failed in 11. Furthermore, in 8 of the 13, the valgus deformity returned. Jones [17] reported on 11 cases, of whom six had recurrence and an unsatisfactory result. Lau [18] presented a rather mixed group with 64 operations in 39 patients. The age at operation was 6–38  years (mean 13.4  years). He reported results of “Good” (hip clinically stable; head coverage greater than 75%, pain free, good range of motion (ROM)) in 46%, “Satisfactory” (hip clinically stable, head coverage less than 75%, pain free; good ROM) in 24%, and “Unsatisfactory” in 18%. Six patients had pain post-operatively (15%) but it is unclear how many had pain pre-operatively. Pelvic Support Using Ilizarov’s Concepts Mahram [21] reported on the use of a pelvic support procedure using Ilizarov concepts and technique which involved a proximal femoral extreme valgus with extension as well as mid-shaft osteotomies to add varus for knee alignment and lengthening for limb shortening. But of the 20 patients (14–30 years old), only two patients had had poliomyelitis, and the ages at surgery were not provided. All the patients had complained

H. G. Watts et al.

pre-operatively of pain on walking, either in the thigh or low back. The mean time in the external fixation was 6.4 months (range 5–9 months). In one of the two patients with polio, knee stiffness proved to be a major complication. Shelf, Salter, Pemberton and Chiari Procedures There are no studies which specifically focus on the efficacy of these procedures in children with hip instability following poliomyelitis. The procedures are mentioned, or listed among those that were tried, but no specific recommendations are given. Parsons and Seddon [27] mention nine “traditional acetabuloplasties” and two Salter types. In their study, eight were followed adequately; six failed and two developed pain post-­op. Lau and colleagues list 39 patients who had had 64 operations (including Salter, Chari, Steel, shelf augmentation, and an unspecified posterior acetabuloplasty) giving little detail that can be used to decide which, if any operation was preferable. Periacetabular Osteotomies (Bernese or Steel) Sierra [32], and Lee [19] have reported on these procedures. Lee and colleagues reported on 62 patients who underwent Steel’s triple innominate osteotomy (some with concomitant transiliac leg lengthening) (Chap. 5) but only 12 were aged less than 21 years. Twenty-five of the 62 patients had other hip procedures (including Campbell; Mustard; external oblique transfers with or without a concomitant Ober procedure) and it is unclear when these were done in relation to the pre-op evaluation of instability. There were only three patients with dislocation and all three had open reductions and all became stable post-­ operatively. Of six children with subluxatable hips, two hips remained unchanged following surgery. There was no discussion as to pain relief. Sierra and colleagues reported on nine patients—all with pain—but only two were children (12 and 13.5 years). Both able to walk unlimited without support but with moderate pain. Three had prior hip surgeries which were not described. The seven adult patients included

23  The Hip in Poliomyelitis

had decreased pain and better X-ray appearance post-operatively but no real improvement in function. The authors stated: “in most cases, walking distance and the use of gait aids was highly dependent on the presence or absence of disability related to other joints”. While the enthusiasm for the Bernese periacetabular osteotomy appears to be growing for use in adolescents with other hip disorders, the evidence for its use in hip instability or pain caused by polio paralysis is scant.

613

“With the hip fused in flexion, the patient might also need an ankle fusion in slight plantar flexion to help stabilize the knee into extension for cases where quadriceps function is weak or absent”.

A walking spica was introduced at 3 months post-operatively, and the total time in plaster was 4–8 months. Sharp and colleagues [30] reported 16 hip fusions in children (ages 7 1/2–16 1/2 years) for gait abnormalities secondary to poliomyelitis but “While the enthusiasm for the Bernese no patient had subluxation or pain pre-­operatively. periacetabular osteotomy appears to be Their list of complications was impressive: 8 growing, the evidence for its use in hip of the 16 had fractures of the fused femur, and instability or pain caused by polio paralyone of the ipsilateral tibia. Three children had sis is scant”. pseudarthrosis and one developed slipped capital femoral epiphysis; one had limited knee motion Arthrodesis from prolonged immobilization and one underArthrodesis of the hip has usually been reserved went amputation for excessive shortening. Their for patients with hip instability and pain who criteria for patient selection were as follows: “… have failed previous operations. The literature is Some hip elevating power is necessary, sufficient scant. Hallock [13, 14] reported on 11 patients, to lift the paralyzed limb with its fused hip off all with hip subluxation or dislocation but only the ground. More important is a good contralatsix had pre-operative instability and pain. Post-­ eral gluteus medius to tilt down on that side and operatively, all were relieved of their pain. In elevate the opposite fused unit. With this, both this study, the patients underwent extra-articular (bilateral) abdominals are necessary to stabilize arthrodesis between the ages of 13 and 30 years. the trunk tilt and prevent a pseudo-antalgic type Hallock stated that “there were no post-operative of gait above the pelvis as the trunk swings over complications”; however two patients required the weak side for balance, to prevent falling to re-fusions and one underwent a subtrochanteric the strong side and also to aid in elevating and osteotomy for repositioning 5  years post-op. holding the pelvis where the gluteus medius initiThere were two patients with unstable knees ated it. Additionally, good abdominal strength is post-operatively. Shortening of 2.5–8 cm (mean needed to prevent scoliosis which would contra4  cm) was reported. The details of length of indicate the procedure. The optimum age for hip follow-­up are absent. There is no mention of the fusion is ten or fourteen years for social reasons possible effects of the hip fusion on back pain and to decrease additional limb shortening to a or ipsilateral knee instability over time. The posi- minimum.” tion the hip joint for arthrodesis of fixation was Stinchfield [35] reported on nine patients reported to be: 35° flexion and neutral abduction/ with polio related hip dislocations. The operaadduction except in females or where consider- tions were done from ages 13 to 31 years of able shortening was expected. For these excep- age (average of 22.5  years) but further detail tions, 10–20° of abduction was selected. were not given. “The primary indication for It should be noted that with the hip fused in surgery was instability rather than pain, flexion the patient might need an ankle fusion in although the pain was severe in three cases.” slight plantar flexion to help stabilize the knee Post-operatively, two had severe, five had modinto extension for cases where quadriceps func- erate and two had mild pain. Stinchfield sumtion is weak or absent. marized the results, stating that “Most of them

H. G. Watts et al.

614

were remarkably improved after operation and led moderately active lives, seven of the nine discarding crutches entirely.” Given the very limited literature, and the significant litany of complications, and the concern for the development of back pain in adulthood, one cannot be enthusiastic about recommending hip arthrodesis following poliomyelitis for anything but hip pain which is unresponsive to bracing or lesser operations, and even then, pain relief is not guaranteed. The use of a standard total hip replacement is not considered to have a role in the management of children and adolescents but may have a place in the treatment of adults with hip instability with pain.

Essential Surgical Techniques

Flexion and abduction contracture • Release of tensor fascia latae (proximal and distal; Ober-Yount procedure) • +/− release of rectus femoris, iliopsoas, sartorius as needed Hip joint instability • Restore muscle balance first to augment abductor weakness if possible • Correct coxa valga with varus proximal femoral osteotomy • Correct acetabular dysplasia with reconstructive or salvage techniques • Consider Colonna capsular arthroplasty for the painful, dislocated hip

Operative Pitfalls

• Lack of supportive hospital and rehabilitation resources required following major hip surgery in polio • Release of the hip capsule may lead to instability and stiffness worse more disabling than the initial deformity

• Not performing femoral shortening to relax tissues during correction of severe deformities (risk of neural stretch injury) • Setting the patient’s expectations too high as to functional improvement post-operatively • Treating the X-ray rather than the patient. Hip dislocations are most often asymptomatic and can be made worse by surgical intervention. • Trading a mobile, painless dislocated hip pre-operatively for a stiff, sometimes painful hip post-operatively. • Being heroic in the treatment of acetabular dysplasia. Complex acetabular reconstructive procedures have a significant risk of complications, often with little functional benefit.

Classic Papers Sharrard WJW.  Posterior iliopsoas transplantation in the treatment of paralytic dislocation of the hip. JBJS 1964;46B:426. Points out the importance of the clustering of the axons for a given muscle within the spinal cord, and why some muscles are more affected than others. Beasley WC.  Quantitative muscle testing: Principles and applications to research and clinical services. Archives of Phys Med & Rehab. June 1961;398–425. Quantifies the degree to which the manual muscle test markedly underestimates the degree of weakness of a muscle and exaggerates the possibility of a benefit of muscle transfer. Yount CC.  The role of the tensor fasciae femoris in certain deformities of the lower extremity. JBJS 1938;20:314. First paper that focuses on the important role that the tensor fascia femoris has of causing both flexion and abduction contractures at the hip and flexion, external rotation contractures at the hip.

23  The Hip in Poliomyelitis

Souter R.  A new operation for hip contractures in poliomyelitis, Boston Med Soc. J 1926;170:380. The first paper to focus on the release of flexion contractures at the hip. Ober FR: An operation for the relief of paralysis of the gluteus maximus muscle, JAMA 1927;88:1063. Provides a less extensive, yet as effective modification of Souter’s hip flexor release. Mustard WT. Iliopsoas transfer for weakness of the hip abductors; a preliminary report. JBJS 1952;24 A(3):647–50. Describes the redirection of the forces of the ileopsoas muscle to convert flexor power to abduction power. Barr JS: Poliomyelitic hip deformity and the erector spinae transplant, JAMA 1950;144:813. Together with Thomas’ paper (see below) describes one of the earliest attempts to use muscle transfer about the hip. Thomas LI, Thompson TC, and Straub LR: Transplantation of the external oblique muscle for abductor paralysis, JBJS 1950;32-­ A:207. Together with Barr’s paper (see above) describes one of the earliest attempts to use muscle transfer about the hip. Hallock H Arthrodesis of the hip for instability and pain in poliomyelitis JBJS 1950;32A:904–9. One of the few papers describing hip arthrodesis for instability, with or without pain.

Key Evidence Joseph B, and Watts H.  Polio revisited: reviving knowledge and skills to meet the challenge of resurgence. J Child Orthop 2015;9:325–338. This recent review article by the chapter authors collects the available evidence and their experience including the treatment of the hip in polio. Though not an ‘evidence-based’ study, it is essential reading for the surgeon uninitiated in the orthopedic management of polio.

615

Take Home Messages

• Problems for children and adolescents involving the hip following poliomyelitis are: –– Deformities resulting from contractures about the hip, • Very common, and easiest to manage –– Problems resulting from either weakness of muscles acting on the hip joint, • Much less common, and harder to manage –– Problems resulting from hip joint instability or dislocation. • Very uncommon and very difficult to manage • The literature concerning treatment options in poliomyelitis as it affects the hip is scant and very old. The quality of the evidence, by current standards, is almost nil. • Problems of the hip must not be evaluated in isolation. The many weaknesses and deformities in the other parts of the limbs, including those in the upper extremities, will affect decision-making. A thorough evaluation of the range of motion and muscle strengths is a sine qua non of the preoperative examination. • While release of contractures in children with poliomyelitis paralysis is very rewarding to the patient and the surgeon, the multiplicities of tendon transfers about the hip are seldom as rewarding.

616

References 1. Barr JS.  Poliomyelitic hip deformity and the erector spinae transplant. JAMA. 1950;144(10):813–7. 2. Beasley WC.  Quantitative muscle testing: principles and applications to research and clinical services. Arch Phys Med Rehab. 1961;42:398–425. 3. Cabaud HE, Weston GW, Connelly S.  Tendon transfers in the paralytic Hip. JBJS. 1979;61A(7):1035–41. 4. Campbell W. Transference of the crest of the ilium for a flexion contracture of the hip. SRN Med J. 1923;16:289. 5. Carreri G. A reassessment of the surgical treatment of severe contractures of the hip joint following poliomyelitis. Postgrad Med J. 1961;37(201):206. 6. Centers for Disease Control and Prevention. Progress toward poliomyelitis eradication—India, January 2009–October 2010. MMWR Morb Mortal Wkly Rep. 2010;59:1581–5. 7. Centers for Disease Control. AFM in the United States. https://www.cdc.gov/acute-flaccid-myelitis/ afm-surveillance.html 8. Colonna PC.  Capsular arthroplasty for congenital dislocation of the hip. A two-stage procedure. JBJS. 1953;35-A:179–97. 9. Crenshaw AH.  Anterior poliomyelitis. In: Azar FM, Terry Canale S, Beaty JH, editors. Campbell’s operative orthopaedics. 4th ed. St. Louis, MO: Mosby; 1963. 10. Dao QS, Zhang YF, Yang QM, Guo BF.  Free gracilis transfer in the treatment of gluteus medius paralysis after poliomyelitis. J Reconstr Microsurg. 1985;1(3):241–4. 11. Elzinga ER, Key JA. Paralytic dislocation of the hip in poliomyelitis, vol. 14; 1932. p. 867–81. 12. Global Polio Eradication Initiative. http://www.polioeradication.org/. Accessed 21 Mar 2012. 13. Hallock H.  Surgical stabilization of dislocated paralytic hips: end result study. Surg Gynecol Obstet. 1942;75:721–8. 14. Hallock H. Arthrodesis of the hip for instability and pain in poliomyelitis. JBJS. 1950;32A:904–9. 15. Hammesfahr R, Topple S, Yoo K, Whitesides T, Paulin AM. Abductor paralysis and the role of the external oblique transfer. Orthopedics. 1983;6(3):315–21. 16. Huckstep RL.  Poliomyelitis: a guide for developing countries. Edinburgh: Churchill Livingston; 1975. 17. Jones BS. Upper femoral osteotomy in the treatment of paralytic subluxation of the hip due to poliomyelitis. SA Med J. 1969;43:1187–92. 18. Lau JH, Parker JC, Hsu LC, Leong JC.  Paralytic hip instability in poliomyelitis. J Bone Joint Surg. 1986;68(4):528–33. 19. Lee DY, Choi IH, Chung CY, Ahn JH, Steel HH. Triple innominate osteotomy for hip stabilization and transiliac leg lengthening after polio myelitis. JBJS. 1993;75B(6):858–64. 20. Legg AT. Tensor fasciae femoris transplantation in cases of weakened gluteus medius. JAMA. 1923;80:242. 21. Mahram MA, ElGebeilly MA, Ghaly NAM, Thakeb MF, Hefny HM.  Pelvic support osteotomy by Ilizarov’s concept: is it a valuable option in managing neglected hip problems in adolescents and

H. G. Watts et al. young adults? Stratagies Trauma Limb Reconstr. 2011;6:13–20. 22. Miller GR, Irwin CE. Paralytic dislocation of the hip. Duke Correspondence Club Letter, 18948, 6 Sept. 23. Mustard WT.  Iliopsoas transfer for weakness of the hip abductors; a preliminary report. JBJS. 1952;34 A(3):647–50. 24. Mustard WT. A follow-up study of iliopsoas transfer for hip instability. JBJS. 1959;41-B(2):289–98. 25. National Polio Surveillance Project. A Government of India–WHO collaboration. Accessed 8 Jun 2012. http://www.npspindia.org/ 26. Ober FR.  An operation for the relief of paralysis of the gluteus maximus muscle. JAMA. 1927;88:1063. 27. Parsons DW, Seddon HJ. The results of operations for disorders of the hip caused by poliomyelitis. JBJS. 1968;50B(2):266–73. 28. Sever JW. The causes and treatment of dislocations of the hip joint. Boston Med Surg J. 1911;165:313–23. 29. Shahcherghi GH, Mahzad J.  Abductor paralysis and external oblique transfer. J Pediatr Orthop. 2000;20(3):380–2. 30. Sharp N, Guhl JF, Sorenson RI, Voshell AF.  Hip fusion in poliomyelitis. JBJS. 1964;46A(1):121–33. 31. Sharrard WJW. Posterior iliopsoas transplantation in the treatment of paralytic dislocation of the hip. JBJS. 1964;46B:426. 32. Sierra RJ, Schoeniger RS, Millis M, Ganz R.  Periacetabular osteomy for containment of the nonarthritic dysplastic hip secondary to poliomyelitis. JBJS. 2012;92:2917–23. 33. Souter R.  A new operation for hip contractures in poliomyelitis. Boston Med Soc J. 1926;170:380. 34. Steel HH.  Triple osteotomy of the innominate bone CORR no. 122. J Bone Joint Surg. 1977;122:116–27. 35. Stinchfield FE, Cavallaro WU. Arthrodesis of the hip joint. A follow-up study. JBJS. 1950;32-A:48–58. 36. Thomas LI, Thompson TC, Straub LR. Transplantation of the external oblique muscle for abductor paralysis. JBJS. 1950;32-A:207. 37. Weissman SL, Yotok G, Khermosh O. Intertrochanteric osteotomy in fixed paralytic obliquity of the pelvis. JBJS. 1961;43A(8):1135–54. 38. Weissman SL. Capsular arthroplasty in paralytic dislocation of the hip. JBJS. 1959;41-A:429–39. 39. Weissman SL. Follow-up notes on articles previously published in the journal. JBJS. 1999;51A(5):1015–7. 40. Weston GW.  Tendon transfers about the foot, ankle, and hip in the paralyzed lower extremity. JBJS. 1963;47A(7):1430–43. 41. Whitman A. Observations on the correction of deformities of long standing. JAMA. 1923;80:18. 42. World Health Organization. Immunization and vaccine development. Polio eradication, VPD surveillance. Geneva: WHO; 2011.. http://www.searo.who. int/en/Section1226/ShowArchive.asp?Year=2011. Accessed 21 Mar 2012. 43. Yount CC.  The role of the tensor fasciae femoris in certain deformities of the lower extremity. J Bone Joint Surg. 1938;20:314. 44. Sharrard WJW. The distribution of the permanent paralysis in the lower limb in poliomyelitis. JBJS. 1955;37B:540.

Part VIII Syndromes and Skeletal Dysplasias

The Hip in Rett Syndrome

24

Deborah M. Eastwood

Introduction Rett syndrome is a rare neurological condition first described in 1966 by an Austrian physician Andreas Rett [1]. He noted that in some girls there was stagnation of normal development and then relatively acute and often severe regression with loss of motor and social skills. The condition showed similarities to those affected by cerebral palsy and with autism but specific differences were identified. People with Rett syndrome have profound and multiple physical and communication disabilities and are totally reliant on others for support throughout their lives [2, 3]. The neuromuscular components of the syndrome develop over time. A low tone and a wide based, dyspraxic gait pattern are apparent early on but give way, by late childhood, to lower limb spasticity and increased tone in many but not all patients [2, 3]. In many neuromuscular conditions, hip problems are common. For instance in both low tone and high tone forms of cerebral palsy, there is often hip subluxation and/or dislocation. In Rett syndrome, there is very little published information about the incidence and natural history of hip dysplasia. In

Naidu’s [3] descriptive paper of 70 cases ranging from age 2.5–34.5 years at review, two cases of neonatal hip dislocation are noted but there is no mention of whether or not treatment was given. The paper makes a comment that during childhood ‘a few developed elbow and knee contractures with some eventually developing hip dislocation’ but there are no details of numbers and/or treatment. The final paragraphs describe adult patients with Rett Syndrome and in 8/9 cases, ambulation was normal or possible with support. Other studies [4–6] have documented abnormal clinical and radiological features of associated dysplastic hip development in some patients with Rett Syndrome but overall there is a lack of evidence on incidence and natural history or functional difficulties, if any, associated with the dysplasia. No treatment guidelines are available for the surgical management of hip problems should they be identified. “Rett syndrome involves stagnation of normal development fol­ lowed by acute and often severe regression.”

Pathophysiology D. M. Eastwood (*) Great Ormond St. Hospital for Children, London, UK The Catterall Unit, The Royal National Orthopaedic Hospital, Stanmore, Middlesex, UK © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_24

Almost all cases of Rett syndrome are caused by one of many defined mutations in the methyl-­ CpG-­binding protein 2 gene (MeCP2) on the X 619

D. M. Eastwood

620

chromosome and therefore the vast majority of affected patients are girls [7, 8]. The MeCP2 protein is essential in regulating gene expression during development particularly in the central nervous system. Some individual mutations are associated with specific functional outcomes including hand use and/or ambulation. The neurological deterioration may lead to a progressive loss of ambulation often associated with spasticity and muscle imbalance leading to joint contractures and deformity (Figs.  24.1, 24.2, and 24.3). A high incidence of fractures was noted in one study: with ten fractures in five patients, 14% of the cohort [5] (Fig. 24.4). The more severe/major the chromosomal deletion, the more severe the clinical picture and the greater the likelihood of the child developing associated problems such as respiratory difficulties or a scoliosis. Scoliosis is present in between 35% and 93% [3, 5, 6, 9] of cohorts assessed depending on whether clinical [6] or radiographic [5] criteria are used. There is no evidence that this is associated with hip displacement. Although joint contractures have been identified in several cohort studies [3–5], it remains unclear how often the hip is involved. Coxa valga is common [5] Fig. 24.1  Time line for the onset and progression of the clinical features of typical Rett Syndrome (from Chahrour M, Zoghbi HY. 2007. The story of Rett syndrome: from clinic to neurobiology. Neuron 56:422–37)

YEARS

0.5

Fig. 24.2  Thirteen year old girl with scoliosis (in a brace) and significant pelvic obliquity. She is already non-ambulant

2

1

3

4

5

10

20

>20

Normal development Developmental stagnation Microcephaly Growth arrest Hypotonic

Rapid regression Autistic features Loss of hand skills, speech, and social interaction Hand stereotypies Mental retardation Motor abnormalties Seizures Respiratory abnormalities Stationary stage Scoliosis Autonomic dysfunction Anxiety

Late motor deterioration Decrease/loss of mobility

Parkinsonian features

24  The Hip in Rett Syndrome

621

described for other neuromuscular conditions, particularly when lower limb spasticity is present (see Chap. 20).

Natural History

Fig. 24.3  AP pelvic radiograph of an 11y old girl: note the coxa valga but overall good hip development. She is chronically constipated, but comfortable and mobile

Fig. 24.4  AP pelvic radiograph of a 12y old girl demonstrating pelvic obliquity, left hip migration and a healed right femoral shaft fracture. The child was pain free and mobile

(Fig. 24.3) but acetabular dysplasia (as assessed by the acetabular index (AI)) less so [6], although the same study noted a lower centre-edge (CE) angle in cases of RS compared to controls: perhaps predicting greater potential for hip instability. Overall, the pathophysiology of hip displacement in Rett syndrome probably mirrors that

Females with minor changes to the MeCP2 gene may be healthy and there have also been case reports of boys with minor changes leading a relatively normal life and life span. Males with more severe genetic changes die in the neonatal/infancy period. Those with an additional X chromosome such as patients with Klinefelter syndrome may be less severely affected. Patients are of short stature and there is no adolescent growth spurt: by age 18, 70–80% of one study group were below the second centile for weight and height [10]. The severity of this growth failure in adolescence is associated with scoliosis, seizures and functional gross motor, fine motor and nonverbal language deficits. In the affected female, life expectancy is not well documented: death is usually due to aspiration pneumonia or seizure associated and the perception is that death is likely in the third decade by which time many women are not independently mobile [3, 9]. However, a recent Australian paper shows likelihood of survival of 71% at 25  years and 59% at 37  years and 61% were mobile with/without assistance; 40% of patients had undergone surgery for a scoliosis [11]. The incidence of scoliosis is known to be high in this patient group but the rate of surgical intervention is unclear: Downs et al. [9] reported that scoliosis surgery was performed in 98 of 140 (70%) cases of severe scoliosis (defined by the ‘need for surgery’ or a Cobb angle >40°). This study [9] suggested an improvement in life expectancy and quality of life in terms of respiratory health with surgery—with the caveat that case selection was important (Fig. 24.5). Unfortunately, studies [9, 11] reporting on the indications for and outcomes of spinal fusion for scoliosis fail to mention the presence/absence of concomitant hip problems.

D. M. Eastwood

622

Fig. 24.6  AP pelvic radiograph in a 14 year old girl. The spine has been fused to L5. The child stands for transfers only: she was ambulant prior to surgery. Her x-ray suggests a flexion deformity of the left hip

showed no difference between the AI of Rett patients vs age matched controls but the CE angle was lower in Rett cases. They postulated that patients who were wheelchair based with increased adductor tone would have a higher likelihood of developing hip problems but their paper acknowledged that no such patient was identified in their cohort.

Epidemiology

Fig. 24.5  AP spinal view showing fusion to the sacrum for scoliosis. The child is wheelchair based but with improved chest function following surgery

Loder et al. [5] documented joint contractures in 36% of their cohort of 36 patients but only six hips in four patients (11% of the cohort) had contractures; none underwent surgical intervention (Fig. 24.6). Hennessey and Haas [6] reviewed 16 patients but little evidence was presented relating to progressive hip dysplasia or subluxation. Their study

Rett syndrome is a rare X-linked genetic syndrome which therefore predominantly affects girls. The incidence is difficult to determine but it probably afflicts 1:10–12,000 girls. Almost all cases are due to a spontaneous mutation. The phenotype varies depending on the degree of gene expression. If the condition affects a male, there is often genetic mosaicism. The incidence in males is probably 1:40,000. “The more severe the genotype, the more severe the phenotype”

Clinical Presentation The clinical picture is often described in four stages (Table 24.1). The classic presentation is at the regression stage (Stage 2) where parents

24  The Hip in Rett Syndrome Table 24.1  Four clinical stages of Rett syndrome Stage 1 Early signs

Description Hypotonia, difficulty feeding, abnormal ‘jerky’ movements, difficulty with mobility, lack of interest in toys 2 Regression Loss of the ability to use hands purposefully with repetitive movements obvious Periods of distress, social withdrawal, unsteadiness when walking Difficulty eating, breath holding 3 Plateau May be improvement in aspects of behavior with less irritability Development of seizures, teeth grinding, breath holding and cardiac arrhythmias 4 Deterioration (in Development of a scoliosis, joint movement) contractures and spasticity, loss of the ability to walk

notice the loss of acquired motor skills at around the age of 1–4  years. In hindsight, subtle early (Stage 1) signs have been present since birth. Characteristic repetitive movements, particularly of the hands and upper limbs become apparent such as hand clapping or wringing. This is often associated with behavioural disturbances similar to children with autism. Musculoskeletal deformity, sufficiently severe to require consideration of orthopedic surgery, is a significant problem in Rett syndrome and usually presents in stage 3/4. By then, difficulties with walking are common and this may or may not be associated with the presence of joint contractures. Loder et al. [5] identified equinus contractures in 25% of patients and these as well as behavioural problems can limit walking ability. Hip symptoms, themselves, appear to be rare (or rarely documented) but clinical signs associated with muscle and joint contracture are present, as in children with other neuromuscular conditions. As they are rarely commented on, it may be that hip flexion and adduction contractures are usually not as marked as in children with GMFCS IV/V spastic bilateral cerebral palsy and patients with non-spastic neurological deterioration would not necessarily show this type of contracture. Loder et al. [5] noted one

623

of four patients with a hip contracture was in fixed abduction and extension. Whilst, as mentioned previously, hip instability is a potential cause for concern, there is little literature to support this idea. In a recent study [4] of 31 patients, at a mean age of 15 years 6 months, 24 (77%) were wheelchair based for all activities. Eight patients (26%) had undergone surgery for a hip migration percentage greater than 30% (four patients) or a dislocated hip (four patients). All were wheelchair based at time of review but the study suggested that regular radiographic review and early surgery should be considered for the prevention of a painful dislocation and increased pelvic obliquity. For a painful hip dislocation, reconstructive or salvage surgery is suggested (see Chap. 20). The same cohort [4] had a high incidence of scoliosis and 35% had undergone a spinal fusion. The diagnosis of Rett Syndrome is often made on observation of the child and their clinical signs and by exclusion of other differential diagnoses such as cerebral palsy. Diagnostic criteria have been established [12] for both typical and atypical forms of the syndrome (Table 24.2): the diagnosis of the typical case requires a period of regression followed by recovery or stabilization, all main criteria and all exclusion criteria. Atypical Rett syndrome requires the same period of regression followed by recovery or stabilization and then two of four main criteria and 5/11 supportive criteria.

Essential Clinical Tests

• Complete neuromuscular examination –– Diagnosis by exclusion of other neurodevelopmental conditions • Specific examination of the hip range of movements, muscle tone and contractures • Assessment of intellectual difficulties/ motivation • Assessment of associated cardiorespiratory morbidity and seizure status • Consultation with a geneticist

D. M. Eastwood

624 Table 24.2  The diagnostic criteria for typical and atypical Rett Syndrome [12] Main criteria 1. Partial or complete loss of acquired purposeful hand skills

2. Partial or complete loss of acquired spoken language 3. Gait abnormalities; impaired (dyspraxic) or absence of ability 4. Stereotypic hand movements such as hand wringing/squeezing, clapping/tapping, mouthing and washing/rubbing automatisms

Supportive criteria for Exclusion criteria for typical RS atypical RS   1. Breathing disturbances 1. Brain injury secondary to when awake trauma (peri- or postnatally) neurometabolic disease or severe infection that causes neurological problems 2. Grossly abnormal psychomotor   2. Bruxism when awake development in first 6 months of life   3. Impaired sleep pattern   4. Abnormal muscle tone

  5. Peripheral vasomotor disturbances   6. Scoliosis/kyphosis   7. Growth retardation   8. Small, cold hands and feet   9. Inappropriate laughing/ screaming spells 10. Diminished response to pain 11. Intense eye communication—“eye pointing”

Imaging

• AP Pelvis –– Migration percentage –– NSA (neck shaft angle) –– AI (acetabular index) • AP/Lateral Spine –– Scoliosis is common

The essential imaging is the same as for children with other forms of neuromuscular hip disease and similar measurements are undertaken (see Chap. 20). Surveillance of hip dysplasia/displacement seems sensible but there is no good quality evidence that progressive displacement occurs. Although attempts have been made to equate the pathophysiology in Rett Syndrome with that occurring in other neuromuscular conditions, this may not be appropriate.

Non-operative Management The pathophysiology of Rett syndrome, which encompasses altered muscle tone and the development of muscle and joint contractures, suggests that conservative treatment with physiotherapy, splinting and posture control programmes would be useful—but there is no evidence for their use or benefit. For overall quality of life, it is probably most important that the child is kept mobile and able to walk (with/without assistance). In many surgeons’ opinion this outcome may be more desirable than having a radiograph showing a well located hip. Mobility is limited by issues which include perceptual difficulties and spatial disorientation leading to balance problems. Physiotherapy programmes may be hampered by the child’s behavioural difficulties and hydrotherapy may thus have a proportionately more important role in helping maintain range of movement and muscle strength.

625

24  The Hip in Rett Syndrome

Fig. 24.7  Oblique view of a foot in a 12 year old girl, non ambulant for 12 m, showing gross osteoporosis

Feeding difficulties may exacerbate the tendency for poor bone health and the associated increased fracture risk: this can be minimized by considering the medical management of osteoporosis whilst ensuring an adequate dietary intake and promoting weight bearing exercise (Fig. 24.7). Essential Non-operative Methods of Management

• Maintain mobility –– Strategies to address abnormal behaviours –– Motivation to walk/move • Physiotherapy programmes –– Stretching, splinting, positioning • Maintenance of bone health.

Pitfalls of Conservative Management

• Behavioural difficulties may limit the success of physiotherapy/postural programmes.

Operative Management In many centres, the operative management of hip displacement in Rett syndrome has followed the principles applied to hip displacement in total-body-involvement CP.  This extrapolation of a surgical technique from one neuromuscular condition to another may not be justified, particu-

larly when the natural history of the condition is not well established. Figure 24.8 shows a case in which surgery analogous to that performed in cases of cerebral palsy associated hip dysplasia was performed (adductor release, proximal femoral varus derotation and Dega osteotomy): at five years follow-­up, the hip remained in joint. In this case the scoliosis (also followed radiologically) was mild, and correctable. In the management of cerebral palsy associated hip subluxation and scoliosis there are arguments as to which element should be addressed first. In Rett Syndrome, severe scoliosis is associated with loss of walking ability but there is no evidence that surgery reverses this: there is a perception that on occasion, surgical intervention has been associated with the loss of walking ability due to motivation/ behavioural issues. This factor must be considered if hip surgery is contemplated. Release of ankle contractures to facilitate a plantigrade foot position and stability in stance may be of theoretical benefit but many of these children who do walk, do so by using their tone and an equinus posture (with a wide based stance): again a cautious approach to surgery is advised.

Essentials of Operative Management

• Consider if it is really indicated • Release of muscle contractures • Reduction of femoral head into the acetabulum • Femoral shortening, varus and derotation osteotomy –– In keeping with degree of femoral deformity –– Stable fixation • Acetabuloplasty as required.

Pitfalls of Operative Management

• Remember to consider the ambulatory status/prognosis for your patient before performing soft tissue releases • Intellectual difficulties must be considered when planning your rehabilitation programme.

D. M. Eastwood

626

a

b

d c

e

Fig. 24.8  Patient with Rett syndrome (case, figures and legend courtesy of Dr JS Huntley, Sidra Medicine, Doha) (GMFCS 5—equivalent, totally dependent), asymmetric adductor contracture and right sided spasticity, bilateral coxa valga, and asymmetric dysplasia. Progressive increase in migration percentage from 4 years 4 months to 4 years 7 months (a, b). At 5 years 7 months (c), the right Reimers Migration Percentage was 40%, with asymmetric

f

acetabular indices and centre-edge angles. At the age of 6 years, she underwent a right adductor release, femoral varus derotation and Dega pelvic osteotomy (d), with 6 week follow-up views immediately before spica removal (e). Follow-up films (f-i; metalwork removed at age 7  years) with accompanying spinal films are given until age 11  years and 7  months, (5  years 7  months after surgery)

24  The Hip in Rett Syndrome

g

h

i

Fig. 24.8 (continued)

627

D. M. Eastwood

628

“The surgical management of hip dysplasia in Rett Syndrome relies on the extrapolation of principles defined and used for other neuromuscular conditions: this may not be appropriate”

ogy. It emphasizes the lack of good quality clinical data in some areas (such as musculoskeletal disability).

Take Home Messages

Classic Papers Hagberg B, Aicardi J, Dias K, Ramos O. Ann Neurol. 1983 Oct;14(4):471–9. A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Rett’s syndrome: report of 35 cases. Thirty-five girls were described with a uniform and striking progressive encephalopathy. After normal early general and psychomotor development (up to 7–18  months), developmental stagnation occurred, followed by rapid deterioration of higher brain functions. The condition is similar to a virtually overlooked syndrome described by Rett in the German literature. The exclusive involvement of females, correlating with findings in family data analyses, suggests a dominant mutation on one X chromosome.

Key Evidence Tay G, Graham H, Graham HK, Leonard H, Reddihough D, Baikie G. 2010. Hip displacement and scoliosis in Rett syndrome - screening is required. Dev Med Child Neurol 2010;52:93–8. A review of 31 cases at mean age 15.5  years noted that 24 patients were wheelchair based; eight patients had required hip surgery for subluxation or dislocation and 11 cases had undergone scoliosis surgery. A surveillance programme with early surgery was proposed. Zoghbi HY. Rett Syndrome and the Ongoing Legacy of Close Clinical Observation. Cell. 2016 Oct 6;167(2):293–297. Fifty years after the description of Rett syndrome, this paper highlights the progress made in understanding the condition and the ramifications that this research has had on the fields of genetics and neurobiol-

• Uncommon neurodevelopmental condition • X-linked disorder therefore affecting girls predominantly • Associated with regression (loss) of motor skills • Development of altered muscle tone: spasticity leads to contractures • Behavioural difficulties also limit mobility and motivation • Scoliosis surgery improves life expectancy and respiratory health • Hip surgery follows principles devised for hip displacement in cerebral palsy although there is no evidence that it is beneficial in Rett syndrome

References 1. Rett A.  On an unusual brain atrophy syndrome in hyperammonemia in childhood. Wien Med Wochnschr. 1966;116:723–6. 2. Hagberg B, Aicardi J, Dias K, Ramos O. A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Rett’s syndrome: report of 35 cases. Ann Neurol. 1983;14:471–9. 3. Naidu S, Murphy M, Moser HW, Rett A.  Rett syndrome – natural history in 70 cases. Am J Med Genet. 1986;24:61–72. 4. Tay G, Graham H, Graham HK, Leonard H, Reddihough D, Baikie G. Hip displacement and scoliosis in Rett syndrome  – screening is required. Dev Med Child Neurol. 2010;52:93–8. 5. Loder RT, Lee CL, Richards BS. Orthopedic aspects of Rett syndrome: a multicenter review. J Pediatr Orthop. 1989;9:557–62. 6. Hennessy MJ, Haas RH.  The orthopaedic management of Rett syndrome. J Child Neurol. 1988;3: S43–8. 7. Chahrour M, Zoghbi HY.  The story of Rett syndrome: from clinic to neurobiology. Neuron. 2007;56: 422–37.

24  The Hip in Rett Syndrome 8. Zoghbi HY.  Rett syndrome and the ongoing legacy of close clinical observation. Cell. 2016;167: 293–7. 9. Downs J, Torode I, Wong K, Ellaway C, Elliott EJ, Izatt MT, Askin GN, Mcphee BI, Cundy P, Leonard H. Rett syndrome spinal fusion group. Surgical fusion of early onset severe scoliosis increases survival in Rett syndrome: a cohort study. Dev Med Child Neurol. 2016;58:632–8. 10. Tarquinio DC, Motil KJ, Hou W, Lee HS, Glaze DG, Skinner SA, Neul JL, Annese F, McNair L, Barrish JO, Geerts SP, Lane JB, Percy AK.  Growth failure

629 and outcome in Rett syndrome: specific growth references. Neurology. 2012;79:1653–61. 11. Anderson A, Wong K, Jacoby P, Downs J, Leonard H.  Twenty years of surveillance in Rett syndrome: what does this tell us? Orphanet J Rare Dis. 2014;9:87. 12. Neul JL, Kaufmann WE, Glaze DG, Christodoulou J, Clarke AJ, Bahi-Buisson N, Leonard H, Bailey ME, Schanen NC, Zappella M, Renieri A, Huppke P, Percy AK, Consortium RS.  Rett syndrome: revised diagnostic criteria and nomenclature. Ann Neurol. 2010;68:944–50.

Hip Problems in Children with Trisomy 21

25

Matthew Lea, Sattar Alshryda, and John Wedge

Introduction Trisomy 21 (T21), commonly known as Down Syndrome, is one of the most common and best known chromosomal disorders in humans. It is caused by an extra copy of chromosome number 21 (trisomy of chromosome number 21). This results in various abnormalities involving nearly every part of human body, including learning disabilities, cardiac anomalies, musculoskeletal problems, thyroid disorders, immunological and haematological disorders. Not all patients show every disorder and there is variation in the severity of these features [1]. Hip problems in children with T21 occur in about 5–8%, the most common type being dysplasia. Ligamentous laxity, hypotonia, and joint hypermobility are thought to be primary causes. It is observed that the genes that encode for type VI collagen are located on chromosome 21, and are believed to be responsible for the ligament laxity [2]. M. Lea Royal Manchester Children Hospital, Manchester, UK e-mail: [email protected] S. Alshryda (*) Clinical Director of Paediatric Trauma and Orthopaedic Surgery, Royal Manchester Children Hospital, Manchester University NHS Foundation Trust, Manchester, UK e-mail: [email protected] J. Wedge Hospital for Sick Children (Sickkids®), Toronto, ON, Canada e-mail: [email protected] © Springer Nature Switzerland AG 2019 S. Alshryda et al. (eds.), The Pediatric and Adolescent Hip, https://doi.org/10.1007/978-3-030-12003-0_25

Pathophysiology The stability of a joint depends on several static and dynamic elements which can be divided into three main anatomic groups: 1. Bones (size, shape, orientation of articular surfaces) 2. Ligaments and joint capsules 3. Muscles around the joint The hip joint is a ball and socket synovial joint, between the pelvic acetabulum and the head of the femur. As opposed to the gleno-­ humeral joint, the hip is designed for stability and weight-bearing rather than for an extended range of motion. The acetabular socket is larger than the femoral head and both are oriented to provide optimum range of motion before impingement or dislocation. The socket is further deepened by the labrum. The joint capsule is thick and strong with circular and longitudinal fibres. The hip joint is further stabilized by two sets of strong ligaments: intra-capsular and extra-capsular. The ligamentum teres is the only intra-capsular ligament. There are three extracapsular ligaments: iliofemoral, pubofemoral and ischiofemoral. Numerous large and small muscles cover and further stabilize the hip joint (see Chap. 2). Several patho-anatomical features have been described in hip dysplasia in children with T21. Hypotonia, ligamentous laxity, and ­hypermobility 631

632

are thought to be the main contributing factors for dysplasia and subsequent dislocation [3, 4]. This is supported by the fact that the genes that encode for type VI collagen are located on chromosome 21, and are believed to be responsible for laxity [2]. These 3 factors allow for continuous pathologic “shucking” of the proximal femur within the acetabulum [5] which gradually leads to structural changes that can compromise hip joint stability (Fig.  25.1). In the process, the hip capsule becomes thin and attenuated [6] (Fig. 25.2). “Hypotonia, ligamentous laxity, and hypermobility are thought to be the main contributing factors for dysplasia and subsequent dislocation in children with Trisomy 21”. Over time, with progressive hip subluxation, the acetabular centre-edge angle reduces and the tear drop widens [5, 7, 8]. Without treatment, the hip may become stiff, dislocated and painful, causing significant functional impairment [9, 10]. The acetabulum in children with T21 is often described as retroverted due to a deficient posterior wall. However, most published reports did not use the same criterion to measure acetabular version. In most cases, there is a deficiency in the posterior and superior aspects of the socket.

M. Lea et al.

Therefore, measuring acetabular version in the upper part usually reveals retroversion—in contrast to the middle or lower part of the socket, which are usually still anteverted (Figs. 25.3 and 25.4). A reference line tangential to the posterior points of both ischial tuberosities was used to measure anteversion in children with T21  in a study by Abousamra [11]. Anteversion was measured on CT by the angle formed between a line perpendicular to the reference and a line tangential to the acetabular edges. Acetabular anteversion was only moderately decreased (12°), and not severely decreased (2°) as reported previously [3, 12]. Interestingly, they found that acetabular anteversion in T21 syndrome was decreased to a similar extent irrespective of hip instability. Similarly, the patho-anatomical changes of the proximal femur in T21 remain in debate because published research did not use validated and reliable radiographic outcome measures. Measuring neck shaft angle (NSA) on plain radiographs is not accurate. Persistent femoral anteversion can be misinterpreted as an increase in NSA (apparent coxa valga) on anteroposterior X-ray [5, 8, 12]. Measuring femoral torsion accurately requires either CT Scan or MRI, rarely used in most published studies [13]. Shaw [12] studied 114 patients

Joint Laxity

Fig. 25.1  Abnormal movement of the hips in a child with T21. Hypotonia, ligamentous laxity, and hypermobility allow for repeated pathologic “shucking” of the proximal femur within the acetabulum

25  Hip Problems in Children with Trisomy 21

633

Fig. 25.2  Thin and attenuated hip capsule in a child with T21. Capsule is thin, attenuated and baggy. It can be easily pinched by forceps. It allows for repeated pathologic “shucking” of the proximal femur within the acetabulum

AP views

PA views

Fig. 25.3  Posterior and superior acetabular deficiency in children with Trisomy 21

634 Fig. 25.4 Acetabular version measurements in children with Trisomy 21. Upper part of the acetabulum: Right retroverted = 30°; Left retroverted = 10°. Middle part of the acetabulum: Right anteverted = 10°; Left anteverted = 5°. Lower part of the acetabulum: Right anteverted = 15°; Left anteverted = 10°

M. Lea et al.

Upper part of the acetabulum Right retroverted = 30 degrees Left retroverted = 10 degrees

Middle part of the acetabulum Right anteverted = 10 degrees Left anteverted = 5 degrees

Lower part of the acetabulum Right anteverted = 15 degrees Left anteverted = 10 degrees

with T21 and found that the proximal femur had a slight increase in femoral anteversion (33.5°: range 0–59°) and a normal neck-shaft angle (134°: range 115–148°). Findings from other studies include increases in proximal femoral valgus, extreme femoral anteversion, femoral retroversion or a combination of these [5, 8, 12]. Figure 25.5 shows the relationship between femoral torsion and neck shaft angle. A recent CT study by Bulat and colleagues [14] investigated the morphology of the acetabulum and femur in patients with T21 (42 patients) and compared measurements of the hips with those of age and sex matched controls. They found a more retroverted acetabulum in patients with T21 (mean acetabular version as measured at the level of the centers of the femoral heads was 7.8°  ±  5.1° compared with 14.0°  ±  4.5°; p 15 years): if untreated, the hip invariably develops a painful fixed dislocation by the time the patient is in their ‘late teens or early twenties’.

M. Lea et al.

636 Table 25.1  Prevalence of hip conditions in Trisomy 21 Study and year Lawhon [29] Aprin [30] Diamond [31] Bennet [10] Shaw [12] Makley [32] Stack [26] Cristofaro [33] Hresko [9] Total

Hips 162 946 265 220 114 110 90 129 130 2166

Dysplasia 9 12 12 10 8 NR 13 22 86

SCFE 2 1 2 1 1 3 3 NR 1 14

AVN 4

LCPD

OA 3

1

Note Average age for the hip dysplasia was 8 years

1 1 3 NR NR 0 8

NR NR 0 2

NR 13 6 22

One patient had unilateral hip protrusio

Y year, SCFE slipped capital femoral epiphysis, AVN avascular necrosis, OA osteoarthritis, NOF neck of femur fracture, LCPD Legg Calves Perthes disease, NR not reported

Epidemiology The incidence of Trisomy 21 (T21) varies depending on the parental age. The incidence is higher in children who are born to older parents. The risk of maternal age on having a child with T21 has been widely studied and the incidence has been found to be around 1  in 1500 at age 15–29  years, 1  in 800 at age 30–34  years, 1  in 270 at age 35–39  years, 1  in 100 at age 40–44 years, and 1 in 50 at age 45 years and older [6]. There has been a long-running debate about the association between paternal age and the incidence of T21. This has been fueled by the discovery that the extra chromosome in about 30% of cases of trisomy 21 is of paternal origin; however, the association (if it exists) is more complex than that with maternal age [18–21]. In the UK, pregnant women are routinely offered a screening test for T21 from 10 to 14 weeks of pregnancy. The test combines information from an ultrasound scan of the baby and a blood test from the mother. The test is a screening test with well documented false positive and negative rates. Findings should be interpreted carefully; however, if the results indicate increased risk of T21, particularly in women older than 35 years, it is reasonable to consider amniocentesis and chromosome analysis. Children with T21 can develop various musculoskeletal problems that may require medical and surgical attention. These include reduced height, atlanto-occipital and atlantoaxial instability, vertebral malformations of the cervical spine, patellar

dislocations, and hip dysplasia. Atlanto-­axial instability is the most serious, occurring in about 14%, and can cause spinal cord compression in about 2% (Fig. 25.6). Atlantoaxial instability must be considered before general anaesthaesia [22, 23]. Hip problems in children with T21 occur in about 5–8%, the most common type being subluxation. Although children with T21 may suffer from hip conditions that affect normal children such as Legg-Calve-Perthes Disease (LCPD), slipped capital femoral epiphysis (SCFE), developmental dysplasia of the hip (DDH) and j­ uvenile idiopathic arthritis, treatment principles of these conditions do not differ substantially from normal. In the small series of patients available in the medical literature, the reported outcomes and complications are not as favourable as seen in normal children. SCFE in children with T21 may occur as a distinct entity from classical (idiopathic) SCFE. Underlying causes such as endocrine disorders (particularly hypothyroidism) are common [24, 25]. In 1966, Stack and Peterson [26] reported on three patients with T21 who presented with SCFE and were treated with a Knowles pin. One patient developed avascular necrosis (AVN). Dietz and colleagues reported on eight SCFEs; five developed AVN. Bosch and colleagues [25] reported on their experience with 11 SCFEs. All patients were overweight and more than half were diagnosed with hypothyroidism. Three developed AVN and 6 of the 11 slips progressed despite the fixation. Additional surgery was necessary in seven hips. Although these

25  Hip Problems in Children with Trisomy 21

Extension view

637

Flexion view

Fig. 25.6  Atlanto-axial instability in a child with Trisomy 21. Flexion view on plain X-ray of the cervical spine shows that the atlanto-axial distance has increased in

comparison to the extension views (top pictures). CT Scan shows the widening of the atlanto-axial distance

are small studies, the message is clear and consistent with what we know about similar challenges in treating hip dysplasia in T21. These patients are challenging and the outcomes are less favourable as compared with classical SCFE. It is difficult to establish the incidence or prevalence of LCPD in children with T21 because most studies erroneously combined AVN and LCPD as a single group. However, studies [10, 27] that did not combine LCPD and AVN showed that LCPD is less common than SCFE

(Table  25.1). Only a single case series that addressed this cohort of patients was identified in the literature. Seven hips developed LCPD in six patients with T21 [28]. The authors’ conclusion was that LCPD in children with T21 does not differ from those without it with regards to its progress clinically or radiologically and should be treated similarly. The main focus of this chapter is on hip instability that develops in children with Trisomy 21, secondary to their unique pathophysiology.

M. Lea et al.

638

Clinical Presentation The diagnosis of a child with T21 and most of the associated features are usually well established before musculoskeletal issues become apparent. Nevertheless, a detailed history to explore the followings is essential: 1 . The current hip problem. 2. Other musculoskeletal problems, such as cervical spine instability and patellar instability. 3. General problems that may have an impact on the proposed treatment such medical comorbidities including cardiac, thyroid or haematological disorders. Hip dislocation usually appears between the ages of 2 and 10 years. This usually starts as a unilateral problem, and then progresses to become bilateral. It usually starts as an infrequent, painless (or occasionally painful) dislocation that reduces spontaneously. The frequency of dislocation, initially reducible, increases with time and progressively evolves into a fixed dislocation. Although hip dislocation in T21 is not usually congenital, children with T21 can have a congenital hip dislocation. This congenital dislocation does not respond as favorably (to either closed or open treatment) as congenitally dislocated hips of non-T21 children. Signs vary according to stage of the disease and whether one side or both sides are involved. The affected limb appears shorter, internally rotated, and adducted. The hip can be stable, dislocatable (a positive Barlow’s test), dislocated but reducible (a positive Ortolani test) or irreducible. The child may exhibit a Trendelenburg or short limb gait. Thorough neurological assessment is important to exclude spinal cord compression from upper cervical spine instability which affect the type of gait. About 13–14% of patients have radiographic evidence of atlantoaxial instability but only 1–2% of patients have symptoms that require treatment [6]. Symptoms include neck pain, limited neck mobility or head tilt, torticollis, difficulty walking, a change in gait pattern and other signs of spinal cord compression. The presence of atlantoaxial instability may cause spinal cord damage during general anaesthesia if appropriate precautions are not followed.

Instability of the patella is less common (0.4%) than hip and neck problems in non-­ institutionalized T21 patients. Conversely, it is much higher (8%) in institutionalized T21 patients [34]. Patellar instability is associated with excessive femoral anteversion (Fig.  25.7). The knee may give way, leading to frequent episodes of falling. This interfered with walking in some patients, but these numbers are generally very low. Almost all patients with patellofemoral instability adapted to the problem and were able to walk.

Essential Clinical Tests

1. General and associated conditions (a) Dry skin, weight gain, hair loss, depression (hypothyroidism) (b) Breathlessness, cyanosis, finger clubbing, chest surgical scars (cardiac involvement) (c) Overweight, snoring, difficulty awaking (sleep apnea) 2. Neck instability tests (a) Neck tenderness, limited neck mobility, head tilt, torticollis (b) Difficulty walking, change in gait pattern, loss of motor skills, incoordination, clumsiness (c) Spasticity, hyperreflexia, up-going Babinski test, clonus, weakness (d) Change in bowel and bladder function (e) Paraplegia, hemiplegia, quadriplegia 3. Hip instability tests (a) Leg length discrepancy (Galeazzi test) (b) Barlow and Ortolani tests 4. Patellofemoral instability tests (a) Increase in passive patellar translation (>2 quadrants of patellar = medial border of patella to lateral edge of trochlear groove) (b) Patellar apprehension (c) Q angle (d) J sign (e) Associated with patella alta (f) Frank patellar dislocation

25  Hip Problems in Children with Trisomy 21

Right Femoral Anteversion = 47 degrees

639

Left Femoral Anteversion = 51 degrees

Fig. 25.7  Patellar dislocation in a child with T21 and excessive femoral anteversion. Right femoral anteversion = 47° (34°+13°). Left femoral anteversion = 51° (32°+19°)

M. Lea et al.

640

Imaging Essential Imaging Tests and Measurements

1. Plain pelvic X-ray (a) Acetabular index (AI) (before the triradiate cartilage closure). The mean AI is 25° in the adult. Lateral CEA (on the false profile view): is >17°. These angles decrease with hip dysplasia. (e) Neck shaft angle (NSA) is very unreliable on plain radiograph. (f) Check for any acetabular or femoral head damage (Crater sign) (Fig. 25.8) 2. CT Scan with 3D reconstruction (a) The anatomical changes of the acetabulum and femoral head including version, inclination, joint congruity, focal damages (b) The rotational profile of the femur 3. MRI scan (Fig. 25.9) (a) The quality of the articular cartilage (b) Labrum tears (c) Impingement signs 4. Essential pre-operative imaging tests (a) Flexion and extension lateral cervical views to rule out upper cervical or subaxial cervical spine instability. (b) Echocardiogram to assess cardiac function given the high rate of associated congenital anomalies.

Non-operative Management Children with T21 pose serious challenges and a multi-disciplinary team approach is essential. These children often have other medical conditions that take priority over the musculoskeletal conditions. This can cause delay in the referral to the orthopaedic team. Moreover, some of these children are non-compliant and do not follow postoperative or physiotherapy instructions. The goal of treatment is to achieve a stable, painless and functional hip joint with minimal complications, lasting the patient’s lifetime. This is often unachievable, partly because all current interventions only treat the consequences of the primary pathology (hypotonia, laxity and hypermobility). For the same reason, non-operative management is not favorable. Early hip dislocations in T21 have been treated with closed reduction and prolonged immobilization in a hip spica cast (2–6 months), followed by an ambulatory abduction orthosis, full time for 4  months and then only at night for an additional 2–4 months, with little success [10, 30, 35]. Although some still recommend non-operative treatment for first dislocation or traumatic dislocation, the concern is that immobilization weakens the muscles even further and is counter-productive. “The goal of treatment is to achieve a stable, painless and functional hip joint with minimal complications, lasting the patient’s lifetime. This is often unachievable, partly because all current interventions only treat the consequences of the primary pathology (hypotonia, laxity and hypermobility)”

Operative Management To overcome the high failure rates in non-­ operative management of hip dislocation in T21, several operations have been advocated, on their own or in combination—with varying success rates. Surgical treatments include capsular plication, femoral osteotomy and acetabular proce-

25  Hip Problems in Children with Trisomy 21

641

Fig. 25.8  Crater sign. Crater sign (arrow) indicates significant damage to the articular cartilage caused by edge loading

dures (Salter; Dega, triple and, Chiari osteotomies) [7, 8, 11, 36–43]. The capsule is often thin, attenuated and baggy. ‘T-V’ capsulorraphy is performed to limit superior migration of the femoral head. Some authors recommend injecting bone material into the capsule or using fibrogenic mesh or sutures to enhance further fibrosis (Fig. 25.10). Capsulorrhaphy without correcting the valgus NSA is prone to failure even with supplementary attempts at inducing scarring of the capsule; the scarred capsule still stretches over time. Mercer Rang devised a procedure he termed as “arthroresis” where he placed autologous cancellous bone chips around the repaired capsule to decrease its pliability—but this too failed with the capsule still stretching over time. Femoral osteotomy is an essential part of the treatment. Preoperative examination under anaesthesia (EUA) and fluoroscopy is important

to establish the correct anatomy of the proximal femur and establish the femoral anteversion and neck shaft angle. It is expected that varisation improves stability. However, the role of derotation is not well established. Most experts recommend restoring femoral NSA to the lowest possible of the normal range (120–135°) or even lower (see evidence below) and anteversion to normal values of about 15° if it is excessive [41, 44]. Pelvic osteotomy is required if the acetabulum is abnormal/deficient, or if the hip remains unstable after capsular plication combined with a varus derotation femoral osteotomy. This is usually the case for surgically treated patients. The type of pelvic osteotomy is often determined mainly by the anatomical features of the acetabulum and the direction of instability. A capacious acetabulum is corrected with either a Dega osteotomy or a modified Pemberton osteotomy (Fig.  25.11). A normal sized congruent acetabulum can be corrected with a re-direc-

Fig. 25.9  MRI scan of both hips of a child with T21 and dislocated left hip. STIR views showed hip joint effusion, articular surface damage and subluxation

642 M. Lea et al.

25  Hip Problems in Children with Trisomy 21

643

Fig. 25.10  Suture mesh to limit capsule stretching (Manchester Technique)

Fig. 25.11  Dislocatable right hip in a child with Trisomy 21 treated with femoral varus derotational osteotomy and Dega pelvic osteotomy

644

M. Lea et al.

Fig. 25.12  A child with Trisomy 21 who had recurrent dislocation. Excision arthroplasty as a salvage procedure for recurrence after hip reconstruction surgery in children with Trisomy 21

tional pelvic osteotomy (Salter, triple or Ganz osteotomies). It is important that the acetabulum is re-directed to increase the coverage where the hip tends to dislocate. A smaller acetabulum is better corrected using a Chiari osteotomy or shelf acetabuloplasty. In a retrospective study comparing re-­ directional pelvic osteotomy (Bernese periacetabular or triple osteotomy, 25 patients) with patients who underwent varus femoral osteotomy  ±  incomplete acetabuloplasty (Dega osteotomy or shelf acetabuloplasty, ten patients), the re-directional pelvic osteotomy showed supe-

rior radiological and clinical results [43]. Bernese periacetabular osteotomy was performed for skeletally mature patients who had acetabular dysplasia whereas femoral osteotomy with or without acetabuloplasty was performed for younger patients with an open triradiate cartilage. Ninety-four percent (17/18) of hips that demonstrated gross instability preoperatively and were treated with re-directional osteotomy remained stable at the time of the latest followup (5.3 years). In contrast, to 50% (3/6) hips that were grossly unstable and were treated with femoral osteotomy with or without incomplete

25  Hip Problems in Children with Trisomy 21

645

Table 25.2  Total hip replacement outcome in patients with Trisomy 21 Harris Hip Score Preop Postop NR NRa

WOMAC Score Preop Postop NR NRa

NR

NR

19

59

82.4

NR

NR

86

41.4

85.2

b

b

b

b

0

NR

92

c

c

5.8

4

91.4

NR

NR

9

5

All revised because of aseptic loosing 38 around 3, 5, 6 and 17 years respectively 42 First: deep infection at 6 months. Second: recurrent dislocation at 1 year, The rest were revised because of aseptic loosing at 6, 10 and 16 years respectively NR

83

NR

NR

NR

NR

NR

(37.1) (87)

(30)

(72)

Study and year Kioschos 1999 [47] Solayar 2009 [48] Kosashvili 2010 [49] Taylor 2012 [50] Ries 1994 [51] Skoff 1987 [52] Amanatullah 2014 [53]

Hips (patients) 9 (6)

F/up (Y) 7.9

2(1)

1.5

Revision Reasons for revision 1 Acetabular lysis after 7 year and 4 months 0

9 (7)

9.9

2

1(1)

2

0

4 (4)

2.9

0

8 (3)

4.8

21(NR)

Gross 2013 [27]

26 (NR)

Boylan 2016 [46] Total (mean)

241 (NR) NR 321

NR

First: periprosethetic fracture 6 years after 40.8 surgery. Second: stem lysis at 16 years 27.5

(5.5) 12

NR not reported, F/up follow up, Y year, Preop: preoperative, Postop, postoperative At the latest follow up all patients were fully mobile and had no limitation of function due to their hips. None seemed to have any discomfort and all had good movement b New York State Functional score was used. Average preoperative and postoperative scores were 141 and 23 respectively c AAOS score was used and universally from Fair to Excellent a

acetabuloplasty. It is of note that this study is not a comparative study as the two groups were dissimilar and there was no clear protocol for the intervention choice. Some patients with recurrent or neglected dislocation of the hip can present with adverse sequelae which render the outcome of hip reconstruction less favourable (persistent pain, stiffness and loss of mobility). An alternative may be safer; (Fig.  25.12) excision arthroplasty, hip arthrodesis and total hip replacement (THR) have been utilized with varying success [40]. THR is increasingly performed in ambulatory and compliant patients with T21. In a systematic

review and patient level exploratory analysis (unpublished) of nine studies (321 patients with Trisomy 21) who underwent hip arthroplasty, the functional hip scores (Harris and WOMAC hip scores) improved substantially after hip replacement. The 5-year cumulative revision rate was 7.5%; twice as high as age matched control in the UK [45]. Medical and surgical complications were 3 times higher than matched controls [46] (Table 25.2). “Following total hip replacement, medical and surgical complications were 3 times higher in patients with trisomy 21 than matched controls reflecting the challenges of operating on these patients”

M. Lea et al.

646

Operative Pitfalls

• Do not delay surgery. Once the hip starts dislocating, it will never get better and gradual articular damage is inevitable. • Appropriately manage the expectation of parents, carers and children. • Treatment of hips should be as a multidisciplinary team. • Assess the patho-anatomy accurately. Clinical assessment, 3D imaging and examination under fluoroscopy screening are often necessary to adequately assess the anatomy of the hip. • Treat the patient and not just the hips. • More is usually better than less (within reasons). Capsule plication, femoral and pelvic osteotomies are better than these operations individually, more varus is better than less varus, 8 weeks in spica is better than 6 weeks in spica etc. We recommend that the NSA on intraoperative imaging with the foot in neutral alignment should be reduced to 100– 105° with a femoral osteotomy to ensure stability. • With habitual dislocation in toddlers, wait until 3–4 years of age when stability of a 100–105° femoral osteotomy is better due to improved bone density. Most T21 children will be able to learn to walk at this age because they are better able to cooperate with postoperative physiotherapy. Bone strength and remodeling should thus improve. • Warn parents that the child will waddle for up to 2 years after surgery, but with the stimulus of ambulation over time the gait improves and the NSA recovers to within normal range within 5 years. • Even if it is only one hip that is habitually dislocating we recommend bilateral, simultaneous femoral osteotomies. This is because of our earlier experience with unilateral osteotomy, where the contralateral hip almost always began to habitually dislocate postoperatively.

This also prevents leg length inequality, which contributes to the development of dislocation on the second side. After unilateral surgery, when the child stands, a lateralizing moment is applied to the uncorrected valgus proximal femur, promoting instability and even habitual dislocation on the unoperated side. • Remove the plates 1–2 years post-operatively because of frequent peri-implant fractures in our published series, due to the inherent osteopenia in T21 children who have delayed ambulation.

Classic Papers Bennet GC, Rang M, Roye DP, Aprin H, Dislocation of the hip in trisomy 21. J Bone Joint Surg Br. 1982;64(3):289–94. Bennet [10] reported on a series of 28 patients (45 hips) with T21 who were treated for hip dislocations. Bennet was first to recognise four phases in the natural history of hip subluxation in T21. This is discussed above. In his series, a variety of methods were utilized including: closed reduction and hip spica (five patients), capsular plication (two patients), femoral osteotomies (nine patients), femoral osteotomy and capsular plication (three patients), Salter osteotomy and capsular plication (four patients), Chiari osteotomy (four patients), Schanz osteotomies (one patient). Double pelvic osteotomy (one patient). They found poor results with closed reduction and hip spica treatment. They also found that capsular plication combined with femoral ± pelvic osteotomy yielded the best results, but even this was associated with a 50% failure rate. Interestingly, they noted a high infection rate (19%).

Key Evidence The evidence that underpins current treatments for dislocated hips in children with T21 is scarce. A Pubmed search using “Hip dislocation in

25  Hip Problems in Children with Trisomy 21

Trisomy 21” reveals just over 50 published papers. Most are either review articles or case reports. There are a few case series with small patient numbers and short follow up. Knight DM, Alves C, Wedge JH.  Femoral varus derotation osteotomy for the treatment of habitual subluxation and dislocation of the pediatric hip in trisomy 21: a 10-year experience. J Pediatr Orthop. 2011 Sep;31(6):638– 43. This is a single surgeon series of nine children (16 hips) aged below 10 years who were operated on between 1998 and 2008. All had a femoral varus derotation osteotomy with a target femoral NSA of 105° and external rotation of