Inherited bleeding disorders in women [Second edition.] 9781119426028, 1119426022

Citation preview

Inherited Bleeding Disorders in Women

Inherited Bleeding Disorders in Women Second Edition

Edited by Rezan A. Kadir MB ChB MRCOG FRCS(Ed) MD

Professor, Department of Obstetrics and Gynaecology and Katharine Dormandy Haemophilia and Thrombosis Centre, The Royal Free Foundation Hospital, London, UK and Institute for Women’s Health, University College London, London, UK

Paula D. James MD FRCPC

Professor, Department of Medicine Queen’s University, Kingston, Canada

Christine A. Lee MA MD DSc FRCP FRCPath FRCOG Emeritus Professor of Haemophilia University College London, London, UK

This edition first published 2019 © 2019 John Wiley & Sons Ltd. Edition History Wiley‐Blackwell (1e, 2009) All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. The right of Rezan A. Kadir, Paula D. James, and Christine A. Lee to be identified as the authors of the editorial material in this work has been asserted in accordance with law. Registered Office(s) John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial Office 9600 Garsington Road, Oxford, OX4 2DQ, UK For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com. Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats. Limit of Liability/Disclaimer of Warranty The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Library of Congress Cataloging‐in‐Publication Data Names: Kadir, Rezan A., editor. | James, Paula D., editor. | Lee, Christine A., editor. | Preceded by (work): Lee, Christine A. Inherited bleeding disorders in women. Title: Inherited bleeding disorders in women / edited by Rezan A. Kadir, Paula D. James, Professor Christine A. Lee. Description: Second edition. | Hoboken, NJ : Wiley-Blackwell, 2019. | Preceded by Inherited bleeding disorders in women / Christine A. Lee, Rezan A. Kadir, Peter A. Kouides. 2009. | Includes bibliographical references and index. | Identifiers: LCCN 2018032345 (print) | LCCN 2018033046 (ebook) | ISBN 9781119426127 (Adobe PDF) | ISBN 9781119426066 (ePub) | ISBN 9781119426028 (hardback) Subjects: | MESH: Blood Coagulation Disorders, Inherited | Women’s Health Classification: LCC RC647.C55 (ebook) | LCC RC647.C55 (print) | NLM WH 322 | DDC 616.1/570082–dc23 LC record available at https://lccn.loc.gov/2018032345 Cover Design: Wiley Cover Image: © Barbara Bruch Set in 10/12pt WarnockPro by SPi Global, Chennai, India 10 9 8 7 6 5 4 3 2 1

v

Contents Preface to Second Edition  ix Preface to First Edition  xi List of Contributors  xiii 1

Hematological Assessment of a Patient with an Inherited Bleeding Disorder  1 Sue Pavord and Henna Wong

1.1 Introduction  1 1.2 Normal Hemostasis  1 1.3 Defects of Hemostasis  5 1.4 Clinical Presentation of Bleeding  5 1.5 Diagnosis  6 1.6 Approach to a Female with a Bleeding History  6 1.7 Summary  10 ­References  11 2

Bleeding Assessment Tools  13 Sarah H. O’Brien and Paula D. James

2.1 Introduction  13 2.2 Evolution of Vicenza‐Based Bleeding Assessment Tools  13 2.3 Women’s Studies Using Vincenza‐Based Bleeding Tools  22 2.4 Tools to Evaluate Menstrual Blood Loss  23 2.5 Young Women’s Hematology and Bleeding Assessment Tools  25 2.6 Clinical Utility of Bleeding Assessment Tools  25 2.7 Conclusion  26 ­References  27 3

Physiology of Menstruation  29 Jane J. Reavey, Jacqueline A. Maybin, and Hilary O.D. Critchley

3.1 Introduction  29 3.2 Normal Menstruation  29 3.3 Neuroendocrine Hormones  30 3.4 Menarche  31 3.5 Ovarian Follicle Development and Endocrine Function  31 3.6 Endometrium  33 3.7 Secretory Phase  34 3.8 Menstruation  35 3.9 Control of Bleeding  37

vi

Contents

3.10 Endometrial Repair  38 3.11 Proliferative Phase  40 3.12 Summary  40 ­References  41 4 Gynecology  45 Joanna S. Davies and Rezan A. Kadir

4.1 Introduction  45 4.2 Heavy Menstrual Bleeding  48 4.3 Management of Acute Episodes of Heavy Menstrual Bleeding  58 4.4 Ovulation Bleeding  59 4.5 Endometriosis  59 4.6 Other Gynecological Conditions  60 4.7 Conclusion  61 ­References  61 5

Carriers of Hemophilia A and Hemophilia B  65 Roseline d’Oiron

5.1 Inheritance  65 5.2 Screening for the Genetic Status of Carriers of Hemophilia  65 5.3 Confusion Between Genetic and Coagulation Testing  66 5.4 When to Perform Genetic Testing  67 5.5 What Reasons might Contribute to Delayed Genetic Diagnosis of Carriership?  68 5.6 Bleeding Disorders in Carriers of Hemophilia  68 5.7 Quality of Life of Carriers of Hemophilia  73 5.8 Carriers of Hemophilia A and B and Pregnancy  73 5.9 How to Improve Care for Carriers of Hemophilia  77 ­References  79 6

Von Willebrand Disease  83 Carolyn M. Millar

6.1 Introduction  83 6.2 Structure and Function of Von Willebrand Factor  83 6.3 Von Willebrand Factor Levels and Prevalence of Von Willebrand Disease  84 6.4 VWF Levels, the Menstrual Cycle and Pregnancy  84 6.5 Von Willebrand Disease Classification and Inheritance  85 6.6 Clinical Presentation  87 6.7 Menorrhagia and Postpartum Hemorrhage  87 6.8 Diagnosis and Laboratory Testing  88 6.9 Management of Von Willebrand Disease  90 6.10 Management of Gynecological Bleeding  91 6.11 Obstetric Management of Von Willebrand Disease  92 6.12 Neonatal Management  94 6.13 Cases  94 ­References  96 7

Factor XI Deficiency  101 Bethan Myers and Rezan A. Kadir

7.1

Factor XI Structure and Function in Coagulation  101

Contents

7.2 Incidence and Inheritance of Factor XI Deficiency  101 7.3 Bleeding Manifestations and Diagnosis of Factor XI Deficiency  103 7.4 Factor XI Deficiency and Gynecological Issues  103 7.5 Factor XI Deficiency and Pregnancy  105 7.6 Neuroaxial Analgesia and Anesthesia  108 7.7 Management of Postpartum Period  108 7.8 Treatment Options  109 7.9 Neonatal Bleeding  111 7.10 Conclusions and Recommendations  111 ­References  112 8

Rare Bleeding Disorders  117 Danijela Mikovic, Marzia Menegatti, and Flora Peyvandi

8.1 Introduction  117 8.2 Clinical Symptoms  118 8.3 Gynecological and Obstetrical Manifestations in Women with Rare Bleeding Disorders  119 8.4 Laboratory Diagnosis  122 8.5 Differential Diagnosis in Women with Menorrhagia  123 8.6 Treatment  124 8.7 Conclusion  128 ­References  129 9

Inherited Platelet Defects  133 Mike Makris and Clare Samuelson

9.1 Introduction  133 9.2 Normal Platelet Function  133 9.3 Presentation  136 9.4 Investigation  137 9.5 Syndromic and Non‐Syndromic Inherited Platelet Disorders with Recognized Causative Genes  143 9.6 Prenatal Diagnosis  143 9.7 Management  144 9.8 Future Directions  147 9.9 Case Histories  147 9.10 Conclusion  148 ­References  149 10

Genetic and Laboratory Diagnosis  153 Anne C. Goodeve

10.1 Introduction  153 10.2 Phenotypic Analysis of Hemophilia A  154 10.3 Phenotypic Analysis of Hemophilia B  156 10.4 Phenotypic Analysis of von Willebrand Disease  157 10.5 Phenotypic Analysis of Inherited Bleeding Disorders  157 10.6 Genetic Analysis of Hemophilia A  160 10.7 Genetic Analysis of Hemophilia B  162 10.8 Genetic Analysis of von Willebrand Disease  163 10.9 Guidelines  164

vii

viii

Contents

10.10 Summary 164 ­References  164 11

Antenatal Diagnosis  167 Rezan A. Kadir, Irena Hudecova, and Claudia Chi

11.1 Introduction  167 11.2 Genetic Counseling  167 11.3 Prenatal Diagnosis  170 11.4 Prenatal Diagnosis of Hemophilia  177 11.5 Prenatal Diagnosis of von Willebrand Disease  180 11.6 Prenatal Diagnosis of Rare Bleeding Disorders  180 11.7 Preimplantation Genetic Diagnosis  181 11.8 Views about and Experiences of Prenatal Diagnosis of Women in Families Affected with Inherited Bleeding Disorders  183 11.9 Termination of Pregnancy  185 ­References  186 12

Analgesia and Anesthesia for Pregnant Women with Inherited Bleeding Disorders  191 Anne‐Sophie Bouthors, Adrian England, and Rezan A. Kadir

12.1 Introduction  191 12.2 Non‐Pharmacological Methods  192 12.3 Pharmacological Methods  192 12.4 Conclusion  201 ­References  202 13

The Newborn  205 Manuel Carcao and Vanessa Bouskill

13.1 Introduction  205 13.2 Developmental Hemostasis  205 13.3 Laboratory Hemostatic Evaluation of the Neonate  208 13.4 When to Suspect a Congenital Bleeding Disorder in a Newborn  209 13.5 Congenital Bleeding Disorders and Their Presentation in Newborns  213 13.6 Management of Bleeding in Neonates with Congenital Bleeding Disorders  218 13.7 Conclusion  221 ­References  222 14

Women with Inherited Bleeding Disorders in Different Cultural Settings  225 Tahira Zafar, Jameela Sathar, Ali T. Taher, Fadi G. Mirza, and Christine A. Lee

14.1 Introduction  225 14.2 Pakistan  225 14.3 Malaysia  227 14.4 Lebanon  229 14.5 Discussion  231 ­References  232 Index  235

ix

Preface to Second Edition It has been a privilege to edit the second ­edition of Inherited Bleeding Disorders in Women. The first edition was published in 2009, almost a decade ago, and during that time there has been enormous endeavor in the research, management, and education of inherited bleeding disorders in women. This is reflected in the contents of this new edition. We have comprehensively updated the chapters covering the gynecological and obstetric issues for carriers of hemophilia, women with von Willebrand disease, rare bleeding disorders, and inherited platelet disorders to provide an evidence‐based, practical approach to management. The enormous developments in genetic analysis are included in the chapters on laboratory and antenatal diagnosis. New chapters include the use of

bleeding assessment tools in the context of women’s health, and a consideration of inherited bleeding disorders in different cultures and marriage within the family. As before, the book is a collaboration, written by hematologists, obstetrician‐­ ­ gynecologists, laboratory scientists, a nurse, and anesthetists who have expertise in the field. Our aspiration continues to be the high quality of care for women with inherited bleeding disorders worldwide and we hope this book will be useful for those providing care and for the affected women themselves. January 2018

Rezan A. Kadir Paula D. James Christine A. Lee Cover image: ‘Menorrhagia Healing’ © Barbara Bruch 1991

xi

Preface to First Edition In 1926, Erik von Willebrand described a large kindred from the Åland Islands, an archipelago in the Baltic Sea, many of whom had a bleeding disorder. The index case was a little girl called Hjordis, who presented with severe epistaxis and died at the onset of her fourth menstrual period. Her maternal grandmother died from hemorrhage after childbirth in her  only pregnancy. Von Willebrand wrote that the condition was ­particularly prevalent in women. This first description of von Willebrand disease underlined the hemostatic challenges of menstruation and childbirth for those women with an inherited bleeding disorder. Until recently, the predominant issue for men with hemophilia has been safe and effective treatment, and most effort has been directed to the resolution of transfusion‐ transmitted disease. Furthermore, since hemophilia is a sex‐linked disorder, there has been a failure to recognize that women have inherited bleeding disorders. Thus, the su­bstantial morbidity caused in women with inherited bleeding disorders has only recently been addressed in a comprehensive way. It is important that collaboration in the care and

research of bleeding disorders in women continue as many challenges remain. The main task now is to identify those women who do not realize they may have a treatable condition. The patient advocacy organizations are crucial to this endeavor. There also remains the challenge of developing more effective, tolerable, and widely available therapies for controlling menorrhagia and postpartum hemorrhage. This book is written by hematologists, obstetrician‐gynecologists, an anesthetist, and those involved in patient advocacy. It covers the gynecological and obstetric issues for carriers of hemophilia, women with von Willebrand disease, rare bleeding disorders, and inherited platelet disorders. We hope that this book is a modest step towards safe motherhood and provision of quality of care for women with bleeding disorders worldwide and that all those providing care for these women, as well as the women themselves, will find it useful. December 2008

Christine A. Lee Rezan A. Kadir Peter A. Kouides

xiii

List of Contributors Vanessa Bouskill, MN RN(EC)

Joanna S. Davies, MB ChB MD

Department of Nursing Hospital for Sick Children Toronto Canada

Department of Obstetrics and Gynaecology and Katharine Dormandy Haemophilia and Thrombosis Centre Royal Free Foundation Hospital London UK

Anne‐Sophie Bouthors, MD

Department of Anesthesia and Intensive Care Maternité Jeanne de Flandre Academic Hospital Lille France Manuel Carcao, MD FRCP(C) MSc

Division of Haematology/Oncology Department of Paediatrics and Child Health Evaluative Sciences Research Institute Hospital for Sick Children Toronto Canada

Roseline d’Oiron, MD

Reference Centre for Hemophilia and Rare Congenital Bleeding Disorders University Hospitals Paris Sud – Bicêtre Hospital – APHP Le Kremlin‐Bicêtre France Adrian England, MBBS FRCA MD

Department of Anaesthesia Royal Free Hospital London UK Anne C. Goodeve, BSc PhD

Department of Obstetrics and Gynaecology National University Hospital Singapore

Department of Infection Immunity and Cardiovascular Disease University of Sheffield Medical School Sheffield UK

Hilary O.D. Critchley, MD FRCOG FMedSci FRSE

Irena Hudecova, PhD

MRC Centre for Reproductive Health University of Edinburgh Edinburgh UK

Cancer Research UK Cambridge Institute University of Cambridge Li Ka Shing Centre Cambridge UK

Claudia Chi, MBBS MRCOG MD FAMS

xiv

List of Contributors

Paula D. James, MD FRCPC

Danijela Mikovic, MD PhD

Department of Medicine Queen’s University Kingston Canada

Haemostasis Department with Registry of Inherited Bleeding Disorders Blood Transfusion Institute of Serbia Belgrade Serbia

Rezan A. Kadir, MB ChB MRCOG FRCS(Ed) MD

Department of Obstetrics and Gynaecology and Katharine Dormandy Haemophilia and Thrombosis Centre Royal Free Foundation Hospital London UK

Carolyn M. Millar, MD FRCP FRCPath

and

Fadi G. Mirza, MD FACOG

Institute for Women’s Health University College London London UK Christine A. Lee, MA MD DSc FRCP FRCPath FRCOG

University College London London UK Mike Makris, MA MBBS MD FRCP FRCPath

Sheffield Haemophilia and Thrombosis Centre Royal Hallamshire Hospital Sheffield UK

Centre for Haematology and Department of Experimental Medicine Imperial College Healthcare NHS Trust Imperial College London UK Faculty of Medicine and Medical Center American University of Beirut Beirut Lebanon and Columbia University New York USA Bethan Myers, MA FRCP FRCPath

Department of Haematology University Hospitals of Leicester NHS Trust and Lincoln County Hospital Lincoln UK Sarah H. O’Brien, MD

MRC Centre for Reproductive Health University of Edinburgh Edinburgh UK

Division of Pediatric Hematology/Oncology Nationwide Children’s Hospital/The Ohio State University College of Medicine Columbus USA

Marzia Menegatti, BSc PhD

Sue Pavord, MB ChB FRCP FRCPath

Luigi Villa Foundation and Angelo Bianchi Bonomi Hemophilia and Thrombosis Center Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico Milan Italy

Department of Haematology Oxford University Hospitals NHS Foundation Trust Oxford UK

Jackie A. Maybin, MB ChB MRCOG PhD

List of Contributors

Flora Peyvandi, MD PhD

Ali T. Taher, MD PhD FRCP

Angelo Bianchi Bonomi Hemophilia and Thrombosis Center Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico and Department of Pathophysiology and Transplantation University of Milan Milan Italy

American University of Beirut Medical Center Beirut Lebanon

Jane Reavey, MA BMBCh MRCOG

MRC Centre for Reproductive Health University of Edinburgh Edinburgh UK Clare Samuelson, MB ChB MA

Sheffield Haemophilia and Thrombosis Centre Royal Hallamshire Hospital Sheffield UK Jameela Sathar, MD MRCP FRCPath

Department of Haematology Ampang Hospital Malaysia

and Emory School of Medicine Atlanta USA Henna Wong, MBBS MRCP FRCPath

Department of Haematology Oxford University Hospitals NHS Foundation Trust Oxford UK Tahira Zafar, MB DCP FRCPath

Haemophilia Treatment Centre Rawalpindi Islamabad Pakistan and Pakistan Haemophilia Patients Welfare Society Rawalpindi Islamabad Pakistan

xv

1

1 Hematological Assessment of a Patient with an Inherited Bleeding Disorder Sue Pavord and Henna Wong Oxford University Hospitals NHS Foundation Trust, Oxford, UK

1.1 ­Introduction The inherited bleeding disorders (IBDs) are a heterogeneous group of disorders affecting the hemostatic system. In individuals in whom the underlying abnormality has been identified, the majority of IBDs are due to von Willebrand disease (VWD) and disorders of coagulation factors; a small proportion are due to abnormalities in platelet count or function or defects in the fibrinolytic system [1]. Around 2% of patients registered with a bleeding disorder do not have a classifiable disorder [1]. Individuals with IBDs may give a life‐long history of excessive bruising or bleeding, but many only manifest when faced with a hemostatic challenge or are picked up incidentally by abnormal coagulation tests. Indeed some, such as certain cases of factor XI (FXI) deficiency, may not have a bleeding phenotype at all, even when exposed to hemostatic challenges. IBDs can affect all genders, but women with IBDs face added challenges related to menstruation, pregnancy, and childbirth. Undiagnosed bleeding disorders can often be the cause of heavy menstrual bleeding and also the cause of or a contributory factor for other gynecological problems, such as bleeding from the corpus luteum [2].

Women with IBDs may present with a positive bleeding history or have a known family history. Manifestations of bleeding can vary, even within the same type of disorder, because of the influence of concomitant inherited and acquired factors. An integrated clinical and laboratory assessment is therefore essential in the diagnostic work‐up. This chapter will cover the mechanisms of normal hemostasis and an approach to the clinical and laboratory hematological assessment of a patient with a suspected IBD.

1.2 ­Normal Hemostasis After damage to the lining of the blood vessel wall, the body responds with physiological mechanisms to stop bleeding and maintain hemostasis, without causing more widespread thrombosis. This co‐ordinated process involves components of the blood, including platelets and clotting factors, with the overall aim of forming a stable blood clot (Figure 1.1). Hemostasis is achieved through a delicate balance of pro‐ and anticoagulant factors to stop bleeding while simultaneously avoiding development of pathological thrombi [3].

Inherited Bleeding Disorders in Women, Second Edition. Edited by Rezan A. Kadir, Paula D. James, and Christine A. Lee. © 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.

Vessel lumen

Endothelium Collagen

Collagen

Vasoconstriction

Tissue factor

Serotonin

Platelet

activation and Activated platelet

Coagulation pathway

VWF

Platelet adhesion

fibrinogen

aggregation

PRIMARY HEMOSTASIS

Antithrombotic mechanisms

Thrombin

Fibrinogen Primary platelet plug

Inhibitors of coagulation

Fibrin Stable fibrin blood clot

SECONDARY HEMOSTASIS

Plasmin

Fibrinolysis and clot breakdown

FIBRINOLYSIS

Figure 1.1  Overview of hemostasis and key components. A vascular injury exposes collagen that allows platelets to adhere via VWF to the subendothelium. Activation of platelets occurs and platelets aggregate together via VWF and fibrinogen. Primary hemostasis results in the formation of the initial platelet plug. Tissue factor activates the coagulation pathway in parallel with platelet activation; both pathways enhance each other. Fibrinolysis prevents excessive thrombus formation, through the generation of plasmin followed by the digestion of fibrin. Source: Modified from www.thrombocyte.com/hemostasis‐definition.

Hematological Assessment of a Patient with an IBD

1.2.1  Primary Hemostasis In the early stages after vessel injury, interactions between platelets, the subendothelium, and adhesive proteins lead to the formation of a platelet plug (primary hemostasis). The three main steps of primary hemostasis are as follows. 1)  Platelet adhesion: following vessel injury, von Willebrand factor (VWF) binds to specific sites on exposed collagen. Platelets adhere to the exposed subendothelial matrix (directly or indirectly via VWF). This is mediated through binding of VWF with the platelet glycoprotein GP1b, while GPVI interacts with collagen, and platelet β1 integrin with laminin, collagen, and fibronectin. These interactions enable firm adhesion of platelets to the exposed subendothelial matrix [3]. 2)  Platelet activation: platelet adhesion to the subendothelium triggers shape change and release of platelet α and dense granule contents. This activation recruits and activates additional platelets to the injured site. (Thrombin, produced by the coagulation pathway, adds to the activation of platelets.)

APTT measures XII, XI, IX, VIII, X, V, II, I

Intrinsic pathway XII XI IX VIII

3)  Platelet aggregation and platelet plug ­formation: thrombin cleaves fibrinogen and the resulting fibrin monomers form a bridge between activated platelets, causing platelets to aggregate together, forming a platelet plug. The GPIIb/IIIa platelet receptor is converted into its high‐affinity conformation, allowing for stable interactions between the receptor and fibrin, VWF, and fibronectin. 1.2.2  Secondary Hemostasis Secondary hemostasis usually occurs simultaneously with primary hemostasis. After endothelial damage, tissue factor (TF) is exposed, which binds to and activates FVII. The TF‐FVIIa complex then stimulates generation of small amounts of thrombin ­ and FXIa through the extrinsic pathway. Thrombin generation is amplified through the intrinsic pathway starting with FXI and through the downstream cascade including co‐factors FVIII and FV (Figure  1.2). These enzymatic reactions occur on the surface of platelets and other cell surfaces. This leads to the formation of FXa on the platelet surface which, aided by its co‐factor, FVa, generates

Extrinsic pathway Tissue factor VII Prothrombin time measures VII, X, V, II, I

Common pathway X V II

Fibrinogen ↓

Fibrin clot

Thrombin time

Figure 1.2  Pathways of coagulation: the intrinsic (in vitro) pathway as measured by the activated partial thromboplastin time (APTT) and the extrinsic (in vivo) pathway as measured by the prothrombin time (PT). Both intrinsic and extrinsic pathways share a common final pathway.

3

4

Inherited Bleeding Disorders in Women

an explosive burst of thrombin. Thrombin catalyzes the conversion of fibrinogen to fibrin. Fibrin polymers form a fibrin network, which is stabilized by FXIII. As the clot forms, circulating red blood cells, white blood cells, and platelets become incorporated into its structure. 1.2.3 Fibrinolysis Fibrinolysis is tightly regulated by the fibrinolytic system. Fibrinolysis is initiated by the proteases tissue plasminogen activator (tPA) or urokinase‐like plasminogen activator (uPA), which convert plasminogen to the active plasmin. Plasmin cleaves fibrin and fibrinogen, leading to clot breakdown. (a)

1.2.4  Cell‐Based Model of Hemostasis The importance of cells and cell surfaces in hemostasis is reflected in the “cell‐based model of hemostasis” described by Hoffman and Monroe [4]. It is more representative of  in vivo coagulation than the traditional “cascade model.” There are three main phases in the cell‐based model (Figure  1.3). The ­initiation phase occurs on the TF‐bearing cell. Injury exposes the TF‐bearing cell to flowing blood and plasma‐based VIIa. The TF‐VIIa complex results in the generation of a small amount of FIXa, Xa, and thrombin. In the second phase, “amplification,” the small amount of thrombin activates platelets,

(b)

INITIATION X

TF VIIa TF

VIIa

AMPLIFICATION

II prothrombin Xa Va

VIII/VWF → VIIIa + VWF V → Va

thrombin IIa

XI → XIa

TF-expressing cell

Platelet

IX (c) X

IXa

II Prothrombin

IX IXa VIIIa Xa Va XIa

IIa thrombin

Activated platelet

PROPAGATION

Fibrin FXIIIa TAFI

Figure 1.3  Cell‐based model of hemostasis. The cell‐based model comprises three overlapping stages: (a) initiation, (b) amplification, and (c) propagation. (a) Initiation phase: this occurs on the TF‐expressing cell and is initiated when injury exposes the TF‐bearing cell to the flowing blood. A small amount of FIXa, Xa, and thrombin is formed. (b) Amplification phase: the small amount of thrombin generated from the initiation phase activates platelets, releases VWF, and leads to generation of activated forms of FV, FVIII, and FXI. (c) Propagation phase: activated coagulation factors from the previous phases aggregate on the platelet surface. FVIII complexes with FIX to form the tenase complex, resulting in FXa generation on the platelet surface. The prothrombinase complex forms and results in large amounts of thrombin generation. This thrombin burst leads to fibrin formation and also activates FXIII and TAFI. FXIIIa cross‐links fibrin strands to form a stable fibrin network, and TAFI protects the clot from plasmin‐mediated fibrinolysis. Source: Adapted from Hoffman and Monroe [5] and Kessler [6].

Hematological Assessment of a Patient with an IBD

releases VWF, and leads to generation of FVa, FVIIIa, and FXIa. The propagation phase is characterized by the migration of large numbers of platelets to the site of injury and the production of the tenase complex, which results in FXa generation on the platelet surface. The FXa generated on platelets rapidly binds to Va and converts prothrombin to thrombin. Large amounts of thrombin are generated, converting fibrinogen to fibrin. The fibrin clot is stabilized by activation of thrombin‐activatable fibrinolysis inhibitor (TAFI) and FXIII. Although the intrinsic and extrinsic pathways may be less representative of in vivo coagulation, an appreciation of the components of each pathway is helpful when interpreting abnormalities in the activated partial thromboplastin time (APTT) and prothrombin time (PT). Deficiencies of clotting factor may prolong the APTT (tests the intrinsic pathway) and PT (extrinsic pathway) (see Figure 1.2).

1.3 ­Defects of Hemostasis The ability to achieve hemostasis and stop bleeding depends on the integrity of all components of the pathway [7] and knowledge of the underlying hemostatic defect can help with categorization of the bleeding disorder. For a patient with an undiagnosed disorder, it may be possible to predict whether the defect affects primary or secondary hemostasis from the type and pattern of bleeding elicited in the history. The main IBDs can be categorized according to the underlying defect. ●●

●●

●●

Hemophilia – deficiency of clotting factors FVIII (hemophilia A) or FIX (hemophilia B). Patients lack amplification of the coagulation cascade by FIX with co‐factor VIII [7]. VWD, where there is deficiency of VWF and therefore impaired platelet adhesion and aggregation. Platelet function disorders, for example affecting the platelet receptor or platelet signaling pathways.

●●

Rare coagulation disorders, such as deficiency of factors V, VII, IX, and XIII or fibrinogen disorders.

1.4 ­Clinical Presentation of Bleeding Inherited bleeding disorders may manifest with a variety of bleeding symptoms. There may be considerable variability in symptom severity even in patients affected by the same disorder.  Acquired defects in the hemostatic system,  for example caused by antiplatelet or antithrombotic medication, may exacerbate any underlying inherited disorder. Disorders of primary hemostasis often present with mucocutaneous bleeding and the early onset of bleeding after injury or trauma, compared with more delayed bleeding or overt bleeding with disorders of secondary hemostasis (abnormalities with clotting factors or fibrinolysis). Bleeding symptoms are common in healthy individuals; over 20% of the general population report at least one bleeding symptom [8]. In addition, by chance alone, 1 in 20 people will have a result outside the “normal” reference range. Under‐ and overdiagnosis of bleeding disorders can have serious sequelae. Underdiagnosis will lead to inappropriate or inadequate medical treatment, but with overdiagnosis, healthy individuals can be needlessly exposed to hemostatic therapy and potential complications [9]. Menstruation, pregnancy, and childbirth present recurrent hemostatic challenges to females with and without IBDs. Menorrhagia is the most common symptom experienced by women with an IBD (up to 80% of women with IBD report menorrhagia) [10]. However, it is also common in the general population; 5–10% of women of reproductive age will seek medical attention for menorrhagia [11]. Menorrhagia may be due to endocrine, inherited bleeding or gynecological disorders but prior to comprehensive hemostatic testing, the underlying etiology was only found in ~50% of cases [12]. An IBD is found in up to

5

6

Inherited Bleeding Disorders in Women

20% of women with menorrhagia and a normal pelvic examination [9, 13–15], the most common being VWD [15]. Laboratory abnormalities of hemostasis, especially platelet function defects, are common among women with unexplained menorrhagia, but their clinical significance requires further study, especially if the abnormality is mild [14].

1.5 ­Diagnosis Making a diagnosis of an IBD has life‐long consequences. An accurate diagnosis is important as the impact is far‐reaching, especially with precautions that affect perioperative management, work, and lifestyle activities as well as implications for screening and investigation of family members. Despite the development and use of standardized bleeding assessment tools, distinguishing how to further investigate patients with bleeding symptoms can be challenging. Joint input from gynecology and hemostasis specialists is recommended to ensure the bleeding status has been thoroughly investigated in women with suspected IBD [16, 17].

1.6 ­Approach to a Female with a Bleeding History 1.6.1 History A woman may present for consultation with positive bleeding symptoms related to the gynecological system or additional bleeding symptoms, or she may have a family history of a bleeding disorder. Assessment should include a history of bleeding symptoms and the site, duration, frequency, and severity of bleeding (Table 1.1). Particular enquiry should include a history of mucosal bleeding, menorrhagia, epistaxis, easy bruising, and any bleeding episodes after a hemostatic challenge such as surgery, dental extraction, childbirth (postpartum hemorrhage (PPH)), and treatment required for any bleeding e­ pisodes. Bleeding that required blood t­ ransfusion, use of

­emostatic adjuncts, and antifibrinolytic h agents, such as tranexamic acid, may indicate a significant or severe bleed. Bleeding history may be subjective and specific tools have been developed to provide a more objective and standardized assessment of bleeding symptoms, for example the International Society on Thrombosis and Haemostasis (ISTH) Bleeding Assessment Tool and pictorial bleeding assessment chart for menorrhagia (see Chapter 2 for bleeding assessment tools). These standardized tools should be used wherever possible. Although women with IBDs are more likely to experience menorrhagia, they are also at risk of other problems that may present with increased bleeding such as hemorrhagic ovarian cysts, bleeding from the corpus luteum, endometriosis, hyperplasia, polyps, fibroids, pregnancy, and childbirth [20, 21]. In women with an IBD, especially VWD, the risk of PPH is increased [22, 23] but primary PPH alone is not a good predictor of IBDs [24]. 1.6.1.1  Past Medical History and Family History

Any other medical problems and pregnancy history should also be established. A family history of a bleeding disorder may be useful but a negative family history does not exclude the presence of an IBD, especially if the underlying genetic abnormality has incomplete penetrance or if a de novo mutation is present. 1.6.1.2  Medication History

Drug history should also include the use of contraceptive medication, antiplatelet drugs, or other anticoagulants, as this may affect the results of hemostatic assays. Platelet function tests can be affected by high concentrations of alcohol and caffeine and other food (for a list of drugs and food interfering with platelet function see [25]). 1.6.2  Physical Assessment In addition to a general systemic examination, a number of physical signs may be particularly relevant in the physical assessment

Hematological Assessment of a Patient with an IBD

Table 1.1  Bleeding history and salient features indicating non‐trivial bleed. Site of bleeding

Features

Epistaxis

Any nosebleed that causes interference or distress with daily or social activities

Oral cavity

Gum bleeding causing frankly bloody sputum, lasts for 10 minutes or longer on more than one occasion. Tooth eruption or spontaneous tooth loss bleeding that requires assistance or supervision by a physician or lasts at least 10 minutes. Bleeding occurring after bites to lips, cheek, and tongue lasting at least 10 minutes or causing a swollen tongue or mouth

Menorrhagia

Any bleeding that interferes with daily activities such as work, housework, exercise, or social activities during most menstrual periods

Postpartum hemorrhage

Requires medical consultation, supportive treatment

Muscle hematoma

Spontaneous hematoma

Hemarthrosis

Spontaneous bleeding into joint

Cutaneous

Bruises are considered significant when five or more (>1 cm) in exposed areas. Significant features of bruising: (a) atraumatic bruising, (b) bruising occurring at least weekly, and (c) bruises greater than 5 cm

Gastrointestinal bleeding

Bleeding not related to peptic ulcer disease

Surgery

Any bleeding judged by the surgeon to be abnormally prolonged that causes a delay in discharge or requires some supportive treatment

Dental extraction

Any bleeding occurring after leaving the dentist’s office and requiring a new, unscheduled visit or prolonged bleeding at the dentist’s office causing a delay in the procedure or discharge

Central nervous system bleeding

Subdural or intracerebral hemorrhage

Source: Modified from [18, 19].

of a patient with a bleeding disorder. Careful inspection of the skin, mouth, and nose may reveal bruising or petechiae. Joint examination may reveal swelling or evidence of contractures. Other inherited disorders that cause a bleeding tendency due to a connective tissue disorder, such as Ehlers–Danlos, should also be considered. Signs of a connective tissue disorder may include hypermobility of joints and loose joints. Although occurring very rarely, some inherited platelet disorders may present with other syndromic features such as ocular albinism with late‐ onset sensorineural deafness. 1.6.3 Investigations The specificity of bleeding symptoms may be poor and accurate laboratory assessment of

the hematological and coagulation system (Table 1.2) is important [7]. A first‐line set of investigations includes the full blood count, blood film, and standard coagulation screen (PT, APTT, Clauss fibrinogen, and thrombin time). In the full blood count, it is particularly important to know if the platelet count is within the normal range, elevated or low, as this can help direct further investigations (see Table 1.2). Although the PT and APTT are used widely, they were not designed to be used as screening tools and only assess part of the hemostatic system (see Figure 1.2). If prolonged, they may indicate the presence of an inhibitor or a clotting factor deficiency. An inhibitor could be due to either an antibody that inhibits the activity of a specific clotting factor or a non‐specific inhibitor, such as a

7

8

Inherited Bleeding Disorders in Women

Table 1.2  Laboratory assessment of hematological and coagulation system. Test

Rationale

Possible problem/differential

Full blood count

Is anemia present? This may indicate ongoing bleeding. Look at the mean corpuscular volume (MCV) – if low, this may indicate iron deficiency and/or hemoglobinopathy. Assess platelet count and platelet volume (mean platelet volume)

Anemia Thrombocytopenia or thrombocytosis Large platelets can be associated with inherited platelet disorders such as Bernard–Soulier syndrome (BSS), myosin heavy chain (MYH9)‐ related disorders or acquired platelet disorders, e.g. immune thrombocytopenia

Blood film

To look at the size and shape of platelets and any other abnormalities present

For example, BSS is a rare autosomal recessive disease associated with bleeding tendency, giant platelets, and thrombocytopenia. MYH9‐related disorders are characterized by large platelets and thrombocytopenia

First‐line standard coagulation tests PT

Prothrombin time

See Figure 1.2

APTT

Activated partial thromboplastin time

See Figure 1.2

Clauss fibrinogen

Assessment of functional fibrinogen

If prolonged, could be due to hypofibrinogenemia, afibrinogenemia, dysfibrinogenemia

Thrombin time (TT)

To assess fibrinogen and presence of heparin

TT prolonged in presence of heparin, direct thrombin inhibitors, fibrinogen disorders

Factor assays

FVIII, IX, XI

Hemophilia A, VWD, hemophilia B, factor XI deficiency

Von Willebrand screen (VWF antigen (Ag) activity (e.g. RiCoF))

To investigate VWD

VWD

Blood group

People with blood group O have 25% lower VWF and FVIII levels

–

Iron studies (ferritin, transferrin saturation, serum iron)

If suspected iron deficiency

Iron deficiency (low ferritin, high transferrin saturation, low serum iron)

Light transmission platelet aggregometry

Assessment of platelet function

Abnormalities in aggregometry with Glanzmann’s thrombasthenia, BSS, afibrinogenemia, VWD

Others, e.g. TEG, ROTEM, thrombin generation

Global assay of hemostasis

Platelet storage pool disorder, platelet release defect

Other clotting factors, e.g. FXIII, α2 antiplasmin

If the above tests are normal and there is a strong clinical suspicion of an inherited bleeding disorder

FXIII deficiency, α2 antiplasmin deficiency

Further tests

More specialized tests

Hematological Assessment of a Patient with an IBD

Table 1.2  (Continued) Test

Rationale

Possible problem/differential

Genetic mutation analysis

Test if known mutation

For example, VWF mutation

Genomics screen

May be in offered in specialized laboratories

RiCoF, ristocetin co‐factor; ROTEM, rotational thromboelastometry; TEG, thromboelastography; VWD, von Willebrand disease; VWF, von Willebrand factor.

lupus anticoagulant (a lupus anticoagulant is associated with increased risk of thrombosis rather than bleeding). However, a PT or APTT in the normal range does not preclude the diagnosis of a bleeding disorder. They will not pick up platelet defects or mild VWD. Therefore, laboratory results should always be interpreted in light of the clinical history. If the PT or APTT is prolonged, the next step is to do a mixing study (50:50 mix). Here, normal plasma containing normal levels of clotting factor are mixed with the patient’s sample. If, after mixing, there is no correction of the prolonged APTT or PT, this suggests the presence of an inhibitor in the patient’s blood. If there is correction, this suggests a factor deficiency and specific factor assays are performed to identify the deficiency. The Clauss fibrinogen should be measured, rather than the PT‐based fibrinogen. The Clauss fibrinogen is a quantitative, functional assay which measures the ability of fibrinogen to form a fibrin clot after being exposed to a high concentration of purified thrombin. For suspected dysfibrinogenemia, there will be a discrepancy between functional activity and antigen level (measured by an enzyme‐linked immunosorbent assay (ELISA)‐based immunological test). The thrombin time can also be used to assess fibrinogen, although it has mostly been superseded by the Clauss fibrinogen. A prolonged thrombin time may indicate low fibrinogen level (hypofibrinogenemia or afibrinogenemia) and/or abnormal fibrinogen function (dysfibrinogenemia). Thrombin

time may also be prolonged if there is heparin present, or if the D‐dimers are elevated. 1.6.3.1  Preanalytical Factors in Coagulation Testing

Preanalytical factors are the leading cause of error in coagulation testing [26, 27]. The preanalytical phase describes all actions and aspects of the medical laboratory diagnostic pathway, from when the test is requested up until the analytical phase. Several preanalytical factors are particularly relevant in women. There have been some reports of lower VWF levels in the first few days of the menstrual cycle. Combined oral contraceptive pills (COCPs) and hormone replacement therapy (HRT) may also affect VWF levels. The elevated VWF levels can mask an underlying VWD [28], although the newer combination OCs (which are of lower dose potency than the estrogen preparations used in the initial case reports) do not appear to have the same effect. Contraceptives can also interfere with other tests of coagulation. They have been reported to lead to increased concentrations of fibrinogen, prothrombin, and factors VII, VIII, and X, and reduction in some coagulation inhibitors [27, 29, 30]. A practical approach would be to test women prior to starting the OC, if possible, but to obtain VWF testing if OCs have  already been started. It is also important to consider the effect of pregnancy where f­actor levels, particularly of VWF, FVIII, and fibrinogen, rise and reach peak levels  in the third trimester and continue to be  elevated to a lesser degree postpartum.

9

10

Inherited Bleeding Disorders in Women

When performing tests of coagulation, the following factors are important to note and steps should be taken to control for these conditions [26]. ●●

●●

●●

●●

●●

Avoid intense physical exercise for at least 24 hours prior to venepuncture. For the diagnosis of VWD in fertile women, blood samples should be obtained on days 1–4 of the menstrual cycle. This may aid in the diagnosis in women with borderline values obtained at other times. Test women before starting combined oral contraceptives and HRT if possible. For the diagnosis of inherited disorders of hemostasis (particularly VWD and FVIII deficiency), samples should be obtained when normal menstrual cycles have returned or at least two months postpartum. All abnormal values obtained in connection with pregnancy should be verified with repeat blood sampling. Avoid long transport times from venepuncture to hemostasis testing. Cold storage of whole blood can lead to artificially low VWF levels.

1.6.3.2  Specialized Tests of Coagulation

Specialized coagulation assays should be ­performed in a hemostasis laboratory with internal and external quality assurance, in conjunction with other tests of hemostasis and interpreted in light of the clinical history. One of the limitations with current standard coagulation tests is that they look at individual components of the hemostatic pathway. This may be a useful starting point in the diagnostic pathway but they do not give an overall measure of global hemostasis. Tests evaluating global hemostatic capacity (thrombin generation and viscoelastic hemostatic assays, e.g. thromboelastography (TEG) or rotational thromboelastometry (ROTEM)) may provide more accurate evaluation of in vivo hemostasis, as they more effectively assess rate/total thrombin generated and whole‐blood clot formation.

Generally, there seems to be a very poor ­correlation between laboratory findings and bleeding genotype‐phenotype in the rare bleeding disorders such as deficiencies of fibrinogen, prothrombin, FV, combined FV and FVIII, FVII, FX, FXI, and FXIII. TEG and ROTEM show particular promise in the evaluation of hemostasis in patients with rare bleeding disorders where they may be clinically informative [31, 32]. Further work is required for validation in IBDs before widespread clinical use. Advances in genetic and molecular diagnostics have also been seen in the field of hemostasis. Next‐generation sequencing (NGS) has transformed the scale and cost‐ effectiveness of genetic testing and has emerged as a valuable tool, particularly for the diagnosis of inherited platelet disorders [33]. This is an evolving field and these tests are already becoming more widely available in the diagnostic work‐up of an inherited platelet defect (IPD) [1]. 1.6.4 Interpretation of Hemostatic Assays Abnormalities should be interpreted in light of the clinical history and taking into account any preanalytical factors. When assessing low VWF and FVIII levels, it is particularly important to consider whether the patient has blood group O as patients with this blood group have lower levels. Any abnormalities in hemostatic assays should be repeated.

1.7 ­Summary The evaluation of bleeding symptoms can be challenging. Definitive diagnosis depends on a unified approach to clinical and laboratory assessment, using objective bleeding assessment tools where possible and considering the potential limitations of tests when interpreting results.

Hematological Assessment of a Patient with an IBD

­References 1 Sivapalaratnam, S., Collins, J., and

2

3

4

5

6

7

8

9

10

11

Gomez, K. (2017). Diagnosis of inherited bleeding disorders in the genomic era. Br. J. Haematol. James, A. H. (2011). Diagnosis and management of women with bleeding disorders – international guidelines and consensus from an international expert panel. Haemophilia 17 (Suppl. 1): 3–5. Haley, K. M., Recht, M., and McCarty, O. J. T. (2014). Neonatal platelets: mediators of primary hemostasis in the developing hemostatic system. Pediatr. Res. 76 (3): 230–237. Hoffman, M. and Monroe, D. M. III (2001). A cell‐based model of hemostasis. Thromb. Haemost. 85 (6): 958–965. Hoffman, M. and Monroe, D. M. (2017). Impact of non‐vitamin K antagonist oral anticoagulants from a basic science perspective. Arterioscler. Thromb. Vasc. Biol. 37 (10): 1812–1818. Kessler, C. M. (2005). New perspectives in hemophilia treatment. Hematol. Am. Soc. Hematol. Educ. Program 429–435. Kouides, P. A. and Philipp, C. (2009). Approach to the patient with an inherited bleeding disorder. In: Inherited Bleeding Disorders in Women, 1–11. Wiley‐Blackwell. Rodeghiero, F., Castaman, G., Tosetto, A. et al. (2005). The discriminant power of bleeding history for the diagnosis of type 1 von Willebrand disease: an international, multicenter study. J. Thromb. Haemost. 3 (12): 2619–2626. Boender, J., Kruip, M. J. H. A., and Leebeek, F. W. G. (2016). A diagnostic approach to mild bleeding disorders. J. Thromb. Haemost. 14 (8): 1507–1516. Byams, V. R., Kouides, P. A., Kulkarni, R. et al. (2011). Surveillance of female patients with inherited bleeding disorders in United States Haemophilia Treatment Centres. Haemophilia 17: 6–13. Kouides, P. A. and Kadir, R. A. (2007). Menorrhagia associated with

12 13

14

15

16

17

18

19

20

laboratory abnormalities of hemostasis: epidemiological, diagnostic and therapeutic aspects. J. Thromb. Haemost. 5: 175–182. Rees, M. (1987). Menorrhagia. BMJ 294 (6574): 759–762. De Wee, E. M., Knol, H. M., Mauser‐ Bunschoten, E. P. et al. (2011). Gynaecological and obstetric bleeding in moderate and severe von Willebrand disease. Thromb. Haemost. 106 (5): 885–892. Miller, C. H., Philipp, C. S., Stein, S. F. et al. (2011). The spectrum of haemostatic characteristics of women with unexplained menorrhagia. Haemophilia 17 (1): e223–e229. Kadir, R. A., Economides, D. L., Sabin, C. A. et al. (1998). Frequency of inherited bleeding disorders in women with menorrhagia. Lancet 351 (9101): 485–489. James, A. H., Kouides, P. A., Abdul‐Kadir, R. et al. (2011). Evaluation and management of acute menorrhagia in women with and without underlying bleeding disorders: consensus from an international expert panel. Eur. J. Obstet. Gynecol. Reprod. Biol. 158 (2): 124–134. Davies, J. and Kadir, R. A. (2017). Heavy menstrual bleeding: an update on management. Thromb. Res. 151: S70–S77. Tosetto, A., Castaman, G., and Rodeghiero, F. (2013). Bleeders, bleeding rates, and bleeding score. J. Thromb. Haemost. 11 (Suppl.1): 142–150. Rodeghiero, F., Tosetto, A., Abshire, T. et al. (2010). ISTH/SSC bleeding assessment tool: a standardized questionnaire and a proposal for a new bleeding score for inherited bleeding disorders. J. Thromb. Haemost. 8 (9): 2063–2065. Peyvanidi, F., Garagiola, I., and Menegatti, M. (2011). Gynecological and obstetrical manifestations of inherited bleeding disorders in women. J. Thromb. Haemost. 9: 236–245.

11

12

Inherited Bleeding Disorders in Women

21 Lavee, O. and Kidson‐Gerber, G. (2016).

22

23

24

25

26

Update on inherited disorders of haemostasis and pregnancy. Obstet. Med. 9 (2): 64–72. James, A. H. and Jamison, M. G. (2007). Bleeding events and other complications during pregnancy and childbirth in women with von Willebrand disease. J. Thromb. Haemost 5 (6): 1165–1169. Govorov, I., Löfgren, S., Chaireti, R. et al. (2016). Postpartum hemorrhage in women with Von Willebrand disease – a retrospective observational study. PLoS One 11 (10): e0164683. Kadir, R. A., Kingman, C. E. C., Chi, C. et al. (2007). Is primary postpartum haemorrhage a good predictor of inherited bleeding disorders? Haemophilia 13 (2): 178–181. Harrison, P., Mackie, I., Mumford, A. et al. (2011). Guidelines for the laboratory investigation of heritable disorders of platelet function. Br. J. Haematol. 155 (1): 30–44. Blombäck, M., Konkle, B. A., Manco‐ Johnson, M. J. et al. (2007). Preanalytical conditions that affect coagulation testing,

27

28

29

30

31

32

33

including hormonal status and therapy. J. Thromb. Haemost. 5 (4): 855–858. Magnette, A., Chatelain, M., Chatelain, B. et al. (2016). Pre‐analytical issues in the haemostasis laboratory: guidance for the clinical laboratories. Thromb. J. 14. Alperin, J. (1982). Estrogens and surgery in women with von Willebrand’s disease. Am. J. Med. 73 (3): 367–371. Franchi, F., Biguzzi, E., Martinelli, I. et al. (2013). Normal reference ranges of antithrombin, protein C and protein S: effect of sex, age and hormonal status. Thromb. Res. 132 (2): e152–e157. Sandset, P. M. (2013). Mechanisms of hormonal therapy related thrombosis. Thromb. Res. 131: S4–S7. Palla, R., Peyvandi, F., and Shapiro, A. D. (2015). Rare bleeding disorders: diagnosis and treatment. Blood 125 (13): 2052–2061. Nogami, K. (2016). The utility of thromboelastography in inherited and acquired bleeding disorders. Br. J. Haematol. 174 (4): 503–514. Westbury, S. K. and Mumford, A. D. (2016). Genomics of platelet disorders. Haemophilia 22: 20–24.

13

2 Bleeding Assessment Tools Sarah H. O’Brien 1 and Paula D. James 2 1

Nationwide Children’s Hospital/The Ohio State University College of Medicine, Division of Pediatric Hematology/Oncology, Columbus, OH, USA 2 Queen’s University, Department of Medicine, Kingston, ON, Canada

2.1 ­Introduction Globally, the World Federation of Hemophilia estimates that 1 in 1000 indi­ viduals has an inherited bleeding disorder. Far fewer are ever diagnosed, however. Women are at particular risk of going undi­ agnosed because of a lack of understanding about normal versus abnormal bleeding, particularly for menstruation and postpar­ tum bleeding. Women bleed physiologically, therefore evaluation tools are beneficial when assessing bleeding symptoms. With only subjective interpretation, abnormal bleeding can be overlooked or interpreted as normal. In contrast, given the high fre­ quency of bleeding symptoms reported by the general population, normal symptoms can be given undue attention [1, 2]. A standardized approach has been devel­ oped in the form of bleeding assessment tools (BATs), which have been developed for both the general assessment of hemorrhagic symptoms and the assessment of women’s bleeding specifically. These tools have been tested in both primary care and referral ­populations, for pediatrics and adults, and have been validated for the diagnosis of von  Willebrand disease (VWD) and in the

assessment of hemophilia carriers. In this chapter, we will review the development of BATs and their evolution over time, as well as published validation studies and the current clinical utility of BATs in the evaluation of women’s bleeding.

2.2 ­Evolution of Vicenza‐Based Bleeding Assessment Tools In 1995, Šrámek et  al. reported a study that used a bleeding questionnaire mailed to patients who had a known bleeding disorder (70% of whom had VWD) in addition to patients who had bleeding symptoms but no laboratory diagnosis, and a group of healthy volunteers [3]. The most discriminatory fea­ tures were questions about bleeding follow­ ing traumatic events such as tonsillectomy/ adenoidectomy or dental extraction (but not childbirth) and questions about family mem­ bers with a known bleeding disorder. The data showed that while this questionnaire had high discriminatory power as a screening tool, it had low performance in a referral set­ ting. A decade later, the International Society on Thrombosis and Haemostasis (ISTH) Scientific and Standardization Committee

Inherited Bleeding Disorders in Women, Second Edition. Edited by Rezan A. Kadir, Paula D. James, and Christine A. Lee. © 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.

14

Inherited Bleeding Disorders in Women

Vicenza 2005 0 to + 3 40 minutes

MCMDM1VWD 2006 −1 to + 4 40 minutes

Condensed MCMDM1VWD 2008 −1 to + 4 5–10 minutes

PBQ 2009 −1 to + 4 20 minutes

ISTH BAT 2010 0 to + 4 20 minutes

Self-BAT 2013 0 to + 4 5–10 minutes

Self-PBQ 2017 0 to + 4 10 minutes

Figure 2.1  The evolution of the different Vicenza‐based bleeding scores from the original Vicenza bleeding score on the left and the subsequent versions. For each bleeding score, the year of publication of the original paper, scoring system used and the approximate administration time are shown [5–8, 11–13].

(SSC) on von Willebrand factor (VWF) pro­ posed consensus criteria for the diagnosis of type 1 VWD, which included guidelines for considering mucocutaneous bleeding symp­ toms as significant [4]. Since then, a number of researchers have developed and validated tools for the quantitative assessment and standardization of bleeding symptoms; the Vicenza‐based tools have been largely focused on VWD. The evolution of the Vicenza‐based bleeding scores (BSs) is shown in Figure  2.1 and reviewed below (see also Table 2.1). Tools specifically designed and tested for menor­ rhagia will be reviewed in the next section.

the bleeding severity: 0 for absent or trivial symptoms and up to 3 for symptoms that required medical intervention. The overall bleeding score was then determined by sum­ ming the severity of all bleeding symptoms reported by the patient. The results of this retrospective study showed that having at least three hemorrhagic symptoms or a bleeding score of 3 in males and 5 in females had a high specificity of 98.6% for type 1 VWD, but had a lower sensitivity (69.1%). The negative predictive value (NPV) was 0.99 and the positive predictive value (PPV) was 0.33.

2.2.1  Vicenza Bleeding Questionnaire and Score

2.2.2  Molecular and Clinical Markers for the Diagnosis and Management of Type 1 VWD (MCMDM‐1VWD)

In 2005, Rodegheiro et  al. in Vicenza, Italy, developed a bleeding questionnaire for the diagnosis of type 1 VWD [5]. Study partici­ pants were asked about a number of bleeding symptoms including epistaxis, menstrual bleeding, and postsurgical bleeding. For each symptom, a score was given depending on

A few years later, changes were made to the scoring of the original Vicenza Bleeding Questionnaire in an attempt to increase the sensitivity of the bleeding score for the diagnosis of VWD. The range of possible ­

20

Inherited Bleeding Disorders in Women

scores was increased to include −1, which was scored if bleeding was absent after significant hemostatic challenges such as two or more dental extractions, surgeries, or childbirth for women, and up to 4, scored for significant medical interventions such as blood transfu­ sion, infusion of clotting factor concentrates or surgery to control bleeding [6]. The revised questionnaire and bleeding score were used to evaluate the bleeding severity in type 1 VWD families enrolled in the European Molecular and Clinical Markers for the Diagnosis and  Management of Type 1 von Willebrand Disease (MCMDM‐1VWD) study. Overall, there was a strong inverse correlation between bleeding score and VWF level (P 13 years with VWD

Case series

47%

Perry and Alving 1990

36 women with VWD

Case series

42%

Greer et al. 1991

8 women with VWD (type 1 and 2)

Case series

100%

Kirtava et al. 2003

102 women with VWD registered at US HTC and 88 controls

Case control study

95% among women with VWD; 61% among controls

De Wee et al. 2011

423 women >16 years with moderate and severe VWD

Nationwide cross‐sectional study

81% with >2 HMB symptoms

Foster et al. 1995

30 women with VWD unresponsive to DDAVP in an international registry

Case series

80% with at least one episode requiring blood product

Federici 2004

Women with VWD in Italian registry

Case series

32% with type 1; 32% with type 2; 56% with type 3

Mauser Bunschoten et al. 1988

102 carriers of hemophilia A and 19 carriers of hemophilia B

Case series

31% in HA; 10% in HB

Greer et al. 1991

18 carriers of hemophilia A and five carriers of hemophilia B

Case series

22% in HA; 40% in HB

Kadir et al. 1999

30 carriers of hemophilia recruited from a HTC

Prospective cohort

57% with PBAC score >100

Plug et al. 2006

274 carriers of hemophilia and 245 controls

Postal survey

23% consulted with GP for HMB; 20% in controls

Meisbach et al. 2011

46 carriers of hemophilia A

Cohort

50%

Bolton‐Maggs et al. 1995

46 women with FXI deficiency

Questionnaire study

41% with symptoms indicating HMB

Brenner et al. 1997

82 women with FXI deficiency

Case series

12% with prolonged menstrual bleeding (>7 days)

Hemophilia carriers

Factor FXI deficiency

(Continued)

49

50

Inherited Bleeding Disorders in Women

Table 4.1  (Continued) Study

Sample size and population

Type of study

Prevalence

Kadir et al. 1999

20 women with FXI deficiency recruited from a HTC

Prospective cohort

59% with PBAC score >100

Platelet function disorders Lopez et al. 1998

35 women with Bernard–Soulier syndrome

Summary of case reports

51%

George et al. 1990

55 women with Glanzmann thrombasthenia

Summary of case reports

98%

McKay et al. 2004

Three out of six in affected women with Quebec platelet disorder

Case series

50%

DDAVP, desmopressin; FXI, factor XI; GP, glycoprotein; HA, hemophilia A; HB, hemophilia B; HMB, heavy menstrual bleeding; HTC, hemophilia treatment center; PBAC, pictorial blood assessment chart; VWD, von Willebrand disease.

higher in women with VWD, carriers of hemophilia, and women with FXI deficiency. A PBAC score of >100 was reported in 74%, 57%, and 59% of women with VWD, hemophilia carriers, and women with FXI deficiency, respectively [18]. In a systematic review of the literature, HMB was reported in 35–98% of women with women with rare bleeding disorders, including deficiency in prothrombin, fibrinogen, factors V, VII, X, and XIII [19]. In women with mild inherited platelet function defects, there are no data on the prevalence of HMB, but high frequency (51–98%) has been reported in women with severe platelet function disorders such as Bernard–Soulier syndrome and Glanzmann thrombasthenia. These women typically present with severe HMB in adolescence [14]. Menstruation has a major influence on women’s lifestyle, education, and employment. Quality of life during menstruation has been shown to be worse in women with bleeding disorders compared to controls. All scales of health‐related quality of life are reduced in women with HMB and IBD, with pain, mental health, vitality, and social functioning being the worst affected parameters [20]. Amongst individuals with VWD, women are less likely than men to undertake postsecondary (college or university) education (5.5% versus 35%) [21]. It is possible that this burden of morbidity reflects the effect of

HMB directly, by physical and social limitations, or indirectly because of chronic iron deficiency secondary to HMB. 4.2.1 Investigations Heavy menstrual bleeding in women with disorders of hemostasis may be due to the underlying hematological problem, but not necessarily exclusively. HMB may be multifactorial or purely due to local gynecological causes in these women. Thus, a full gynecological evaluation is required prior to instigating treatment. This usually entails pelvic examination, cervical smear test, and transvaginal pelvic ultrasound. Hysteroscopy is required if ultrasound examination is inconclusive or to determine the presence and location of uterine fibroid or polyp. The possibility of endometrial hyperplasia or malignancy must be excluded  by histological sampling of the endometrium, especially in at‐risk women, such as those aged 45 or more, not responding to treatment or those with irregular pattern of bleeding. The initial investigations should include a full blood count and ferritin level to exclude  iron deficiency anemia. Correction of iron depletion and treatment for anemia are important aspects of management. Correction of iron status alleviates s­ ymptoms,

Gynecology

leads to improvement of quality of life, and reduces the requirement for blood transfusion [22, 23]. The risk of bleeding complications during invasive investigations must always be assessed, and need for hemostatic prophylaxis including blood products should be considered. 4.2.2  Management of Heavy Menstrual Bleeding in Women with IBD Ideally, women with bleeding disorders should be managed in a multidisciplinary clinic, including both hematologist and gynecologist, within the network of hemophilia treatment centers. This ensures provision of comprehensive care with accurate on‐site hemostasis testing and availability of appropriate hemostatic agents when required. In a survey by the Centers for Disease Control

and Prevention (CDC) in the US, 95% (71 out of 75) of women receiving care in hemophilia treatment centers reported a strong positive opinion and satisfaction [24]. Figure 4.2 presents an algorithm for management of HMB in women with bleeding disorders. 4.2.2.1  Medical Treatment of Heavy Menstrual Bleeding

Non‐surgical management for HMB entails both hormonal and hemostatic therapies. Non‐surgical therapies allow a woman to retain reproductive potential, improve quality of life, and reduce the burden of unnecessary surgical procedures. 4.2.2.1.1  Hormonal Treatments Levonorgestrel‐Releasing Intrauterine System

The levonorgestrel‐releasing intrauterine system (LNG‐IUS) provides the most effective non‐surgical treatment for HMB, and is

Would the patient like to preserve fertility?

Yes

No

Would the patient like to become pregnant now?

Yes

Hemostatic measures 1. Anti-fibrinolytic therapy a. tranexamic acid b. epsilon amino caproic acid 2. DDAVP a. intranasal b. subcutaneous

No

Hormonal measures 1. Levonorgestrel IUS 2. Combined oral contraceptive 3. Progestins 4. GNRH therapy with add-back therapy

Figure 4.2  Algorithm of management of heavy menstrual bleeding.

Can also consider surgical options 1. Hysterectomy 2. Endometrial ablation

51

Inherited Bleeding Disorders in Women

considered first‐line treatment if contraception is required. It is an intrauterine device that steadily releases 20 μg of levonorgestrel into the endometrial cavity per 24 hours over a recommended duration of five years. The LNG‐IUS suppresses endometrial growth, causing the glands of the endometrium to become atrophic, thus reducing menstrual loss. The data on efficacy are evident from numerous randomized controlled trials (RCTs) and systematic reviews [25]. The mean reported reduction in PBAC scores from 17 RCTs involving 712 patients and 10 non‐randomized trials involving 380 patients exceeded 70% during the first three months of treatment, with further reductions during the first year of treatment that were sustained throughout the four years of use. The need for further surgical intervention, although inconsistently reported across studies, was around 9% [25]. The LNG‐IUS is also an effective reversible contraception method. It is an attractive treatment option for women who require contraception and wish to preserve their fertility and for those who are unlikely to comply with drug treatments. The LNG‐IUS has also been shown to provide effective long‐term treatment for HMB in women with bleeding disorders not responding to other medical treatments [26]. The median PBAC score decreased significantly, from 255 before insertion to 35 with the LNG‐IUS (Figure  4.3), and 42% of

women were amenorrheic at a median of three months (range 0–24) following insertion of the LNG‐IUS. There was also a significant improvement in hemoglobin level, menstrual pain, and quality of life among IUS users. The main adverse side‐effects reported with the use of the LNG‐IUS include irregular spotting and hormonal symptoms, breast tenderness, abdominal/pelvic pain, and headache. A high discontinue rate due to unscheduled bleeding, pain, and/or systemic progestogenic adverse effects is reported in women using the LNG‐IUS for contraception. However, the LNG‐IUS has high rates of patient satisfaction and tolerability when it is used for treatment of HMB. In a systematic review, the one‐year continuation rate of LNG‐IUS to treat HMB was 79%, reported from 14 studies [25]. In addition, proper patient assessment, counseling, and education regarding these side‐effects can improve patient tolerance and continuation rate. Expulsion of the IUS from the uterine cavity can occur after insertion, usually within the first six weeks. Women are encouraged to check the threads by digital examination or, if they are not happy to do so, by speculum examination six weeks following insertion to ensure the IUS is correctly in place. Uterine perforation is a rare but potentially serious complication of LNG‐IUS insertion, occurring in 1 in 1000 cases.

PBAC Scores before and with Mirena

PBAC score

52

450 400 350 300 250 200 150 100 50 0

Mean PBAC Pre insertion 299 At follow-up

Pre-insertion

24

At follow-up

Figure 4.3  Levonorgestrel‐IUS and PBAC scores: long‐term follow‐up ≥24 months after insertion.

Gynecology

Combined Hormonal Contraceptives

Combined hormonal contraceptives (COCs) include combined oral contraceptive pills (OCPs), transdermal contraceptive patches, and vaginal rings. COCs are a highly reliable method of contraception. They reduce MBL by inducing shorter, regular shedding of a thinner endometrium, as well as reducing menstrual pain or dysmenorrhea. A systematic review of eight studies among 438 women, including six RCTs (five with the OCP, one with the vaginal ring), assessed the effect of COCs in treatment of HMB [27]. All six studies reported a reduction in menstrual loss (measured through either the alkaline hematin method or PBAC score) following seven cycles or six months of treatment [27]. However, hormonal side‐effects including breast tenderness, headache, mood changes/ depression, nausea/vomiting, and weight gain resulted in one‐year continuation rates that ranged between 72% and 84% [27]. The new estradiol valerate and dienogest (E2V/ DNG) contraceptive pill is the only formulation that has provided sufficient evidence to receive US Food and Drug Administration and European Union approval to treat HMB. A pooled analysis from two placebo‐controlled trials of E2V/DNG in 269 women with HMB showed a significant reduction in MBL after six months of use compared to baseline (88% in the E2V/DNG treatment group versus 24% for placebo) [27]. The efficacy of COCs in reducing MBL in women with bleeding disorders is not well studied. In a survey of women with VWD (type 2 and 3) unresponsive to 1‐desamino‐ 8‐D‐arginine vasopressin (DDAVP), the use of COC was reported to be effective in controlling HMB in 88% [15]. On the other hand, in type 1 patients, a standard‐dose and a higher‐dose COC were effective in only 24% and 37% of cases, respectively [16]. COCs suppress ovulation so in women with bleeding disorders, they have the added advantage of preventing ovulation bleedings that can be recurrent and potentially life‐ threatening, especially in those with severe disorders.

Increased risk of thrombosis is the main concern associated with the use of COCs. However, women with bleeding disorders are at low risk for thrombotic complications. Serious side‐effects of COCs include hypertension and, rarely, impaired liver ­ function and hepatic tumors. Other less serious side‐effects include nausea, vomiting, headache, breast tenderness, breakthrough bleeding, fluid retention, depression, and skin reactions. Extended‐use COC, which entails continous use of levonorgestrel and low‐dose ethinyl estradiol without the seven‐day break for a duration of three months, has been evaluated, with amenorrhea reported in 79% of women [28]. While unscheduled (breakthrough) bleeding is reported frequently with extended COC use, it decreases with each successive cycle of therapy. Progestogens

Oral progestogens are one of the most commonly prescribed medications for HMB. There are two different cyclical regimes of oral progestogens: a short luteal phase treatment (days 19–26 or days 15–25) or a longer 21‐day course starting from day 5 of the cycle. The available data on short luteal phase oral progestogens show limited efficacy in reducing MBL. In a systematic review, involving 157 patients in four studies, the median MBL reduction ranged between 2% and 30% with a six‐month treatment course [25]. Therefore this regime should not be used for the treatment of HMB. In contrast, the longer course 21‐day oral progestogens consistently reduced PBAC scores in the systematic review among 178 women [25], with a median reduction in MBL of 63%, and 78% during cycles 1 and 3 of treatment. Adverse side‐effects reported with oral progestogens include headaches, breast tenderness, and erratic bleeding problems. Of note, none of the studies included in the systematic review were placebo controlled , and thus it was not possible to ascertain whether these side‐effects could be attributed to the nocebo effect [25].

53

54

Inherited Bleeding Disorders in Women

However, the discontinuation rate with long‐phase cyclical oral progestogen treatment is high (78% after three months), suggesting that tolerability is a problem with this method of treatment. Cyclical 21‐day progesterone therapy can be considered as a second‐line treatment for patients who do not respond to other medical therapies previously discussed or in whom such treatments are contraindicated. Intramuscular or subcutaneous injection of high‐dose depot medroxyprogesterone acetate (DMPA) given every 12 weeks can induce amenorrhea in up to 50% of women. However, due to follicular suppression, long‐ term estrogen production is reduced, resulting in reduction in bone mineral density and increased risk of fracture, limiting its use in adolescent girls and perimenopausal women with HMB. The progesterone‐only subdermal implant is a highly effective form of long‐ acting contraception that alters menstrual bleeding patterns and may result in amenorrhea in some women. However, its use as a treatment for HMB has not been studied. In addition, these modes of administration can be associated with a risk of excessive bruising and hematoma in women with bleeding disorders, especially severe forms. The progestin‐only “mini‐pill,” containing 75  μg desogestrel (Cerazette®), may induce amenorrhea in up to 20% of users [29]. However, it is associated with unpredictable and irregular blood loss and not usually considered as a first‐line treatment for HMB, but may be used in combination with other therapies if no other options are acceptable. Danazol and Gonadotrophin‐Releasing Hormone Agonist

Danazol, a synthetic androgen that has antiestrogen and antiprogestogen actions, is used primarily for the treatment of endometriosis. Although effective in reducing MBL, it is rarely used for the treatment of HMB due to perceived and real adverse effects including low mood, irritability, weight gain, hirsutism, and irreversible voice changes. Gonadotrophin‐releasing hormone (GnRH)

agonists cause ovarian suppression and amenorrhea but are not used for the treatment of HMB due to significant side‐effects. However, due to high efficacy, their use with simultaneous hormone replacement therapy (add‐back therapy) with combined estrogen/ progesterone or tibolone may be an alternative option to surgery for women with severe bleeding disorders such as type 3 VWD or Glanzmann thrombasthenia not responding to other treatments. Ulipristal Acetate

Ulipristal acetate (UPA) (Esmya®) is a selective progesterone receptor modulator that is clinically proven to reduce the size of uterine fibroids, reduce menstrual loss, or induce amenorrhea in women with fibroids. In the PEARL IV study, an RCT involving 46 centers across Europe, 451 women with fibroids and HMB were administered 5 or 10 mg UPA per day and matching placebos for up to four 12‐week courses. Both treatment doses were associated with a significant reduction in MBL and high amenorrhea rate. In addition, there was significant reduction in fibroid volume from baseline, and the percentage of women with fibroid volume reduction >50% increased from course 1 to course 4 in both treatment groups [30]. Common side‐effects reported included headache and breast tenderness. Abnormal bleeding (prolonged, frequent, or irregular) was reported in women with submucous fibroids, which could limit the efficacy of UPA at controlling bleeding. Endometrial thickness increased after the first course, but returned to below screening levels in subsequent treatment courses. UPA did not increase the occurrence of endometrial features of concern. Non‐physiological changes were observed in 18% and 13% of endometrial biopsies taken after treatment courses 2 and 4, respectively, and were reversible after treatment cessation [30]. In a recent study in women treated with UPA prior to hysterectomy, UPA administration altered expression of sex steroid receptors and progesterone‐regulated genes and was associated with low levels of glandular

Gynecology

and stromal cell proliferation. This indicates that UPA could potentially be useful for the treatment of HMB even in women without uterine fibroids [31]. Due to recent concerning reports regarding rare but serious risk of liver injury, the European Medicines Agency recommend UPA is contraindicated in women with known liver disease, and liver tests are recommended in all women before, during and after stopping treatment with Esmya. 4.2.2.1.2  Non‐Hormonal Treatments Antifibrinolytic Therapy (Tranexamic Acid)

Tranexamic acid is an antifibrinolytic agent that reversibly blocks lysine binding sites on plasminogen and prevents fibrin degradation. There is increased fibrinolysis in the endometrium in women with HMB. Tranexamic acid treatment has been shown to significantly reduce endometrial tissue plasminogen activator (t‐PA) and plasmin activity in the menstrual as well as the peripheral blood of women with HMB [32]. In a Cochrane systematic review [33], two trials showed a significant reduction in MBL with antifibrinolytic therapy compared to placebo. Tranexamic acid was also shown to be superior to non‐steroidal anti‐inflammatory drugs (NSAIDs) in the reducing MBL. Oral tranexamic acid is generally well tolerated by women with HMB. Nausea and diarrhea are the most common side‐effects. Adverse events with tranexamic acid treatment for HMB were not higher than with placebo, NSAIDs, or cyclic progestogens in the Cochrane systematic review [33]. There have been isolated reports of thromboembolic complications with the use of tranexamic acid, which have led to reluctance to use it. However, the incidence of thrombosis over 19 years and 238  000 patient‐years of treatment with tranexamic acid was shown to be similar to the spontaneous frequency of thrombosis in women in the general population [34]. A Royal College of Obstetricians and Gynaecologists’ guideline (in collaboration with NICE) in the UK recommends tranexamic acid for three

months as a first‐line medical treatment for HMB [5]. If successful, tranexamic acid can be used indefinitely in patients not requiring contraception or who prefer non‐hormonal treatment. In women with IBDs, tranexamic acid is widely used (orally, intravenously, or ­topically alone or as an adjuvant therapy) in the prevention and management of oral cavity bleeding, epistaxis, gastrointestinal bleeding, and HMB. In a recent Cochrane review that assessed medical therapies for treatment of HMB in three cross‐over RCTs involving 175 women with IBD, tranexamic acid was superior to DDAVP for reduction of MBL (mean difference 42  mL, 95% CI 19.6–63  mL, P 6 days (odds ratio (OR) 2.5; 95% CI 1.1–5.9) [54]. To date, there is a lack of research to evaluate the link between endometriosis and IBD. In a case control study conducted at the Royal Free Hospital, women with endometriosis had a higher frequency of platelet aggregation defects (31%) compared to controls (4%), and women with severe (stage IV) endometriosis diagnosed at laparoscopy had significantly reduced VWF activity level [55]. In a study that assessed the reproductive experience of women with VWD, endometriosis was reported in 30% of women with VWD compared to 13% of controls [17]. In a survey that interviewed 168 women, dysmenorrhea and its interference with daily life were reported significantly more often in women with bleeding disorders compared to controls. Of the 99 women with IBD included in the survey, 50% reported moderate, severe or very severe pain during their menses [56]. Surveillance data from the US CDC reported that over 50% of 217 women with IBD experienced dysmenorrhea, and the prevalence of a confirmed diagnosis of endometriosis in these women was 13% [57]. The findings of these studies could have implications for the pathogenesis of endometriosis – increased retrograde menstruation promoting endometriosis formation and impaired local hemostasis within endometriotic implants resulting in recurrent cyclical internal bleeding. Further research is required to determine the association of local or systemic hemostatic defects with endometriosis and whether hemostatic therapies have any beneficial effect in alleviating endometriosis symptoms or on disease progression.

4.6 ­Other Gynecological Conditions Apart from HMB and ovulation bleeding, there is no evidence of a higher prevalence of other gynecological problems. However, most gynecological conditions present with bleeding or symptoms secondary to bleeding, so women with these disorders are more likely to be symptomatic with gynecological conditions. Uterine polyps, uterine fibroids, and endometrial hyperplasia can present with HMB and/or irregular intermenstrual bleeding. A case control study of 102 women with VWD reported a higher prevalence of endometrial polyps (8%) and endometrial hyperplasia (10%) in women with VWD versus 1% in controls [17]. Presence of these gynecological conditions may unmask a previously undiagnosed bleeding disorder. In the perimenopausal period, many women experience significant vasomotor symptoms. Estrogen‐containing hormone replacement therapy (HRT) is effective but unscheduled or irregular vaginal bleeding is commonly encountered in HRT users. HRT may induce increased vascular fragility in endometrial blood vessels and stroma leading to such bleeding [58]. This vascular fragility is more likely to manifest as abnormal bleeding in women with IBD due to their deficient hemostasis, leading to inconvenience and possibly unnecessary repeated invasive investigations to exclude underlying endometrial malignancy. Postmenopausal bleeding (PMB) is another common gynecological symptom and accounts for 5% of referrals to gynecology outpatient clinics. Endometrial hyperplasia/ cancer is the underlying pathology in up to 10% of cases. However, the most common cause for PMB is atrophy of the genital tract which accounts for 60–80% of cases. There are no data in the literature on PMB in women with bleeding disorders, but they may be more likely to present with PMB with any of the underlying causes. However,

Gynecology

bleeding disorders should not be presumed to be the underlying cause for PMB or abnormal bleeding with HRT use. Management of these women requires urgent referral and investigation to rule out malignancy and provide appropriate treatment.

4.7 ­Conclusion Women with IBD are more likely to suffer from gynecological disorders due to the bleeding nature of these conditions. They have a high prevalence of HMB, including acute HMB and HMB from adolescence, which is often the presenting symptom. The presentation of HMB and a positive bleeding

history can be used as a screening tool to identify women with IBD. Women with IBD can present with HMB at perimenopausal age due to concomitant presence of gynecological pathology, which requires full investigation prior to instigating treatment. Other presentations of gynecological disorders, including hemorrhagic cysts and ruptured ovarian cyst with life‐threatening hemoperitoneum, require multidisciplinary collaboration, especially in severe and rare cases, as these presentations can be challenging to manage. Early recognition, accurate diagnosis, and appropriate management of bleeding disorders should improve not only the quality of care for these women but also their quality of life.

­References 1 UKHCDO (2016). A Report from the UK

2

3

4

5

6

National Haemophilia Database. London: UK Haemophilia Centres Doctors’ Organisation. Chi, C., Shiltagh, N., Kingman, C. et al. (2006). Identification and management of women with inherited bleeding disorders: a survey of obstetricians and gynaecologists in the United Kingdom. Haemophilia 12 (4): 405–412. Hallberg, L. and Nilsson, L. (1964). Determination of menstrual blood loss. Scand. J. Clin. Lab. Invest. 16: 244–248. Warner, P.E., Critchley, H., Lumsden, M. et al. (2004). Menorrhagia II: is the 80‐mL blood loss criterion useful in management of complaint of menorrhagia? Am. J. Obstet. Gynecol. 190 (5): 1224–1229. NICE (2016). Clinical Guideline [CG44]. Heavy Menstrual Bleeding: Assessment and Management. London: National Institution for Health and Clinical Excellence. Higham, J.M., O’Brien, P.M., and Shaw, R.W. (1990). Assessment of menstrual blood loss using a pictorial chart. Br. J. Obstet. Gynaecol. 97 (8): 734–739.

7 Beksinska, M.E., Smit, J., Greener, R. et al.

8

9

10

11

(2015). Acceptability and performance of the menstrual cup in South Africa: a randomized crossover trial comparing the menstrual cup to tampons or sanitary pads. J. Women’s Health (Larchmt) 24 (2): 151–158. Shankar, M., Lee, C., Sabin, C. et al. (2004). Von Willebrand disease in women with menorrhagia: a systematic review. Br. J. Obstet. Gynaecol. 111 (7): 734–740. Miller, C.H., Philipp, C., Stein, S. et al. (2011). The spectrum of haemostatic characteristics of women with unexplained menorrhagia. Haemophilia 17 (1): e223–e229. Kouides, P.A., Conard, J., Peyvandi, F. et al. (2005). Hemostasis and menstruation: appropriate investigation for underlying disorders of hemostasis in women with excessive menstrual bleeding. Fertil. Steril. 84 (5): 1345–1351. ACOG (2013). ACOG Committee Opinion No. 557: management of acute abnormal uterine bleeding in nonpregnant reproductive‐aged women. Obstet. Gynecol. 121 (4): 891–896.

61

62

Inherited Bleeding Disorders in Women

12 NICE (2007). Heavy Menstrual Bleeding:

13

14

15

16

17

18

19

20

21

Assessment and Management. London: National Institute for Health and Clinical Excellence. Kadir, R.A., Economides, D., Sabin, C. et al. (1998). Frequency of inherited bleeding disorders in women with menorrhagia. Lancet 351 (9101): 485–489. Davies, J. (2016). Women with Inherited Bleeding Disorders and Their Offspring – the Unresolved Issues. EGA Institute for Women’s Health. London: University College London. Foster, P.A. (1995). The reproductive health of women with von Willebrand disease unresponsive to DDAVP: results of an international survey. On behalf of the Subcommittee on von Willebrand Factor of the Scientific and Standardization Committee of the ISTH. Thromb. Haemost. 74 (2): 784–790. Kouides, P.A., Phatak, P., Burkart, P. et al. (2000). Gynaecological and obstetrical morbidity in women with type I von Willebrand disease: results of a patient survey. Haemophilia 6 (6): 643–648. Kirtava, A., Drews, C., Lally, C. et al. (2003). Medical, reproductive and psychosocial experiences of women diagnosed with von Willebrand’s disease receiving care in haemophilia treatment centres: a case‐control study. Haemophilia 9 (3): 292–297. Kadir, R.A., Economides, D., Sabin, C. et al. (1999). Assessment of menstrual blood loss and gynaecological problems in patients with inherited bleeding disorders. Haemophilia 5 (1): 40–48. James, A.H. (2005). More than menorrhagia: a review of the obstetric and gynaecological manifestations of bleeding disorders. Haemophilia 11 (4): 295–307. Shankar, M., Chi, C., and Kadir, R. (2008). Review of quality of life: menorrhagia in women with or without inherited bleeding disorders. Haemophilia 14 (1): 15–20. Barr, R.D., Sek, J., Horsman, J. et al. (2003). Health status and health‐related quality of

22

23

24

25

26

27

28

29

life associated with von Willebrand disease. Am. J. Hematol. 73 (2): 108–114. Liu, Z., Doan, Q., Blumenthal, P., and Dubois, R. (2007). A systematic review evaluating health‐related quality of life, work impairment, and health‐care costs and utilization in abnormal uterine bleeding. Value Health 10 (3): 183–194. Froessler, B., Palm, P., Weber, I. et al. (2018). The important role for intravenous iron in perioperative patient blood management in major abdominal surgery: a randomized controlled trial. Ann. Surg. 267 (2): e39–e40. Kirtava, A., Crudder, S., Dilley, A. et al. (2004). Trends in clinical management of women with von Willebrand disease: a survey of 75 women enrolled in haemophilia treatment centres in the United States. Haemophilia 10 (2): 158–161. Bitzer, J., Heikinheimo, O., Nelson, A. et al. (2015). Medical management of heavy menstrual bleeding: a comprehensive review of the literature. Obstet. Gynecol. Surv. 70 (2): 115–130. Chi, C., Huq, F.Y., and Kadir, R.A. (2011). Levonorgestrel‐releasing intrauterine system for the management of heavy menstrual bleeding in women with inherited bleeding disorders: long‐term follow‐up. Contraception 83 (3): 242–247. Fraser, I.S., Parke, S., Mellinger, U. et al. (2011). Effective treatment of heavy and/or prolonged menstrual bleeding without organic cause: pooled analysis of two multinational, randomised, double‐blind, placebo‐controlled trials of oestradiol valerate and dienogest. Eur. J. Contracept. Reprod. Health Care 16 (4): 258–269. Archer, D.F., Jensen, J., Johnson, J. et al. (2006). Evaluation of a continuous regimen of levonorgestrel/ethinyl estradiol: phase 3 study results. Contraception 74 (6): 439–445. Benagiano, G. and Primiero, F.M. (2003). Seventy‐five microgram desogestrel minipill, a new perspective in estrogen‐free

Gynecology

30

31

32

33

34

35

36

37

38

39

contraception. Ann. N.Y. Acad. Sci. 997: 163–173. Donnez, J., Donnez, O., Matule, D. et al. (2016). Long‐term medical management of uterine fibroids with ulipristal acetate. Fertil. Steril. 105 (1): 165–173 e4. Whitaker, L.H., Murray, A., Matthews, R. et al. (2017). Selective progesterone receptor modulator (SPRM) ulipristal acetate (UPA) and its effects on the human endometrium. Hum. Reprod. 32 (3): 531–543. Dockeray, C.J., Sheppard, B., Daly, L., and Bonnar, J. (1987). The fibrinolytic enzyme system in normal menstruation and excessive uterine bleeding and the effect of tranexamic acid. Eur. J. Obstet. Gynecol. Reprod. Biol. 24 (4): 309–318. Lethaby, A., Farquhar, C., and Cooke, I. (2000). Antifibrinolytics for heavy menstrual bleeding. Cochrane Database Syst. Rev. (4): CD000249. Rybo, G. (1991). Tranexamic acid therapy: effective treatment in heavy menstrual bleeding: clinical update on safety. Therap. Adv. 4: 1–8. Ray, S. and Ray, A. (2016). Non‐surgical interventions for treating heavy menstrual bleeding (menorrhagia) in women with bleeding disorders. Cochrane Database Syst. Rev. (11): CD010338. Lukes, A.S., Moore, K., Muse, K. et al. (2010). Tranexamic acid treatment for heavy menstrual bleeding: a randomized controlled trial. Obstet. Gynecol. 116 (4): 865–875. Hrometz, S.L. (2012). Oral modified‐ release tranexamic acid for heavy menstrual bleeding. Ann. Pharmacother. 46 (7–8): 1047–1053. Kouides, P.A., Byams, V., Philipp, C. et al. (2009). Multisite management study of menorrhagia with abnormal laboratory haemostasis: a prospective crossover study of intranasal desmopressin and oral tranexamic acid. Br. J. Haematol. 145 (2): 212–220. Dunn, A.L., Powers, J., Ribeiro, M. et al. (2000). Adverse events during use of

40

41

42

43

44

45

46

47

48

intranasal desmopressin acetate for haemophilia A and von Willebrand disease: a case report and review of 40 patients. Haemophilia 6 (1): 11–14. Munro, M.G. (2018). Endometrial ablation. Best Pract. Res. Clin. Obstet. Gynaecol. 46: 120–139. Middleton, L.J., Champaneria, R., Daniels, J. et al. (2010). Hysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system (Mirena) for heavy menstrual bleeding: systematic review and meta‐analysis of data from individual patients. BMJ 341: c3929. Cooper, K., Lee, A., Chien, P. et al. (2011). Outcomes following hysterectomy or endometrial ablation for heavy menstrual bleeding: retrospective analysis of hospital episode statistics in Scotland. Br J. Obstet. Gynaecol. 118 (10): 1171–1179. Huq, F.Y., Al‐Haderi, M., and Kadir, R.A. (2012). The outcome of endometrial ablation in women with inherited bleeding disorders. Haemophilia 18 (3): 413–420. Pynna, K., Vuorela, P., Lodenius, L. et al. (2014). Cost‐effectiveness of hysterectomy for benign gynecological conditions: a systematic review. Acta Obstet. Gynecol. Scand. 93 (3): 225–232. Maresh, M.J., Metcalfe, M., McPherson, K. et al. (2002). The VALUE national hysterectomy study: description of the patients and their surgery. Br J. Obstet. Gynaecol. 109 (3): 302–312. James, A.H., Myers, E., Cook, C., and Pietrobon, R. (2009). Complications of hysterectomy in women with von Willebrand disease. Haemophilia 15 (4): 926–931. Leung, P.L., Ng, P., Lok, I., and Yuen, P. (2005). Puberty menorrhagia secondary to inherited bleeding disorders. Acta Obstet. Gynecol. Scand. 84 (9): 921–922. Yoost, J., LaJoie, A., Hertweck, P., and Loveless, M. (2013). Use of the levonorgestrel intrauterine system in adolescents with endometriosis. J. Pediatr. Adolesc. Gynecol. 26 (2): 120–124.

63

64

Inherited Bleeding Disorders in Women

49 DeVore, G.R., Owens, O., and Kase, N.

50

51

52

53

(1982). Use of intravenous Premarin in the treatment of dysfunctional uterine bleeding – a double‐blind randomized control study. Obstet. Gynecol. 59 (3): 285–291. Bowkley, C.W., Dubel, G., Haas, R. et al. (2007). Uterine artery embolization for control of life‐threatening hemorrhage at menarche: brief report. J. Vasc. Intervent. Radiol. 18 (1 Pt 1): 127–131. Sharief, L.A. and Kadir, R.A. (2013). Congenital factor XIII deficiency in women: a systematic review of literature. Haemophilia 19 (6): e349–e357. Vigano, P., Parazzini, F., Somigliana, E., and Vercellini, P. (2004). Endometriosis: epidemiology and aetiological factors. Best Pract. Res. Clin. Obstet. Gynaecol. 18 (2): 177–200. Macer, M.L. and Taylor, H.S. (2012). Endometriosis and infertility: a review of the pathogenesis and treatment of endometriosis‐associated infertility. Obstet. Gynecol. Clin. North Am. 39 (4): 535–549.

54 Darrow, S.L., Vena, J., Batt, R. et al. (1993).

55

56

57

58

Menstrual cycle characteristics and the risk of endometriosis. Epidemiology 4 (2): 135–142. Davies, J., Hussein, B., Riddell, A. et al. (2017). The prevalence of laboratory abnormalities of haemostasis in women with endometriosis: a case‐control study. Thromb. Res. 151 (Suppl 1): S103–S140. Kadir, R.A., Sabin, C., Pollard, D. et al. (1998). Quality of life during menstruation in patients with inherited bleeding disorders. Haemophilia 4 (6): 836–841. Byams, V.R., Kouides, P., Kulkarni, R. et al. (2011). Surveillance of female patients with inherited bleeding disorders in United States Haemophilia Treatment Centres. Haemophilia 17 (Suppl 1): 6–13. Hickey, M. and Ambekar, M. (2009). Abnormal bleeding in postmenopausal hormone users – what do we know today? Maturitas 63 (1): 45–50.

65

5 Carriers of Hemophilia A and Hemophilia B Roseline d’Oiron Reference Centre for Hemophilia and Rare Congenital Bleeding Disorders, University Hospitals Paris Sud, Bicêtre Hospital – APHP, Le Kremlin‐Bicêtre, France

Two main questions have to be answered for women in families with hemophilia. Are they carriers of the mutation responsible for the familial hemophilia, with the associated risk of passing on the affected gene to their offspring, and are they exposed to a higher bleeding risk, as some carriers may have a low factor VIII (FVIII) or factor IX (FIX) level and/or abnormal bleeding tendency? The answers rely on two different types of biological tests, genetic testing and coagulation assays, and need to be addressed in order to organize support for women in families with hemophilia. Based on the results of both genetic testing and factor level assays, three situations can be described for girls or women from families with hemophilia: being a carrier of the F8 or F9 gene mutation with normal factor levels; being a carrier of the F8 or F9 gene mutation with low factor levels; or being negative for the F8 or F9 gene mutation (non‐carrier) and not exposed to low factor levels in relation to the familial history of hemophilia. A range of different medical specialists, including geneticists, hematologists, gynecologists, pediatricians, surgeons, and dentists, may be involved at different key moments of life: menarche, invasive procedures, dental procedures, reproductive choices, and pregnancy, to

provide comprehensive care for women from families with hemophilia.

5.1 ­Inheritance Hemophilia is a hereditary X‐linked bleeding disorder resulting from a deficient or defective coagulation FVIII or FIX due to heterogeneous genetic events in the F8 or F9 genes. Males who have only one allele of the F8 or F9 gene are mostly affected. Females with two alleles may be heterozygous for the mutation and it was assumed until recently that they would have a normal FVIII or FIX provided by the unaffected allele. A female carrier has a 50% probability of transmitting the abnormal allele to her child (Figure 5.1).

5.2 ­Screening for the Genetic Status of Carriers of Hemophilia The techniques used for genetic screening and diagnosis have dramatically improved during the last three decades (see Chapter 10). The majority of the mutations responsible for hemophilia can now be identified by

Inherited Bleeding Disorders in Women, Second Edition. Edited by Rezan A. Kadir, Paula D. James, and Christine A. Lee. © 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.

66

Inherited Bleeding Disorders in Women

Female carrier Female (not carrier) Male with hemophilia Male without hemophilia

Figure 5.1  The inheritance of hemophilia.

direct sequencing of F8 or F9 genes. For carrier testing, even if the DNA of the patient is not available, or if the mutation in the family is unknown, diagnosis of the mutation is feasible in most cases. The most common genetic abnormality in severe hemophilia A is the inversion of intron 22, which is responsible in about 45% of affected families [1]. Currently, almost all other variants can be identified, regardless of the severity and type of hemophilia. The variants are recorded in the European Association for Haemophilia and Allied Disorders (EAHAD) international database that can be accessed at http://www.eahad‐ db.org. Technically, the confirmation of carrier status is therefore straightforward provided that access to molecular testing can be offered. However, in sporadic families, that is, those without history of affected members or carriers, the potential risk of somatic mosaicism

should not be overlooked for counseling ­purposes. Indeed, a study showed varying degrees (0.2–25%) of mosaicism by mutation enrichment procedure in eight (13%) of 61 families with sporadic severe hemophilia A [2]. Mothers of sporadic cases, apparently non‐carriers with standard techniques on white blood cells, may carry uncertainty about the recurrence risk and should be offered counseling accordingly. Another uncommon situation is the discovery of an undescribed variant of the F8 gene in a woman with low FVIII:C levels, without other diagnoses  –  such as von Willebrand disease (VWD) or combined FV‐ FVIII deficiency – and without a familial history of hemophilia A. Predicting the potential hemophilia severity that can be passed on to children for such a woman is more difficult in the absence of a male index case. Although not validated for current care, a research strategy based on a combination of standard in silico bioinformatics approaches and site‐ directed mutagenesis in vitro studies using a model for cellular expression of the variant FVIII proteins may be useful to improve counseling [3].

5.3 ­Confusion Between Genetic and Coagulation Testing It is often unclear from the literature what the terms diagnosis of carrier or carrier testing exactly cover, as some reports imply the genetic diagnosis of the carriership of the F8 or F9 gene mutation, while others refer to diagnosis of the bleeding disorder related to low FVIII or FIX levels expressed by some carriers of hemophilia. Genetic testing and assessment of the bleeding risk are not always performed at the same age. For example, a young girl who is potentially a carrier of hemophilia may be tested for FVIII or IX levels for a preoperative work‐up, a bleeding episode or a specific inquiry because of a family history of hemophilia, but

Carriers of Hemophilia A and Hemophilia B

the genetic testing may be performed after informed consent in adulthood. Alternatively, both coagulation and genetic testing of a potential carrier might be tested simultaneously at a first visit to a hemophilia treatment center when seeking medical counseling before a pregnancy. The lack of testing for a potential bleeding disorder during childhood may not only impair access to care but also contribute to losing the opportunity of contacts between the hemophilia team and females from families with hemophilia and therefore participate in delayed genetic testing and diagnosis of carriership. Confusion in understanding between genetic and coagulation testing is frequent. These designations, although implicitly understood by those treating hemophilia, are not easily differentiated by patients, families, and even non‐specialist healthcare professionals. Some actual carriers may mistakenly believe that they are not carriers based on the erroneous argument of a normal FVIII/C or FIX:C level. In this regard, it is crucial to explain to families and potential carriers that a normal FVIII:C or FIX:C level does not exclude a status of carrier of hemophilia. Only a genetic test can confirm or exclude the diagnosis of carriership. In contrast to the two or three hours needed for coagulation assays, genetic analyses may require days or weeks, especially if the mutation has not been identified in affected male relatives. This may contribute to delayed requests for testing and may jeopardize access to prenatal approaches in the appropriate time frame. The time requested to ascertain diagnosis of the F8 or F9 gene mutation also depends on the local medical environment and resources. A typical situation may be observed in women with a familial history of hemophilia A who are already pregnant but unaware of their carrier status. As the FVIII level will rise quite quickly after the beginning of the pregnancy, the chance to detect a low basal FVIII level is missed. Clarifications on both genetic and coagulation testing for girls and women in families with hemophilia should

be offered by those treating hemophilia to ensure proper understanding of the challenges related to the transmission of the affected gene and to being affected by a bleeding disorder. 

5.4 ­When to Perform Genetic Testing Awareness of carrier status before pregnancy has numerous advantages, including time to cope with the diagnosis, access to information about reproductive options, and proper genetic counseling [4]. Further complex decisions will be facilitated, especially those dealing with prenatal diagnosis (PND) or preimplantation genetic diagnosis (PGD) in families affected by the most severe forms of hemophilia. However, delayed diagnoses of carriership have been frequently reported. According to the literature, 30–90% of carriers are unaware of their genetic status at the time of a first pregnancy and 50% of carriers have not yet been diagnosed at 25 years old [5–8]. Interestingly, it has been reported that carriers unaware of their status were younger for their first pregnancy compared to those who knew their risk of passing on the affected gene (25 versus 29 years, P = 0.03) [7]. This is consistent with a Swedish study about reproductive choices in carriers of hemophilia [9]. Even if tested quite late, about half of carriers believed that the timing of the genetic testing was right for them, as reported in a qualitative study [6]. This is discordant from the parental desire for early carrier testing of girls before 10 years old for 40% of parents [10]. Opinions of families may evolve with time. More recently, a survey performed in the USA showed that 65% of obligate carriers preferred testing to be performed before 14 years of age while 72% of healthcare providers recommended testing after this age [11]. Parental concerns, national guidelines, or regulations regarding genetic testing in ­children may explain why coagulation assays alone are proposed for young potential

67

68

Inherited Bleeding Disorders in Women

c­ arriers. Indeed, carrier testing in children deals notably with fears about psychosocial effects, distortion of perception or stigmatization of the child by the family [6]. One of the main concerns is for the autonomous informed decision of the child and lack of reproductive privacy in the future [12]. An overall agreement in the recommendation is to wait until the child can give informed ­consent and understand the implications of being a carrier [11–15]. In some countries, only adults may have access to genetic testing unless there is a direct benefit for the child. Receiving a diagnosis of carriership is never easy and may imply psychosocial issues that need to be anticipated by the hemophilia treatment center (HTC) team in order to provide adequate support [15]. Adolescents or young adults should receive adequate information from the HTC on the consequences of being a carrier and how this medical condition can be supported  and should be given the choice of when to request testing [14]. Readiness to receive information is very variable and can be different according to age or personality. The time for actual testing can be left pending until the adolescent or adult is willing to face the diagnosis.

5.5 ­What Reasons might Contribute to Delayed Genetic Diagnosis of Carriership? The reasons for delayed diagnosis of carriership are multiple and need to be recognized in order to be addressed. Ignorance or misunderstanding of the possibility of being a carrier is frequently reported and observed in the case of tense or distant family relationships,  as for example where there is difficulty of dialogue between a father with hemophilia and his obligate carrier daughter. Care should be taken during the transition period when young adolescents move to a different center because the adult center may not see the parents and

f­urther updates of the family pedigrees may be more difficult to obtain. Some women know that they might be a carrier of hemophilia but do not appreciate the benefits of a diagnosis performed before a pregnancy. Such women may not be aware that solutions can be implemented to secure their own bleeding risk during delivery and in the postpartum period. Similarly, a large majority are not informed of the risk of intracranial hemorrhage (ICH) after birth for male neonates with hemophilia and the crucial role of precautionary measures to avoid this. Others may avoid genetic testing for fear of facing a definite diagnosis or because it is a painful reminder of the burden of hemophilia in the family. Cultural issues should be considered as carriership diagnosis may affect the family choices (see Chapter  14 and section 5.9.1).

5.6 ­Bleeding Disorders in Carriers of Hemophilia 5.6.1  Why Carriers of Hemophilia may have FVIII or FIX Deficiency Girls and women with affected expression of both X‐chromosomes often have moderate or severe FVIII or FIX deficiencies. This is observed in extremely rare contexts such as homozygous status for a given F8 or F9 gene mutation, a compound heterozygous state where two different mutations are inherited from each parent, a structural or numerical aberration of the X‐chromosome (such as Turner syndrome, Swyer syndrome, translocation), a familial non‐random skewed X‐ chromosome inactivation pattern (XIP) or an extreme lyonization [16, 17]. In mammalian females, X‐chromosome inactivation induces a dosage compensation for X‐linked genes. The phenomenon, called lyonization, is mainly the consequence of a methylation process occurring randomly in early embryogenesis and leading to a roughly 50% expression of each parental allele [18]. Inactivation is initiated from a

Carriers of Hemophilia A and Hemophilia B

X‐linked locus, the X‐inactivation center (Xic), and spreads along the chromosome toward both ends. Once established, inactivity is stably maintained in subsequent cell generations. In general, the cellular mosaicism in expression of the paternal or maternal X‐ chromosome allows expression of normal FVIII or FIX in half of the cells, while the other half expresses the mutated FVIII or FIX. Some carriers of hemophilia may express a majority of non‐mutated X‐chromosomes and have higher levels of FVIII or IX. Others may have a majority of cells expressing the mutated X‐chromosome and present with low factor levels. Contrary to the extremely rare conditions quoted above, the occurrence of FVIII or FIX deficiencies in female carriers is frequent, although insufficiently recognized [19]. 5.6.2  Lower FVIII or FIX Levels in Carriers Before the availability of specific molecular diagnosis, low levels of FVIII of FIX were used to detect carriers in families of hemophilia A or B. However, low levels are not observed in all carriers, making this approach unsuitable to ensure diagnosis [20, 21]. It has long been recognized that for families with hemophilia A, the FVIII/VWF:Ag ratio is more useful to detect female carriers than FVIII deficiency alone [20, 21]. Indeed, a recent publication confirms that the median FVIII/VWF:Ag ratio was significantly lower in carriers (0.71; range: 0.18– 2.20) than in non‐carriers (1.39; range: 0.58–2.48) (P  A p. (Arg1708His) substitution in exon 14 of F8. Further analysis using array comparative genomic hybridization (CGH) diagnosed Turner syndrome, identifying hemizygosity for a single X‐chromosome. The child was phenotypically normal. The trauma resulted in swelling of the right knee extending into  the thigh. There was no previous history  of bleeding. Ultrasound and magnetic resonance imaging (MRI) provided evidence of effusion of the joint and rupture of the capsule. A large clot was removed and low‐ grade non‐specific synovitis was diagnosed. Genomic DNA from each parent lacked any F8 variant. The authors comment that the hemarthrosis was unexpected in a mild HA patient given the clotting factor level. She later experienced two further falls resulting in substantial hematomas to her forehead. Most females affected by Turner syndrome have a single X‐chromosome, but further causes include extreme lyonization and the inactivation of the X‐chromosome associated with a de novo variant on the paternal X‐chromosome in hemophilia carriers [14]. Further examples include a maternally inherited de novo variant in F8 in an obligate carrier [15] and a daughter of a carrier who inherited a de novo variant from her father, along with an inherited variant from her mother [16].

10.3 ­Phenotypic Analysis of Hemophilia B Phenotypic testing is needed before a potential HB carrier has any clinical intervention, although testing may not necessarily reveal carrier status. The main phenotypic test is for FIX:C activity, while FIX:Ag is rarely analyzed. For HB, menorrhagia is the most ­commonly identified symptom of bleeding. Treatments can include oral contraceptives and intrauterine hormone‐releasing devices. The Centers for Disease Control Universal Data Collection (UDC) database demonstrated that females with factor levels in the moderate–severe range displayed bleeding histories similar to their male counterparts. Although there were only 22 female HB carriers having or lacking normal FIX:C levels in this database, the proportion of HB in these individuals was consistent with the male population prevalence (females: 5/22, 22%; males: 2413/11 053, 22%). A consistent difference between female and male cohorts was that diagnosis was made at an older age amongst females. Unlike HA carriers, who may experience rises in FVIII:C level during pregnancy, carriers of HB have no similar benefit. This may affect the extent of postpartum bleeding, as FIX:C activity does not undergo a significant rise during pregnancy, in contrast to the twofold and threefold rise observed in HA carriers before the end of their third trimester [17]. Musculoskeletal issues in carriers of ­hemophilia with normal clotting factor levels or a mild–severe deficiency are not well described. A study evaluated the joint range  of motion (ROM) among 451 HA ­ carriers (n = 303) and HB carriers (n = 148). Participants aged between 2 and 69 years were enrolled in the Centers for Disease Control (CDC) UDC database (1998–2010). Individuals were compared to the Normal Joint Study dataset containing age‐related joint ROM of non‐carrier females developed by the CDC [17]. Females reporting one or more joint bleeds in the

Genetic and Laboratory Diagnosis

­revious six months rose with increasing p hemophilia severity: 2% mild, 33% moderate, and 37% severe FIX deficiency. FVIII, in ­contrast, demonstrated 15% mild, 36% moderate, and 54% severe FVIII deficiency respectively. Total joint ROM changes across all ages of HB carriers were affected in all but three of 148 females studied.

10.4 ­Phenotypic Analysis of von Willebrand Disease Von Willebrand disease results from quantitative and/or qualitative defects in VWF. Investigation of a patient with possible VWD includes a range of phenotypic analyses. A minimum of three measurements is usually required: VWF:Ag, a VWF activity measurement (with the newer assays frequently replacing the previously used VWF:RCo assay with VWF:GPIbR), and VWF:GPIbM [18]. The assays are based on the ristocetin‐ induced binding of VWF to a recombinant wild‐type GPIb fragment (VWF:GPIbR) or those that are based on the spontaneous binding of VWF to a gain‐of‐function mutant GPIb fragment along with a FVIII level. Dependent on the type of VWD, ristocetin‐ induced platelet aggregation (RIPA) may also be utilized to identify or exclude a diagnosis of type 2B VWD. For patients with type 1 VWD, VWF:Ag and activity should demonstrate approximately equivalent levels. Individuals with types 2A, 2B, and 2M generally have reduced levels of activity in comparison to VWF:Ag. However, a portion of 2B patients may have elevated VWF levels. For this patient group, the upper limit of 30 IU dL−1 can be ignored for VWF:Ag, activity, and collagen binding (VWF:CB) determinations as highest levels may exceed 100 IU dL−1. Further information is provided in the VWF Gene Review [19]. Type 3 VWD patients are defined as having < 1 IU dL−1 VWF:Ag. However, not all patient groups are categorized using this cut‐off. A small proportion of type 3 patients will develop inhibitory antibodies

to VWF (7.5–9.5%) [20], making treatment difficult. The development of alloantibodies to VWF is a rare complication of VWD treatment. Affected individuals may demonstrate various symptoms that can include loss of hemostatic response to VWF concentrates and, in rare cases, anaphylactic reactions. Multitransfused patients are at the greatest risk of this complication. Risk factors include large partial or complete deletions of the VWF gene, in addition to having a family history of anti‐VWF antibodies. Treatment for these patients can be undertaken using rFVIII in addition to bypassing agents and immune tolerance [21]. Additionally, there is a single report of VWF alloantibodies in a patient with type 2B VWD [22]. Table  10.2 shows the numbers of females with VWD represented on the UKHCDO database for 2016–2017.

10.5 ­Phenotypic Analysis of Inherited Bleeding Disorders Srivaths et  al. [23] reported on phenotypic differences in postmenarchal ( 7 days with pads and tampons becoming soaked in under two hours, passing clots, and ferritin reduced under the normal range or with anemia. Adolescent girls from menarche to 21 years with HMB symptoms were eligible for the program, as were those with a family history of bleeding disorders. The approach for testing included a history, physical examination, and laboratory tests; the pictorial bleeding assessment chart  (PBAC) was used for assessing HMB, along with a provider‐administered bleeding assessment tool (BAT) from the International Society on Thrombosis and Haemostasis (ISTH). VWD analysis was repeated at least twice. Platelet aggregation was tested only once, unless it was abnormal. Girls aged 9–15 were seen by an adolescent medicine specialist (AMS) whereas those aged 16–21 were seen by a gynecologist. The ISTH‐BAT has

been shown to distinguish between no bleeding and a possible bleeding disorder, with a score of < 2 in children suggesting that bleeding is unlikely. Quality of life is often impaired in this patient group, so a patient‐reported outcomes questionnaire was sent to participants to complete prior to their clinic visit. These analyses were conducted at each visit, while waiting for the diagnosis of a bleeding disorder, anemia being corrected, or HMB controlled. The care team included specialists from a number of disciplines: adolescent medicine, gynecology, hematology, coagulation medicine, nutrition, nursing manager, and nursing co‐ordinator. The AMS undertakes a psychosocial assessment at each visit to highlight any high‐risk behavior. The AMS also undertakes a physical exam including therapy assessment, such as suitability for a levonorgestrel IUD. Contraceptive management is individualized. Patients can call the clinic for problems, including breakthrough bleeding. A weekly pathology conference examines conflicting results and timing for additional testing. The nurse co‐ordinator takes all patient calls, educates patients and families, and co‐ordinates their care. Once a diagnosis is made, the HTC nurse takes over management of follow‐up care. The nutritionist reviews diet information to highlight iron‐rich foods and iron replacement therapy. Overweight and obese females are examined and referred to a childhood obesity clinic where appropriate. HMB treatment protocols are obtained from professional and scientific organizations. The team has standardized their own therapy and also recommended its use to other related staff within the institution. HMB management includes first‐line therapy of hormonal and non‐hormonal therapy, antifibrinolytics, packed red blood cells and intravenous iron, intravenous desmopressin, clotting factor replacement, and balloon tamponade. Second‐line therapy includes estrogen oral contraceptive pills and, rarely, dilation and curettage. Follow‐up occurs every 6–8 weeks; PBAC, coagulation testing, and a medication review

159

160

Inherited Bleeding Disorders in Women

are undertaken. Patients are seen at the clinic until a diagnosis is obtained or excluded. In diagnosed patients, HMB is controlled, iron is replenished or anemia corrected and patients are then seen at 6–12‐month intervals. Each is provided with an individual “period plan” that includes instructions in case of breakthrough bleeds, an escalation of therapy or non‐hormonal adjuvant therapy.

10.6 ­Genetic Analysis of Hemophilia A Variants resulting in reduced FVIII:C levels incude point mutations, deletions, duplications, insertions, insertions and deletions, intron 1 and intron 22 inversions, synonymous variants, intronic variants, and complex changes, along with neutral variants; the proportions are currently listed on the F8 database (Table  10.3). Before embarking on genetic analysis of F8, the clotting factor level needs to be known to determine whether to include analysis of the intron 22 and intron 1 inversions in those with severe HA. In individuals with FVIII:C levels > 1 IU dL−1, all 26 exons along with exon/intron boundaries should be sequenced. Different laboratories will include a ­proportion of flanking intronic

sequences, possibly including up to 25 base pairs (bp). If no candidate sequence variant is found, large deletions and duplications can be sought using copy number variation (­generally using m ­ ultiplex ligation‐dependent probe amplification (MLPA) or microarray analysis) and in certain cases, deep intronic sequence variants can also be sought [26–28]. F8 sequence variants have been identified from c.−1175 to c.8527 within the exonic sequence. The American Thrombosis Hemostasis Network (ATHN) undertook surveys which established that only ~ 20% of USA hemophilia patients had been genotyped. The My Life, Our Future team began a project aimed at genotyping USA patients. F8 and F9 genes were simultaneously targeted using molecular inversion probes. Results from the first 3000 HA and HB patients enrolled represent ~ 15% of the HA and HB USA population. Reportable genetic variants were seen in 98.1% (2357/2401) of HA patients and in 99.3% (595/599) of HB patients. One hundred females were included: eight had severe disease, seven had moderate, and 66 mild disease. F8 variants were reported in 81 (93%), of whom 10 had two variants that could be in cis/trans (two severe, one moderate, seven mild), while 19 females had HB; a

Table 10.3  Variant types in in hemophilia A, B, and VWD. Hemophilia A

Hemophilia B

von Willebrand disease

Missense

Missense

Missense

Nonsense

Nonsense

Nonsense

Splice‐site

Splice‐site

Splice‐site

Small insertion/deletion/indel

Small insertion/deletion/indel

Small insertion/deletion/indel

Intron 22 inversion

Synonymous

Synonymous

Intron 1 inversion

Promoter variants

Gene conversiona)

Deep intronic

Chromosomal rearrangement

Promoter variants

Synonymous

–

Chromosomal rearrangement

Promoter variant

–

–

Chromosomal rearrangement

–

–

a) Gene conversion events affect the 3′ end of intron 27 and 5′ end of exon 28. The original sequence is replaced by VWF pseudogene (VWFP1) sequence.

Genetic and Laboratory Diagnosis

single F9 variant was found in all 19 [29]. The study has now enrolled 7498 individuals with the plan to include more than 9000 patients, carriers and potential carriers, including > 2000 female carriers. Severe and moderate HA and HB were also  studied in USA females by di Michele et al. [30]. A multicenter retrospective study of girls and women enrolled 22 individuals (HA, 13 severe, four moderate; HB, one severe, four moderate). Females were older than males at diagnosis but experienced similar symptoms, including joint hemorrhages. Gynecological/obstetric bleeding was seen infrequently. F8 and F9 variants often demonstrated skewed X‐chromosome inactivation (XCI). Arthropathy and viral infections contributed to health‐related quality of life; 13/14 (93%) severe females experienced joint bleeding, first joint hemorrhage was between 12 and 14 years, and 10/13 had experienced mucocutaneous bleeding. In the moderate group, 7/8 (88%) had their first joint hemorrhage at age 7–8 years and all eight had experienced mucocutaneous bleeding. Amongst 15 postmenarchal females, 3/7 moderate patients reported heavy menses while 0/8 severe females fitted these criteria. Gynecological bleeding history was reported, with median age at menarche being 11–13 years in the severe group and 10–13 years in the moderate group. In the severe group, two females reported light and the remaining six moderate menorrhagia. Four of the seven with moderate hemophilia reported moderate and three of seven HMB. Menstruation duration was more than seven days in 1/8 severe females and 2/6 moderate. Iron deficiency anemia was present in 2/7 severe and 3/6 moderate patients. Target joints were present in 7/14 severe and 2/8 moderate females. In comparison with US affected males, the prevalence was higher amongst the females, but the group is rather small. As complications of hemophilia in the severe and moderate groups, one female has had joint replacement (0/7 moderate), 1/14 reported disability, 2/13 hepatitis transmission (1/8 moderate), 5/14 were

­ epatitis C antibody positive (4/8 moderate), h 2/14 were HIV positive, 2/13 had inhibitors, both of which were high titer that developed at 15 and 24 months. One moderate female developed an inhibitor at the age of 16 years. Renault et al. [31] investigated a Canadian family with three HA affected males and three affected females amongst 18 family members. Skewed XCI can result in very reduced levels of FVIII:C in females. In most females, the ratio of active paternal X‐chromosome to active maternal X‐chromosome is 50:50. X‐inactivation appears very early in gestation at the eight‐ or 16‐cell stage. XCI of 80:20 has been used to determine skewed inactivation. The female proband had FVIII:C of 0.02 IU mL−1 with a severe bleeding disorder diagnosed at age 2.5 months, two further affected females had FVIII:C of 0.28 and 0.39 IU mL−1, and other relatives from three generations were also studied. Affected males had FVIII:C of < 0.01– 0.04 IU mL−1, the causative variant being a type II intron 22 inversion. Assay of the HUMARA gene enables calculation of the relative activity of maternal and paternal X‐ chromosomes. Three paternal aunts had normal FVIII:C and also skewed XCI. Two were skewed toward the normal X, the other two had no F8 variant. Six of seven informative females demonstrated an extent of skewed XCI. Investigation of the maternal family members also demonstrated bias toward skewed XCI. In total, three females had very skewed low FVIII:C levels, six were less skewed, and only two not skewed. The distribution of XCI indicates a probable genetic influence over XCI. Discordant phenotypes in female family members suggest that XCI skew resulting from an XIST variant was unlikely. The authors propose that the range of X‐skewing could result from a mechanism seen in mice associated with the Xce gene. Four differing complementation groups each had a different probability of being on the active chromosome. Weak and strong alleles are both required for significant skewing. This is suggested as the most likely mechanism, and is consistent with a novel

161

162

Inherited Bleeding Disorders in Women

heritable human X controlling element. However further work is required to demonstrate how the mechanism may operate. Symptomatic HA carriers generally have one unaffected allele and one bearing a variant. They may present with bleeding symptoms of mild, moderate, and occasionally severe HA. Occasional carriers maybe homozygous or compound heterozygous for F8 variants. Lyonization can also result in skewed FVIII:C levels, from those resulting in symptomatic severe HA to those with levels at the top end of the normal range. Olsson et al. [32] investigated the relationship between variant type and tendency to bleed in severe and moderate HA carriers. A secondary aim examined genotype and FVIII:C levels. Females at potential risk of being HA carriers were recruited from three Swedish hemophilia centers. Genetic analysis identified carrier status in 106 (mean age at carrier diagnosis 27.2 years, range 6–58 years) in only those who were not obligate carriers. Carriers emanated from 81 families. The BAT previously designed for the Molecular and Clinical Markers for the Diagnosis and Management of type 1 von Willebrand disease study (MCMDM‐1VWD) was utilized to determine bleeding scores (BS) [33]. F8 variants were divided into null and non‐null types; 75 (71%) had null variants (48% inversions, 8% nonsense, and 15% small deletions/insertions outside poly A runs), while 31 (29%) had non‐ null variants (missense 24%, small deletions/ insertions in poly A runs 5%, splice variants affecting non‐conserved nucleotides, 1%). No significant difference was determined between bleeding scores in the two groups. Carriers with null variants had FVIII:C of 0.73 ± 0.39 IU mL−1, those with non‐null variants had FVIII:C of 0.54 ± 0.26 IU mL−1. Mean bleeding scores were two in the null variant group (range −3–10) and three (range −3–10) in the non‐null variant group. Miesbach et  al. [34] also investigated HA carriers, aiming to document the occurrence of bleeding in this group and to examine possible relationships between sequence variant types and bleeding.

Forty‐six carriers were included, with median age of 36.3 years (range 15–80 years). Disease severity was known in 44 of 46 families (12 had male relatives with severe HA, seven had moderate and 14 mild HA (FVIII:C 5–10%), while a further group of 11 had FVIII:C levels above 10%). Median FVIII:C amongst females was 55% (4–114%, mean 59 ± 24.5%). VWF:Ag levels were normal in all participants. Three groups were defined: (i) carriers lacking a bleeding tendency; (ii) weak bleeding tendency (easy bruising with normal menstruation); and (iii) strong bleeding tendency (easy bruising and prolonged menstruation). The latter group bled after giving birth and following surgical intervention, as did those in group 2; 23 of 46 carriers had children, of whom 10 (43%) experienced postpartum bleeding. Carriers with an intron 22 inversion experienced more bleeding than those with ­missense variants (hematoma 9/9 intron 22 vs 20/33 missense; prolonged menses 7/9 vs 13/33; prolonged bleeding after birth 3/6  vs  7/15, bleeding following tooth ­extraction 5/5 vs 10/14). FVIII:C level was reduced in those with an inversion (median 39%, mean 39 ± 15.06%) compared with carriers having missense variants (median 61%, mean 65 ± 24.5%). Seventy percent (n = 32) of carriers had experienced spontaneous bleeding symptoms, including menorrhagia (50%), recurrent nose bleeding (15%), and recurrent bleeding (4%). Many reported ≥ 1 symptom, while 30% did not experience any spontaneous bleeding symptoms. The authors comment that bleeding from minor wounds and following tooth extraction, surgery, and tonsillectomy were prolonged in those with FVIII:C levels between 40% and 60%.

10.7 ­Genetic Analysis of Hemophilia B Genetic analysis is more straightforward for HB than for HA. Sanger sequencing is ­relatively quick for the small F9 gene and

Genetic and Laboratory Diagnosis

alternatively, next‐generation sequencing (NGS) can be used. All patients can be examined first for sequence variants that include those in the promotor region and throughout the gene, and, if none are found, for copy number variants. Several large deletions from an exon to more than the entire F9 gene have been observed. Sequence variants have been identified from c.−55 to c.2864. However, large duplications are relatively uncommon. The majority of individuals will have one of these two variant types. Variants resulting in reduced FIX:C levels comprise point mutations, including those within the promoter region, deletions, duplications, insertions, insertions, and deletions, synonymous variants, and complex changes, along with neutral variants; the proportions are currently listed on the F9 database (see Table 10.3). Lavin et  al. [35] characterized the time course for FIX:C levels in female symptomatic carriers of HB with genetically confirmed HB Leyden. Two hundred and twelve patients with HB are documented on the Irish HB database. Ten patients had the same variant in F9, c.−35G > A, and analysis of their families identified 23 affected males having a median of 0.10 IU mL−1, range 0.05– 0.49 IU mL−1. Prior to puberty (< 11 years) and following puberty (> 18 years), mean FIX:C was a median of 54 IU mL−1, range 0.46–1.16 IU mL−1. Thirteen carriers were identified, of which two had a significant reduction in FIX:C levels along with a significant personal bleeding history. Carrier 1 was an obligate carrier; at age 3, her plasma FIX:C was 0.18 IU mL−1 (normal range 0.57– 1.89 IU mL−1). Menorrhagia and easy bruising affected her early teenage years. Carrier 2 was the index case in her family. She had experienced significant bleeding post dental extraction in addition to menorrhagia. At age 10, her FIX:C was 0.24 IU ml−1. Age‐related normalization in both carriers was observed in plasma FIX:C levels (1.17 and 0.93 IU mL−1) associated with a correction of bleeding phenotype. The authors also referred to a second publication on two further carriers of HB Leyden [36]. Both had reduced FIX:C levels

resulting from c.−22T > C. By 14 and 16 years, both had experienced a significant rise in FIX:C levels, although still below the normal range. Sharathkumar et  al. [37] reported on females with reduced levels of FVIII:C and FIX:C, that have been monitored by the CDC. Five hundred and thirty‐two females were present in the registry by December 2016. Variability of bleeding within a large Amish family from Ohio with a single initial founder was investigated in 64 HB carriers, resulting from a F9 c.886G > A, p.Thr296Met. Menstrual symptoms were investigated in 32 females. The median FIX:C level was 45% (range 19–114%); 19 (29.6%) females had FIX:C baseline levels of < 40%. Bleeding was assessed using an unknown‐8 scale. FIX:C was in the range of 13–122 IU mL−1. Twelve females had total bleeding scores of ≥ 3. Each reported bleeding from ≥ 2 sites, and their mean FIX:C levels were significantly reduced in comparison with those with scores < 3 (42 ± 10.3% vs 54.7 ± 22%, respectively; P = 0.002). Predominant bleeding symptoms in this group included those resulting from postpartum hemorrhage, HMB, and dental extraction. Differences in lyonization pattern plus the effect of autosomal regulatory elements that have an influence on FIX:C levels were suggested to contribute to the range of symptoms.

10.8 ­Genetic Analysis of von Willebrand Disease Analysis of VWF can be time‐consuming if Sanger sequencing is used to interrogate the 51 coding exon sequence. NGS can significantly reduce the analysis time, as genetic analysis of F8, F9, and VWF can be carried out simultaneously along with both F8 inversions using the method of Bastida et al. [38]. Patients with 2N VWD are not always readily discriminated from those with mild HA, as both disorders result in reduced FVIII:C ­levels. Analysis of both VWF and F8 can also be undertaken simultaneously using NGS.

163

164

Inherited Bleeding Disorders in Women

Alternatively, specific regions of VWF can be selected for initial analysis of the exons most likely to harbor causative variants and these include exons 18–20 and exon 28. The remaining exons can subsequently be analyzed for candidate sequence variants. Variants have been identified from the promoter region where there is a 13 bp deletion (c.−1522 to −1510del13) that results in mild type 1 VWD [39] to c.8419–8422dup in exon 52. Large deletions and duplications are present in a small proportion of VWD patients (2.26% of variants on the VWF database) and can be sought using copy number variation, using either MLPA or microarray analysis.

10.9 ­Guidelines Genetic analysis guidelines are available for  both HA and HB, but these are largely

directed at males with HA and HB [40, 41]. Keeney et al. present two guidelines on VWD diagnosis and management and on molecular analysis [42, 43]. Fijnvandraat et  al. [44] provide a guideline on the diagnosis and management of hemophilia. Perry et al. [45] describe the UK national external quality assessment scheme for molecular genetic testing in hemophilia. Neff [46] and Bowman and James [47] examine current controversies in VWD.

10.10 ­Summary The three disorders reviewed demonstrate a diverse range of disease severity and of pathogenic mechanisms, in patients with very mild–severe disease. A range of therapeutic interventions is available to manage the disorders, which is currently rapidly expanding.

­References 1 Factor FVIII database. www. 2

3 4

5

6

factorviii‐db.org. Payne, A.B., Miller, C.H., Kelly, F.M. et al. (2013). The CDC Hemophilia A mutation project (CHAMP) mutation list: a new online resource. Hum. Mutat. 34 (2): E2382–E2391. Factor IX database. www.factorix.org. Rallapalli, P.M., Kemball‐Cook, G., Tuddenham, E.G. et al. (2013). An interactive mutation database for human coagulation factor IX provides novel insights into the phenotypes and genetics of hemophilia B. J. Thromb. Haemost. 11 (7): 1329–1340. Li, T., Miller, C.H., Payne, A.B., and Craig Hooper, W. (2013). The CDC Hemophilia B mutation project mutation list: a new online resource. Mol. Genet. Genomic Med. 1 (4): 238–245. von Willebrand factor database. https:// grenada.lumc.nl/LOVD2/VWF/home. php?select_db=VWF.

7 Hampshire, D.J. and Goodeve, A.C. (2011).

8

9

10

11

The International Society on Thrombosis and Haematosis von Willebrand disease database: an update. Sem. Thromb. Hemost. 37 (5): 470–479. UK Bleeding Disorder Statistics for April 2016 to March 2017, A report from the UK National Haemophilia Database. Available at: www.ukhcdo.org/wp‐content/ uploads/2018/02/Bleeding‐Disorder‐ Statistics‐for‐April‐2016‐to‐March‐2017‐ for‐UKHCDO.pdf. Bowyer, A.E., Goodeve, A., Liesner, R. et al. (2011). P.Tyr365Cys change in factor VIII: haemophilia a, but not as we know it. Br. J. Haematol. 154 (5): 618–625. Canadian Hemophilia Registry and Inherited Bleeding Disorders Registry. Available at: https://fhs.mcmaster.ca/chr/. Trickey, R.C., Percy, C., Jenkins, P.V. et al. (2017). Experience of immune tolerance in a carrier of severe haemophilia a with inhibitor development

Genetic and Laboratory Diagnosis

post‐surgery. Haemophilia 23 (3): e234–e235. 12 Byams, V.R., Kouides, P.A., Kulkarni, R. et al. (2011). Surveillance of female patients with inherited bleeding disorders in United States haemophilia treatment centres. Haemophilia 17 (Suppl 1): 6–13. 13 Williams, V.K., Suppiah, R., Coppin, B. et al. (2012). Investigation of inflicted injury in a young girl reveals mild haemophilia a and Turner’s syndrome. Int. J. Lab. Hematol. 34 (1): 98–101. 14 Afifi, A.M. (1974). Spontaneous haemophilia in a genotypically normal female. A family study. Acta Haematol. 52 (2): 112–119. 15 Cai, X.H., Wang, X.F., Dai, J. et al. (2006). Female hemophilia a heterozygous for a de novo frameshift and a novel missense mutation of factor VIII. J. Thromb. Haemost. 4 (9): 1969–1974. 16 Seeler, R.A., Vnencak‐Jones, C.L., Bassett, L.M. et al. (1999). Severe haemophilia a in a female: a compound heterozygote with nonrandom X‐inactivation. Haemophilia 5 (6): 445–449. 17 Staber, J., Croteau, S.E., Davis, J. et al. (2018). The spectrum of bleeding in women and girls with haemophilia B. Haemophilia 24 (2): 180–185. 18 Bodo, I., Eikenboom, J., Montgomery, R. et al. (2015). Platelet‐dependent von Willebrand factor activity. Nomenclature and methodology: communication from the SSC of the ISTH. J. Thromb. Haemost. 13 (7): 1345–1350. 19 Goodeve A, James P. Von Willebrand Disease Gene Review. 2017. Available at: www.ncbi.nlm.nih.gov/books/NBK7014. 20 James, P.D., Lillicrap, D., and Mannucci, P.M. (2013). Alloantibodies in von Willebrand disease. Blood 122 (5): 636–640. 21 Federici, A.B. (2016). Current and emerging approaches for assessing von Willebrand disease in 2016. Int. J. Lab. Hematol. 38 (Suppl 1)): 41–49. 22 Baaij, M., van Galen, K.P., Urbanus, R.T. et al. (2015). First report of inhibitory von

23

24

25

26

27

28

29

30

31

Willebrand factor alloantibodies in type 2B von Willebrand disease. Br. J. Haematol. 171 (3): 424–427. Srivaths, L.V., Zhang, Q.C., Byams, V.R. et al. (2018). Differences in bleeding phenotype and provider interventions in postmenarchal adolescents when compared to adult women with bleeding disorders and heavy menstrual bleeding. Haemophilia 24 (1): 63–69. Zia, A., Lau, M., Journeycake, J. et al. (2016). Developing a multidisciplinary young Women’s blood disorders program: a single‐centre approach with guidance for other centres. Haemophilia 9. Munro, M.G., Critchley, H.O., Broder, M.S. et al. (2011). FIGO classification system (PALM‐COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age. Int. J. Gynaecol. Obstet. 113 (1): 3–13. Castaman, G., Giacomelli, S.H., Mancuso, M.E. et al. (2011). Deep intronic variations may cause mild hemophilia a. J. Thromb. Haemost. 9 (8): 1541–1548. Pezeshkpoor, B., Zimmer, N., Marquardt, N. et al. (2013). Deep intronic ‘mutations’ cause hemophilia a: application of next generation sequencing in patients without detectable mutation in F8 cDNA. J. Thromb. Haemost. 11 (9): 1679–1687. Bach, J.E., Wolf, B., Oldenburg, J. et al. (2015). Identification of deep intronic variants in 15 haemophilia a patients by next generation sequencing of the whole factor VIII gene. Thromb. Haemost. 114 (4): 757–767. Johnsen, J.M., Fletcher, S.N., Huston, H. et al. (2017). Novel approach to genetic analysis and results in 3000 hemophilia patients enrolled in the my life, our future initiative. Blood Adv. 1 (13): 824–834. Di Michele, D.M., Gibb, C., Lefkowitz, J.M. et al. (2014). Severe and moderate haemophilia a and B in US females. Haemophilia 20: e136–e143. Renault, N.K., Dyack, S., Dobson, M.J. et al. (2007). Heritable skewed X chromosome inactivation leads to haemophilia A

165

166

Inherited Bleeding Disorders in Women

32

33

34

35

36

37

38

39

40

expression in heterozygous females. Eur. J. Hum. Genet. 15 (6): 628–637. Olsson, A., Ljung, R., Hellgren, M. et al. (2016). Phenotype and genotype comparisons in carriers of haemophilia A. Haemophilia 22 (3): e235–e237. ISTH bleeding assessment tool. Available at: http://c.ymcdn.com/sites/http://www.isth. org/resource/resmgr/ssc/ isth‐ssc_bleeding_assessment.pdf, accessed 30 January 2018. Miesbach, W., Alesci, S., Geisen, C., and Oldenburg, J. (2011). Association between phenotype and genotype in carriers of haemophilia A. Haemophilia 17 (2): 246–251. Lavin, M., Jenkins, P.V., Healy, M.L. et al. (2015). Age‐related factor IX correction in symptomatic female carriers with haemophilia B Leyden. Haemophilia 21 (6): e498–e500. Hildyard, C. and Keeling, D. (2015). Effect of age on factor IX levels in symptomatic carriers of Haemophila B Leyden. Br. J. Haematol. 169 (3): 448–449. Sharathkumar, A., Hardesty, B., Greist, A. et al. (2009). Variability in bleeding phenotype in Amish carriers of haemophilia B with the 31008 C‐‐>T mutation. Haemophilia 15 (1): 91–100. Bastida, J.M., Gonzalez‐Porras, J.R., Jimenez, C. et al. (2017). Application of a molecular diagnostic algorithm for haemophilia A and B using next‐generation sequencing of entire F8, F9 and VWF genes. Thromb. Haemost. 117 (1): 66–74. Othman, M., Chirinian, Y., Brown, C. et al. (2010). Functional characterization of a 13‐bp deletion (c.‐1522_‐1510del13) in the promoter of the von Willebrand factor gene in type 1 von Willebrand disease. Blood 116 (18): 3645–3652. Guidelines prepared by Steve Keeney, Mike Mitchell and Anne Goodeve on behalf of the UK Haemophilia Centre Doctors’ Organisation (UKHCDO), the Haemophilia Genetics Laboratory Network

41

42

43

44

45

46

47

and the Clinical Molecular Genetics Society. Practice Guidelines for the Molecular Diagnosis of Haemophilia A, updated July 2010. Available at: www.acgs. uk.com/media/774613/haemophilia_a_ bpg_revision_sept_2011_approved.pdf. Guidelines prepared by Mike Mitchell, Steve Keeney and Anne Goodeve on behalf of the UK Haemophilia Centre Doctors’ Organisation (UKHCDO), the Haemophilia Genetics Laboratory Network and the Clinical Molecular Genetics Society. Practice Guidelines for the Molecular Diagnosis of Haemophilia B, updated December 2010. Available at: http://ukhcdo.org/docs/ Haemophilia%20B%20BPG%20revision %20Sept%202011%20APPROVED.pdf. Keeney, S., Bowen, D., Cumming, A. et al. (2008). The molecular analysis of von Willebrand disease: a guideline from the UK haemophilia Centre Doctors’ Organisation Haemophilia Genetics Laboratory Network. Haemophilia 14 (5): 1099–1111. Keeney, S., Collins, P., Cumming, A. et al. (2011). Diagnosis and management of von Willebrand disease in the United Kingdom. Sem. Thromb. Hemost. 37 (5): 488–494. Fijnvandraat, K., Cnossen, M.H., Leebeek, F.W., and Peters, M. (2012). Diagnosis and management of haemophilia. Br. Med. J. 344: e2707. Perry, D.J., Goodeve, A., Hill, M. et al. (2006). UK NEQAS for Blood Coagulation. The UK National External Quality Assessment Scheme (UK NEQAS) for molecular genetic testing in haemophilia. Thromb. Haemost. 96 (5): 597–601. Neff, A.T. (2015). Current controversies in the diagnosis and management of von Willebrand disease. Ther. Adv. Hematol. 6 (4): 209–216. Bowman, M.L. and James, P.D. (2017). Controversies in the diagnosis of type 1 von Willebrand disease. Int. J. Lab. Haematol. 39 (Supp 1): 61–68.

167

11 Antenatal Diagnosis Rezan A. Kadir1,2, Irena Hudecova 3, and Claudia Chi 4 1

Institute for Women’s Health, University College London, London, UK Department of Obstetrics and Gynaecology and Katharine Dormandy Haemophilia and Thrombosis Centre, The Royal Free Foundation Hospital, London, UK 3 University of Cambridge, Cancer Research UK, Cambridge, UK 4 National University Hospital, Department of Obstetrics and Gynaecology, Singapore 2

11.1 ­Introduction

11.2 ­Genetic Counseling

Inherited bleeding disorders are lifelong ­conditions that present with a broad spec­ trum of bleeding manifestations, ranging from easy bruising and epistaxis to poten­ tially debilitating musculoskeletal bleeding and life‐threatening intracranial hemor­ rhages. Despite advances in their treatments, they remain incurable and can be associ­ ated  with significant long‐term morbidity. Women with inherited bleeding disorders can pass on the gene defect to their offspring and therefore are at risk of having an affected child, depending on the inheritance pattern of the condition. The decision regarding reproduction is fundamentally complex and challenging and further complicated for these women because of their genetic risks. Developments in molecular genetics and technologies have created new opportunities and expanded the reproductive options for these women. The aim of this chapter is to explore the reproductive choices of women with inherited bleeding disorders.

It is essential for women who are affected or carriers (known carriers or have the possi­ bility of carrying the disorder) to have the opportunity to consider the implications for themselves and their offspring. The aims of genetic counseling are to provide them with sufficient information to make a decision appropriate to their own situation and to support them throughout the process. It should ideally be carried out before concep­ tion to allow sufficient time for the individ­ uals to gain a full understanding of the information provided and for the laboratory to perform genetic testing. Furthermore, pre‐pregnancy planning prevents the par­ ents from having to make difficult decisions under time pressure during early pregnancy and avoids limitations to their reproductive choices. Genetic counseling should be pro­ vided by individuals with good communica­ tion skills and a detailed knowledge of the disorder, including the molecular genetics, care, and treatment involved in the relevant

Inherited Bleeding Disorders in Women, Second Edition. Edited by Rezan A. Kadir, Paula D. James, and Christine A. Lee. © 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.

168

Inherited Bleeding Disorders in Women

inherit the condition whereas females are affected as carriers. The incidence of hemo­ philia A is approximately 1 in 5000 male births and of hemophilia B, 1 in 25 000 male births. Given that females have two X‐chromosomes, the birth incidence of hemophilia carriers by inference would be 1 in 2500 and 1 in 12 500 females, respectively. For each affected male in a particular family, there are potentially many more carrier females. Assuming a nor­ mal genotype in the partner, these women have a 1:4 chance of having a child with hemo­ philia in each pregnancy; there is a 50% chance that a son will be affected and a 50% chance that a daughter will also be a carrier of the condition (Figure  11.1). Since the severity of hemophilia is fairly consistent among family members, carriers can be informed whether their risk is for severe, moderate, or mild hemophilia, although for some missense mutations, the range can span two categories. Von Willebrand disease (VWD) is an autosomal condition that affects both males and females. It is the most common inher­ ited bleeding disorder, with a reported prev­ alence in published studies ranging from 1 in 100 to 1 in 10 000, depending on the method used to identify patients [1]. It dis­ plays both d ­ ominant and recessive inherit­ ance. Type 1, the most common (60–80%)

country and the available reproductive options. 11.2.1  Confirmation of Diagnosis To make an informed decision on reproduc­ tive options, it is vital to first establish the diagnosis of the prospective parents and their genetic risk of having an affected child. This involves obtaining an accurate and detailed family history as well as carrying out phenotypic and genotypic assessments to ascertain the diagnosis, the coagulation fac­ tor level, and the mutation within the family. Individuals undergoing genetic testing need to have an understanding of the purposes and implications of the test. They should also be aware of its accuracy and limitations. Aspects of carrier testing and genetic diag­ nosis are discussed in detail in Chapters 5 and 10. 11.2.2 Inheritance For prospective parents, the risk of having an affected child is determined by the inheritance pattern of the condition. Hemophilia A and B are X‐linked recessive bleeding disorders caused by a deficiency in coagulation factor VIII (FVIII) and IX (FIX), respectively. Males

X

Y

X

X

Carrier female

X

X

Carrier female (50%)

X

X

Normal female

Figure 11.1  Inheritance of hemophilia.

X

Y

Affected male (50%)

X

Y

Normal male

Antenatal Diagnosis

and generally milder form of VWD, is ­transmitted as an autosomal dominant trait. It is characterized by a partial quantitative deficiency of von Willebrand factor (VWF). Type 2 refers to a  qualitative deficiency of VWF and is sub­divided into four variants (2A, 2B, 2M, and 2N). Types 2A, 2B, and 2M are mainly i­nherited in an autosomal domi­ nant pattern, whereas type 2N is inherited recessively. Type 3 also has an autosomal recessive inheritance and is the most severe form of VWD with a virtually complete deficiency of VWF. Factor XI deficiency is also an autosomal condition but has a complex pattern, with both recessive and dominant forms. Although FXI deficiency occurs in all populations, severe FXI deficiency is a rare condition in  most, with an incidence of 1/100 000. However, in the Ashkenazi Jewish popula­ tion, the heterozygous form occurs in 8% of  individuals and the homozygous form in 0.2–0.5% [2]. Other rare bleeding disorders include afi­ brinogenemia, hypoprothrombinemia, and deficiencies of factor V, combined factors V and VIII, factor VII, factor IX, factor X, and factor XIII. They are inherited as autosomal recessive traits and generally are expressed in homozygotes or compound heterozygotes. Their estimated incidence in the severe (homozygous) form ranges from 1 in 500 000 (factor VII deficiency) to 1 in 2 million (fac­ tor II deficiency). However, they are more prevalent in racial groups, communities, or countries where consanguineous marriages are common, such as the Middle East, south­ ern Asia, and North African countries. 11.2.3  Clinical Phenotype Other factors that may influence reproduc­ tive decision making include the likely clini­ cal phenotype of the disorder, its severity and potential complications, the potential risk of inhibitor development, the expected quality of life of affected children, and the possible impact on the family. Thus, these predictors should be provided as part of the genetic

counseling. Details on the efficacy, safety, and side‐effects of current treatment should also be addressed. When discussing these issues, it is important to take into considera­ tion the individual’s and family’s previous experience with the condition. The implica­ tions of a genetic disorder can undoubtedly cause great anxiety and distress so genetic counseling should not only encompass the technical issues but also the emotional aspects. Adequate support should be pro­ vided throughout the whole process. 11.2.4  Reproductive Options Women with or carriers of inherited bleeding disorders may choose to abstain from having children for fear of passing the genetic defect to their offspring, especially when adequate treatment and/or the option of prenatal ­diagnosis (PND) is not available. They may choose to use non‐biological methods such as adoption or fostering. On the other hand, they may decide to accept any outcome by taking a chance of having an affected child. Over time, reproductive options have been broadened by advances in molecular tech­ nology and developments in the techniques for PND and assisted reproduction. Women can choose to have antenatal diagnosis and selective termination of an affected preg­ nancy or consider assisted conception with the use of donor gametes. There is even the possibility of selecting embryos that are free of the genetic disorder before implantation to prevent the birth of an affected child. In order for the prospective parents to make an  informed choice on their reproductive options, it is vital that they are provided with adequate information regarding the availa­ bility of these options and the procedures involved (how and where they would be per­ formed), including their accuracy, limita­ tions, and risks. The options for women with inherited bleeding disorders, in general, include: ●● ●●

not having children adoption or fostering

169

170

Inherited Bleeding Disorders in Women ●●

●●

●● ●●

conceiving naturally and accepting the outcome of the pregnancy conceiving naturally and having PND with the option of termination of affected pregnancy assisted conception with donor gametes preimplantation genetic diagnosis (PGD).

The advantages and disadvantages of each option should be explored, including the psy­ chological effects on the mother and the family as whole.

11.3 ­Prenatal Diagnosis Prenatal diagnosis should be undertaken fol­ lowing adequate counseling and in centers with full genetic, hematological, and obstet­ ric expertise. It is generally considered in pregnancies at risk of moderate or severe inherited bleeding disorders and usually when termination of pregnancy is contem­ plated if an affected fetus is identified. This is because the techniques for definitive PND are usually invasive and carry risks to the pregnancy. However, women’s attitudes to PND and termination of pregnancy differ considerably, depending on various personal, social, and cultural factors. They may opt for antenatal diagnosis for other reasons such as psychological preparation in the case of an affected child and planning for place and methods of delivery. Prenatal diagnosis of inherited bleeding disorders entails non‐invasive and/or inva­ sive tests. At present, in clinical practice the definitive PND of inherited bleeding disor­ ders can only be achieved through invasive procedures such as chorionic villus sam­ pling (CVS), amniocentesis, and fetal cord blood sampling (cordocentesis), whereas non‐­invasive tests are mainly used for diag­ nosis of fetal gender in case of hemophilia through fetal ultrasound examination or free fetal DNA (ffDNA) in the maternal circulation. The options of screening for fetal chromo­ somal and structural abnormalities and of

testing for common chromosomal abnor­ malities such as trisomy 21, 18, and 13, in the event of invasive testing, should also be dis­ cussed and offered, as appropriate, in the pretest counseling. It is very likely that more than one labora­ tory will be involved in the PND diagnostic pathway. Good communication between all the parties, co‐ordinated by the hemophilia center, is essential for a successful and rapid process [3]. 11.3.1  Invasive Prenatal Diagnostic Tests These techniques are mainly based on obtaining fetal materials for genetic analysis or clotting factor assays. Thus, identification of the causative mutation or informative markers is a prerequisite for any DNA‐based PND. As this can take up to several weeks, it needs to be done well before considering PND. Advanced planning and careful co‐ ordination between the obstetrician and hematologist as well as the genetic laboratory and fetal medicine unit are necessary to ensure the successful attainment of accu­ rate  results and to minimize the risks of complications. Furthermore, before embarking on these options, the mother/couple should be informed about the procedure: how it will be performed, the possibility of not obtaining an adequate sample, non‐diagnostic results, and potential side‐effects for both mother and fetus. Processes for any long‐term sam­ ple storage and quality control should be dis­ cussed. It should also be agreed with them what tests will be performed and in what order. In particular, it should be agreed whether tests unrelated to the bleeding dis­ order will be performed. The mother’s rhesus status should be assessed and standard precautions for the  prevention of rhesus isoimmunization should be adopted. Invasive testing can be associated with bleeding complications in women with bleeding disorders. The preg­ nancy‐induced rise in factor levels may not

Antenatal Diagnosis

be sufficient to achieve normalization of ­factor levels in early pregnancy. Thus, rele­ vant coagulation factor should be assessed and appropriate hemostatic cover should be administered prior to performing any ­invasive PND procedure in women with low factor levels or those with bleeding phenotype. It should be agreed in advance to whom, how, and where the results of the PND tests will be given. Once the results are known, the options available to the woman should be discussed. It may be necessary to allow time for the results to be considered before a deci­ sion is reached. Invasive PND procedures should be car­ ried out by skilled operators who should be trained to the competencies expected of ­subspecialty training in maternal and fetal medicine or other international equivalent. Competency should be maintained by per­ forming at least 30 procedures per annum and continuous audit of the operator’s per­ formance should be carried out [4]. 11.3.1.1  Chorionic Villus Sampling

Chorionic villus sampling is currently the method of choice for obtaining fetal materi­ als for the PND of inherited bleeding ­disorders. It offers the advantage of attaining an early diagnosis and thus a shorter period of uncertainty than amniocentesis. Early diagnosis is of particular importance when termination of an affected pregnancy is being considered, as there may be personal or religious pro­hibitions on late termination of pregnancy. Furthermore, termination at a  later gestation is likely to be more trau­ matic. CVS is usually performed between 11 (11 + 0) and 13 (13 + 6) weeks of gestation under direct ultrasound guidance to obtain a sample of the chorionic villi (placenta) for genetic analysis. The chorionic villus tissue is a rich source of fetal DNA that can be extracted and used for polymerase chain reaction (PCR)‐based testing for fetal sexing, in the case of hemophilia, if the fetus is male, then direct mutation detection or ­polymorphism linkage analysis for PND of

hemophilia as appropriate. Results are ­usually available within 48–72 h of receipt of samples. Prior to the CVS procedure, an ultrasound assessment is performed to confirm the via­ bility of the pregnancy, the gestation, the number of fetuses, and the positions of the fetus and placenta. CVS can be performed using the percutaneous transabdominal or transcervical approach (Figure  11.2). The choice is currently based on the experience and preference of the operator. However, the transabdominal route is preferable and more widely used because the transcervical route is associated with more vaginal bleeding and  is technically more demanding, with a higher rate of multiple insertions and sam­ pling failure [5]. Chorionic villus sampling is performed under continuous ultrasound screen using different methods of aspiration (negative ­ pressure with syringe/vacuum aspirator or ­ biopsy forceps). Needles used include 18 gauge, 20 gauge, double needle 17/19 gauge, or d ­ ouble needle 18/21 gauge. As there are no published studies comparing clinical outcomes using dif­ ferent techniques, it is advisable for clinicians to use a technique that they are familiar with [4]. Local anesthetic infiltration is usually used with the transabdominal route. Chorionic villus sampling is associated with an approximately 1–2% risk of preg­ nancy loss. The 2010 Royal College of Obstetricians and Gynaecologists (RCOG) guideline reports the additional risk of mis­ carriage following amniocentesis to be around 1% and that the risk may be slightly higher following CVS [4]. A Cochrane review in 2017 comparing transabdominal and tran­ scervical approaches identified five studies with 7978 women and found a higher total pregnancy loss and spontaneous miscar­ riages after transcervical CVS, but this was mainly due to excess loss in the transcervical arm of one study. There was no clear differ­ ence for total pregnancy loss and miscarriage rate in the other four trials. In addition, the differences in pregnancy loss and miscarriage between transabdominal and transcervical

171

172

Inherited Bleeding Disorders in Women

Probe

Needle

Bladder Placenta

Vagina

Womb (uterus) Amniotic fluid

Entrance of womb (cervix) Rectum

Figure 11.2  Chorionic villus sampling (transabdominal). Source: Reproduced with permission from the Royal College of Obstetricians and Gynaecologists.

CVS were no longer clear when a more con­ servative random effects statistical model was applied to overcome heterogeneity of the data [5]. Chorionic villus sampling should not be performed before 10 weeks of gestation because of the potential risk of oromandibu­ lar limb hypoplasia and isolated limb disrup­ tion defects when carried out before this gestation. Furthermore, CVS before 11 weeks is technically difficult to perform due to the small size of the uterus and a thinner pla­ centa [4]. Sampling failure can occur because of technical difficulties. There is a small (< 1%) chance of failing to obtain a result from the laboratory test. False‐negative or ‐positive results may result from maternal cell con­ tamination or placental mosaicism. There is also a very small (less than 1 in 1000) risk of serious infection from inadvertent puncture of the bowel or from contaminants on the skin or the ultrasound probe/gel. Standard procedures for infection control are recom­ mended to avoid this complication. The risk of injury to the fetus is minimal and is

decreased by the use of real‐time ultrasound guidance. 11.3.1.2 Amniocentesis

Amniocentesis is performed from 15 (15 + 0) weeks onwards. It involves the insertion of a fine needle through the maternal abdominal wall into the amniotic cavity to obtain a sample of amniotic fluid through needle ­ aspiration (Figure  11.3). Ultrasound assess­ ment is carried out prior to the procedure to ascertain the positions of the placenta and fetus. Local anesthetic is not usually used for this procedure as it does not appear to affect the level of pain experienced. Needles of 18, 20, and 22 gauge are used, with one experi­ mental model suggesting less amniotic fluid flow from the puncture site with smaller gauge needles [4]. Amniotic fluid contains fetal cells (amnio­ cytes) from which rapid detection of specific chromosomes, including the sex chromo­ somes in cases of hemophilia, can be achieved by fluorescent in situ hybridization (FISH). DNA can also be extracted directly and used for PCR‐based testing for linkage analysis or

Antenatal Diagnosis

Ultrasound probe

Syringe Bladder

Vagina Placenta

Womb Amniotic fluid

Cervix (entrance to womb) Rectum

Figure 11.3  Amniocentesis. Source: Reproduced with permission from the Royal College of Obstetricians and Gynaecologists.

direct mutation detection, with results usually available within 48–72 hours. However, there is often insufficient DNA present in the sam­ ple for analysis. Therefore, testing has to be delayed until cultured cells are available, which takes approximately two weeks. A selection of polymorphic markers unrelated to the gene under investigation should always be included in the analysis of results to ensure that mater­ nal contamination is detected. The RCOG recommends that amniocente­ sis is performed under direct ultrasound con­ trol with continuous needle tip visualization to reduce the chance of obtaining a “bloody tap” as the presence of blood can interfere with cell culture. This also helps to minimize the risk of fetal trauma, which is rare. The incidence of procedure‐related pregnancy loss is approximately 1% [4]. Other complica­ tions include a small chance (< 1%) of not obtaining a definitive diagnosis due to incon­ clusive results or culture failure and an even smaller risk (< 1/1000 procedures) of serious infection caused by inadvertent puncture of the bowel, skin contaminants, or organisms present on the ultrasound probe or gel. Amniocentesis has the disadvantage of providing a diagnosis in the second trimester,

so the option of termination of pregnancy in affected cases is provided at a later gestation. Although amniocentesis can be performed before 14 weeks’ gestation (early amniocen­ tesis), this is associated with a higher preg­ nancy loss rate and a higher incidence of talipes and respiratory morbidity and should be avoided [4]. In a Canadian randomized trial, early (between 11 + 0 and 12 + 6 gesta­ tional weeks) amniocentesis was associated with an increased incidence of fetal loss (7.6% vs 5.9%), talipes (1.3% vs 0.1%), and amniotic fluid leakage post procedure (3.5% vs 1.7%) compared with midtrimester (between 15 + 0 and 16 + 6 gestational weeks) amniocentesis [6]. Furthermore, early amniocentesis was found to be technically more difficult, with higher rates of multiple needle insertions and cytogenetic culture failure [6]. This may be attributed to the presence of two separate membranes (amnion and chorion) until the 15th gestational week. 11.3.1.2.1  Third‐Trimester Amniocentesis

Amniocentesis can also be carried out in the third trimester in certain circumstances, including late karyotyping and evaluation of infection in preterm labor or rupture of the

173

174

Inherited Bleeding Disorders in Women

membranes. The deferment of amniocente­ sis for PND until the third trimester is a pos­ sible option in certain situations where the risk of pregnancy loss following second‐­ trimester amniocentesis may be higher (for example, previous pregnancy loss or in mul­ tiple pregnancy) to avoid the risk of miscar­ riage or severe prematurity associated with ­second‐trimester amniocentesis. However, third‐trimester amniocentesis is  an invasive procedure and there is an approximate 1% risk of procedure‐related complications such as preterm delivery, pre­ mature rupture of membranes, and placen­ tal abruption [7]. Complications such as multiple attempts (> 5%) and bloodstained fluid (5–10%) are also more common compared with midtrimester procedures. ­ Furthermore, there is an approximate 1% chance of failing to obtain a sample and a higher culture failure rate in amniotic fluid samples taken in the third compared to the second trimester (10% vs < 1%) [7]. However, a case–control study of 167 women under­ going amniocentesis after 32 weeks for lung maturity assessment reported no com­ posite poor outcome (urgent birth, abrup­ tion, ­premature rupture of the membranes and five‐minute Apgar score less than seven) compared with one adverse outcome in con­ trols [7]. Other drawbacks of late amniocen­ tesis include the risk of unexpected delivery before the availability of the test result. In addition, a positive diagnosis could become a psychological burden in advance of deliv­ ery. It may also raise several ethical and moral dilemmas that are beyond the scope of this chapter. In pregnancies at risk of severe inherited bleeding disorders, PND by third‐trimester amniocentesis (after 34 weeks’ gestation) has been utilized as a possible option that ena­ bles appropriate planning of mode and place of delivery for parents who are unwilling to accept the risk of fetal loss associated with earlier prenatal testing [8]. If the fetus is unaffected, labor and delivery can be man­ aged without any restrictions in the local

maternity unit. However, published data are limited to two small case series including nine [8] and five cases [9]. The majority of mutations in hemophilia A and B and VWD can usually be reliably detected in DNA extracted from amniotic fluid samples. However, amniotic fluid in later pregnancy may contain few viable fetal cells, so poor‐quality DNA may be avail­ able for extraction. High‐quality DNA is particularly important for successful PND ­ of  some mutations such as F8 intron 22‐­inversion [10]. In a single‐center experi­ ence, third‐­trimester amniocentesis was per­ formed in nine pregnancies in four carriers of severe hemophilia A, one carrier of severe hemophilia B, and one woman at risk of car­ rying a child with type 2B VWD. In one of nine cases no result could be obtained [8]. No hemostatic cover was required for any of  the procedures as all the mothers had corrected their coagulation defect during ­ pregnancy. None of the women suffered ­ any  complications and all delivered after 37  weeks of gestation. The procedure was very helpful in five pregnancies where ­confirmation of an unaffected fetus enabled routine obstetric management at units geo­ graphically convenient for the mother. All women reported psychological benefits from knowing the genetic status of the baby prior to delivery. Three women subsequently had a second pregnancy, and all requested late amniocentesis and mutation analysis from the outset, based on their positive experience in the first pregnancy. On the other hand, in a survey of 41 carriers of hemophilia regis­ tered with a tertiary hemophilia center in London, the majority (80%) reported that they would not consider the option of third‐ trimester amniocentesis due to its invasive nature and associated risks [11]. 11.3.1.3 Cordocentesis

Cordocentesis, or percutaneous umbilical cord blood sampling, is another invasive diagnostic technique performed after 18 (18 + 0) weeks of gestation. It involves the

Antenatal Diagnosis

insertion of a 22 gauge needle under con­ tinuous ultrasound guidance through the maternal abdomen and uterine walls and into the umbilical cord to obtain a sample of fetal blood for laboratory analysis. It is per­ formed under aseptic conditions. Local anesthesia is not required when cordocente­ sis is performed for diagnostic procedures. Ideally, the umbilical vein at the placental insertion of the umbilical cord is chosen for sampling because it is fixed (Figure  11.4). Alternatively, a midsegment (free loop) may be used. It is preferable to avoid transplacen­ tal needle insertion as this increases the risk of fetal‐maternal hemorrhage and possible sensitization but this approach may be una­ voidable if the placenta is anterior. Cordocentesis carries an approximately 1–2% risk of procedure‐related fetal loss when performed by an experienced operator [12]. There is a risk of cord hematoma or bleeding from the puncture site. This is ­usually transient and self‐limiting in fetuses

with normal coagulation. However, fetuses affected with bleeding disorders are at par­ ticular risk of such bleeding complications and this should be taken into consideration when planning for the procedure [10]. Other potential complications include failure in obtaining a sample, fetal bradycardia that may require urgent delivery, infection, pre­ mature rupture of membrane, and premature birth. There is also a risk of the sample being contaminated by maternal blood or amni­ otic fluid, leading to dilution of the sample and false results with low levels of all coagu­ lation factors. Measures should be taken to check for maternal blood contamination, including the assessment of red cell mean corpuscular volume (MCV) and compared to the maternal MCV (> 120 fL for fetal MCV and ~ 90 fL for maternal MCV), the Kleihauer–Betke test showing resistance of fetal hemoglobin to acid elution and/or anal­ ysis of molecular markers. Dilution with

Ultrasound transducer Umbilical cord

Placenta

Figure 11.4  Cordocentesis.

175

176

Inherited Bleeding Disorders in Women

amniotic fluid may also cause activation and consumption of coagulation factors, again resulting in artefactually low levels. The plasma should be tested for the coagulation factor under investigation and one other control coagulation factor level. Fibrinogen should be tested, as a markedly reduced fibrinogen level would suggest activation of the sample. Gestational age should be con­ sidered when interpreting the results and comparison should be made with the fetal blood normal ranges derived from the appropriate gestation. If the coagulation factor under investigation is undetectable ­ and a control coagulation factor level and fibrinogen level are within the expected range then a severe deficiency of that coagu­ lation factor is confirmed. Cordocentesis has largely been superseded by CVS and amniocentesis in obtaining fetal material for genetic studies. The latter two procedures allow earlier diagnosis, are less technically demanding, and are associated with a lower risk of complications. Cordo­ centesis to investigate hemostatic disorders is only considered if all other possible tech­ niques are not available or do not give ­conclusive results. In developing countries, genetic services for hemophilia may not be widely available and cordocentesis remains the main method for PND. In a study looking at indications for fetal blood sampling per­ formed form 1990 to 2009 in a tertiary refer­ ral center in India, 1342 procedures were performed, including 36 procedures for PND of hemophilia [13]. On the other hand, in developed countries, molecular techniques are available for the vast majority of cases and give more reliable results with lower risks of complications. Therefore, cordocentesis is only considered if a woman wishes to ensure that she does not have a child affected with severe hemophilia but a causative mutation cannot be identified, and is generally reserved for cases with severe deficiencies of coagula­ tion factors where an undetectable level would suggest an affected fetus [10]. Pretest counseling must cover the possibility of arte­ fact, incorrect or uninterpretable results.

11.3.2  Non‐Invasive Prenatal Diagnostic Tests 11.3.2.1  Ultrasound Assessment

Prenatal determination of fetal gender can be achieved through ultrasound assessment. This is valuable in the management of preg­ nancies at risk of X‐linked genetic disorders such as hemophilia. Knowledge of fetal sex enables carriers of hemophilia to avoid inva­ sive diagnostic procedures with their associ­ ated risks in 50% of pregnancies where the fetus is female, and relieves their anxiety. It also allows planning for the management of labor and mode of delivery. Ultrasound has provided accurate assess­ ment of fetal gender in the second trimester for over 20 years, based on ultrasound visu­ alization of the external genitalia [14]. Since the development of the external genitalia is similar in both sexes until 14 weeks’ gesta­ tion, fetal gender determination by ultra­ sound was initially limited to the second trimester onwards. It enabled the exclusion of female fetal gender prior to amniocentesis. Improvements in the resolution of ultra­ sound have subsequently allowed early ­identification of fetal gender from as early as 11 weeks’ gestation by assessing the direction in which the genital tubercle develops [15]. Fetal gender can be determined by measur­ ing, in a midsagittal plane, the angle between the genital tubercle and a line drawn through the lumbosacral vertebrae. The fetal sex is considered to be male if the phallus is directed cranially with an angle to the lum­ bosacral vertebrae > 30° and female if the phallus is directed caudally with an angle < 30° [15]. However, visualization of the geni­ tal tubercle may not always be possible at the first attempt due to fetal lie, and a repeat examination may be required. In addition, differentiation of the genital tubercle into the  male or female phallus only begins at 11 weeks’ gestation and its accuracy is l­ imited before 14 weeks [16]. Therefore, in clinical practice, this sign cannot be used to deter­ mine fetal gender for exclusion of female fetuses prior to CVS.

Antenatal Diagnosis

11.3.2.2  Cell‐Free Fetal DNA (cffDNA) in Maternal Circulation

The discovery of cell‐free fetal DNA (cffDNA) in maternal plasma through identification of Y‐chromosome‐specific DNA sequences in pregnant women carrying male fetuses opened up new opportunities for non‐invasive prenatal diagnosis (NIPD) [17]. Initially, much effort was devoted to exploring intact fetal cells in maternal circulation as the source of fetal material for DNA analysis. However, this has proven to be challenging due to extremely low amounts of fetal cells found in maternal blood. Additionally, the persistence of fetal cells in maternal circulation from the previous pregnancy has been shown to be a potential source of false‐positive results. In comparison to fetal cells, the concentrations of cffDNA in maternal plasma DNA are much higher, com­ prising approximately 10% of total cell‐free DNA in the first trimester of pregnancy [18], and show an increasing trend with advancing gestational age. Using sensitive molecular methods for its detection in maternal plasma, cell‐free DNA (cfDNA)‐based testing is cur­ rently offered starting from the 10th week of gestation. The rapid clearance of cffDNA fol­ lowing delivery makes this analysis ideal for the prenatal assessment of the fetal status reflecting the current pregnancy. Specifically, cffDNA is cleared in two phases: an initial rapid phase occurring within one hour after delivery is followed by a slow phase within 13 hours and a final clearance about 1–2 days post partum [19]. The clinical uptake of cfDNA‐based testing of fetal aneuploidies worldwide has been remarkable and the progress in NIPD of single‐gene disorders ­ shows a potential to implement a wide range of conditions into current prenatal diagnostics frameworks.

11.4 ­Prenatal Diagnosis of Hemophilia Prenatal diagnosis of hemophilia was ini­ tially restricted to fetal sex determination by amniocentesis in the second trimester. Some

carriers utilized this opportunity to inter­ rupt pregnancies involving a male fetus, but most found this unacceptable because in half of these cases, the male fetus would have been unaffected. The definitive diagnosis of hemophilia became possible toward the end of the 1970s when techniques were intro­ duced for obtaining fetal blood samples under fetoscopic guidance alongside the development of sensitive assays for deter­ mining coagulation factor VIII and IX on samples so obtained. PND of hemophilia was achieved by means of coagulation or immu­ noradiometric assays for factor VIII or IX on  fetal blood samples obtained during the  18–20th weeks of gestation in male pregnancies identified by ultrasound assess­ ment or amniocentesis [20]. Fetal blood sampling under fetoscopic guidance was later replaced by ultrasound‐guided cordo­ centesis [21] with a lower rate of procedure‐ related fetal loss (approximately 5–6% vs 1–2%, respectively). The characterization of the genes for ­factors VIII and IX together with the devel­ opment of CVS techniques and PCR tech­ nology eventually made first‐trimester PND of hemophilia feasible during the 1980s. At first, the genetic diagnosis was achieved indirectly through linkage studies using restriction fragment length polymorphisms (RFLPs) [22, 23]. This approach has several limitations, including uninformative poly­ morphisms, unavailability of blood samples from key pedigree members, lack of prior family history, mosaicism, risk of recombi­ nation with extragenic markers, plus the sen­ sitive and ethical issue regarding paternity. Subsequently, advances in molecular tech­ nology and further knowledge of genes for  factors VIII and IX have facilitated the antenatal diagnosis of hemophilia through direct identification of the causative muta­ tion [24, 25]. Direct mutation detection cir­ cumvents the problems associated with linkage analysis and has become widely available in developed countries following the widespread use of direct automated sequencing technology. It is the most reliable

177

178

Inherited Bleeding Disorders in Women

method for PND and should be used if avail­ able. However, in situations where the causa­ tive mutation cannot be identified because of the complexity of the genes involved, the  heterogeneity of mutations or, in many developing countries, the lack of resources and facilities required, linked polymorphic markers can be utilized to achieve PND. When both direct and indirect genetic anal­ yses are not available or fail, cordocentesis provides an alternative method of PND through evaluation of fetal clotting factors. The analysis of cfDNA in maternal plasma has brought significant developments in the management of pregnancies at risk for hemophilia. Following the introduction of a cfDNA‐based test for fetal sex assessment in several clinical centers across the United Kingdom [26], the number of invasive proce­ dures in at‐risk pregnancies has significantly decreased, and are now required only in male pregnancies for a subsequent diagnos­ tic follow‐up [27]. The reasons for PND in hemophilia is twofold, firstly to provide the option of

t­ ermination of an affected pregnancy if cho­ sen by the mother. In current clinical prac­ tice, this is achieved by analysis of cfDNA in the maternal plasma at 10 weeks’ gestation and CVS at 11–13 weeks’ gestation if the fetus is male. Carriers of hemophilia who present later in pregnancy are offered ultra­ sound assessment of fetal sex and amnio­ centesis from 15 weeks’ gestation. The second indication for PND is to help safe management of labor and planning safe mode of delivery. Fetal gender is diagnosed during routine anomaly scan at 18–22 weeks’ gestation and in case of a male fetus, the mother is offered third‐trimester amnio­ centesis at 34 weeks’ gestation to estab­ lish  the hemophilia status of the fetus. Management of labor and mode of delivery for affected fetuses are discussed in detail in Chapter 5. When invasive testing is declined due to the associated risks, all male fetuses should be considered as potentially affected and managed accordingly [28]. Figure  11.5 demonstrates an algorithm for PND of hemophilia.

First trimester ffDNA - 10 weeks gestation 10% repeat test in a week

F

M Decline

Confirmation with second trimester USS F 50% chance of being a carrier

Avoid CVS – 50% Female fetuses – not subjected to CVS

Offer CVS Accept

M

CVS (or amniocentesis from 16 weeks)

50% chance of being affected

M

M Caution: - Avoid invasive monitoring techniques in labor - Avoid instrumental deliveries - Avoid postnatal surgical intervention - Oral vitamin K

Affected

M Unaffected

Option of continuing the pregnancy No Yes

TOP

Figure 11.5  Algorithm for PND of hemophilia. CVS, chorionic villus sampling; TOP, termination of pregnancy; USS, ultrasound scan.

Antenatal Diagnosis

11.4.1  New Developments in NIPD of Hemophilia Recent progress in NIPD for pregnant hemo­ philia carriers represents a milestone in the currently established prenatal diagnostic program, and is expected to refine the pre­ sent care in hemophilia families [29, 30]. Unlike fetal sex determination through the detection of fetal‐specific sequences derived from chromosome Y, direct iden­ tification of hemophilia mutations using maternal plasma is technically more chal­ lenging. This approach requires interro­ gation of a single disease‐causing gene alteration inherited by the fetus against the background of maternally derived DNA molecules, for which the introduction of more sophisticated techniques and the anal­ ysis of substantially higher amounts of DNA were essential. In 2008, Lun et  al. proposed a strategy for  plasma DNA‐based detection of mono­ genic disease, which allowed determination of whether the fetus has inherited the maternal mutant allele despite the co‐existence of the maternal background of plasma DNA [31]. This method, named relative mutation dosage (RMD), has subsequently been applied for direct detection of maternally inherited hemophilia sequence variants in maternal plasma [29]. In this study, RMD specifically assessed an allelic imbalance of the mutant and wild‐type allele in maternal plasma of pregnant hemophilia carriers. The concept of RMD has only become feasible by taking advantage of highly sensitive molecular tech­ nologies, such as digital PCR (dPCR), which are able to precisely determine their respec­ tive amounts in maternal plasma. The diag­ nostic accuracy of the digital RMD‐based approach is then dependent on the number of interrogated DNA molecules and the fetal DNA fraction that governs the degree of allelic imbalance. The higher levels of fetal DNA fraction enable a more accurate determina­ tion of the imbalance of the respective alleles. Tsui et  al. analyzed maternal plasma from seven mostly third‐trimester pregnancies at

risk for hemophilia by using microfluidics dPCR and RMD, and correctly predicted fetal inheritance of F8 and F9 sequence variants starting from the 11th week of pregnancy [29]. In comparison with microfluidics dPCR, which is limited by the number of digital reac­ tions achieved per chip, droplet digital PCR (ddPCR) and massively parallel sequencing (MPS) technologies currently offer the high­ est accuracy for plasma DNA quantification. Hudecova et al. investigated the use of ddPCR coupled with RMD in 15 pregnant hemo­ philia carriers, and accurately determined fetal inheritance of sequence variants scat­ tered across the F8 and F9 genes starting from the 8th week of pregnancy [30]. Clinically, ddPCR analysis simultaneously enables fetal sex assessment and, in male fetuses, a muta­ tion‐specific assay based on a  priori knowl­ edge of the maternal mutation assists in direct  determination of fetal hemophilia status. For samples with low fetal DNA ­ ­fractions, collected typically at earlier gesta­ tional ages, an additional number of digital analyses are necessary to enhance the muta­ tion detection. F8 int22h‐related inversions are the most common mutations associated with the most severe clinical outcome. Thus, carriers of these mutations are more likely to opt for PND and benefit from cfDNA‐based NIPD testing. Although NIPD for hemophilia car­ riers of fetal F8 int22h‐related inversions has been for a long time hindered by the molecular complexity of these mutations, targeted MPS coupled with a haplotype‐ based strategy, namely relative haplotype dosage (RHDO), has recently assisted in resolving its fetal inheritance from maternal plasma [30]. The authors first established a  haplotype linkage of the set of single nucleotide polymorphisms (SNPs) within a 7.6 Mb region spanning the F8 gene inferred from genotypes of the mother and proband. Subsequently, accurate measurement of maternally inherited fetal haplotypes facili­ tated fine mapping of the F8 region, and ultimately enabled correct prediction of fetal hemophilia status of F8 int22h‐related

179

180

Inherited Bleeding Disorders in Women

inversions in three families starting from the 12th week of pregnancy. In RHDO anal­ ysis, the combination of alleles inherited by the fetus from its mother is deduced as a series of inheritance blocks whose resolu­ tion depends on the number and distribu­ tion of identified SNPs [32]. Consequently, the sequencing depth, the number of informative SNPs, and the fetal DNA frac­ tion determine the diagnostic accuracy of RHDO analysis through targeted MPS. In the future, linked‐read haplotype sequenc­ ing technology can be utilized to establish the mutation haplotype signature in the sample from a pregnant carrier directly, and thus can offer a universal NIPD approach without the availability of the proband [33]. With the decrease in sequencing costs and by examining larger cohorts of pregnant carriers of F8 int22h‐related inversions, tar­ geted MPS approaches coupled with linked‐ read haplotype sequencing technology may become an essential part of NIPD for hemo­ philia carriers. Validation via a larger study cohort would be particularly important to evaluate the test performance before its introduction into the clinical setting.

11.5 ­Prenatal Diagnosis of von Willebrand Disease The data on PND of VWD are limited to a few case reports. Antenatal diagnosis is not often required or requested in pregnancies at risk of type 1 because the bleeding tendency is usually mild. It is only applicable to fami­ lies with severe VWD, mainly type 3, particu­ larly where the parents already have one affected child. Mutation analysis of the index case can identify the familial mutation, which should be confirmed to be present in each parent and can subsequently be sought in a fetus using chorionic villus or amniocentesis samples [34]. Antenatal diagnosis by direct mutation detection has been performed in certain laboratories for type 2 variants where clusters of mutations have been identified.

However, direct mutation detection is not practical in VWD because of the large size of  the VWF gene and the heterogeneity of the mutations. The causative mutation is unknown in the majority of cases and, in these cases, linkage studies with RFLP or variable number of tandem repeat (VNTR) sequences may be used for PND. Cordo­ centesis and assessment of fetal FVIII and VWF assays provide an alternative option for PND when genetic diagnosis is not avail­ able. The risk of serious fetal bleeding and compromise associated with this procedure should be considered and discussed with the parents for fetuses at risk of severe VWD. In one case report, PND of severe type 3 VWD was made by DNA analysis on CVS sample at 12 weeks’ gestation but further confirmed by assessment of fetal clotting factors by cordo­ centesis at 20 weeks. The latter procedure was complicated by a massive fetomaternal hemorrhage, fetal hypovolemia, and persis­ tent bradycardia requiring intracardiac blood transfusion [35].

11.6 ­Prenatal Diagnosis of Rare Bleeding Disorders Rare bleeding disorders are generally inher­ ited as recessive traits and are more com­ mon in consanguineous marriages. In most cases, they are due to mutations in the genes that encode the relevant clotting factors. Multiple mutations have been described for each rare bleeding disorder and they are often unique for each kindred and in some cases, no causative mutation is found [36]. This may be due to defects in the non‐­coding regions or in genes that encode for the regulators of intracellular transport and ­ post‐translational modification of clot­ ting factors. PND based on genetic analysis is only feasible if the causative mutation is known or if there are informative genetic markers. There have been a small number of reports on the PND of rare bleeding disor­ ders, including factor VII [37], factor X [38], and factor XIII deficiencies [39] based on

Antenatal Diagnosis

direct mutation detection, linkage analysis or a combination of these methods. Similar to the PND of hemophilia, cordocentesis can be considered when genetic diagnosis in the fetus is unavailable or unfeasible [40]. A recent systematic review of literature, reported only 39 cases of PNDs in rare bleeding disorders. Majority were for severe FXIII or FVII deficiency in families with pre­ vious affected siblings who suffered serious bleeding complications mainly intracranial haemorrhage. While PND was achieved by genetic testing in the majority; factor assay of fetal blood obtained via cordocentesis was required in 10 cases where the genetic muta­ tion was not known or genetic analysis was not available [41].

11.7 ­Preimplantation Genetic Diagnosis Preimplantation genetic diagnosis is an alter­ native reproductive technique available for couples at risk of having a child with a certain genetic disorder. It is a very early form of PND, in which embryos created in vitro are analyzed for the specific genetic abnormality and only unaffected embryos are transferred to the uterus. This can prevent the birth of an  affected child and obviates the need for PND with selective termination of an affected pregnancy, a procedure that can be very trau­ matic for the parents both emotionally and physically. PGD is an option for couples who would not consider termination of pregnancy for religious or personal reasons and for those with concurrent infertility. The first successful clinical application of PGD was in couples at risk of having children with X‐linked disorders by sex determina­ tion from the preimplantation embryo. Male embryos were discriminated by PCR amplifi­ cation of Y‐chromosome sequences and only female embryos were transferred. Since then, PGD has been used for an increasing number of single‐gene disorders (monogenic dis­ eases, MGD) as well as chromosomal abnor­ malities. These conditions can be diagnosed

provided the causative mutation has been identified, the chromosome that carries the gene can be tracked through linkage studies or the specific chromosomal rearrangement is known. A review of 10 years (1997–2007) of data collected by the European Society for Human Reproduction and Embryology (ESHRE) PGD consortium, including over 27 000 cycles which reached oocyte retrieval, reported that 17% of the cycles were per­ formed for single‐gene disorders and 4% were for sexing of X‐linked disease [42]. In the latest published data collection from the ESHRE [43], the clinical pregnancy rate for MGD is 30% per embryo transfer and deliv­ ery rate is 26%. The corresponding rates for sexing only in X‐linked diseases are 29% and 27% respectively [43], with no cases of misdiagnosis. Hemophilia is the third most common X‐linked disorder to be tested by PGD, after fragile X and Duchenne muscular dystrophy [44]. Initially, FISH was used in PGD of hemophilia to provide a diagnosis of fetal sex with reimplantation of female embryos only. This leads to unnecessary disposal of healthy male embryos and reduces the number of embryos suitable for transfer and thus a lower pregnancy success rate. More recently, disease‐specific DNA amplification‐based tests have become available with the advan­ tages of identification and transfer of unaf­ fected male embryos. In some circumstances, although controversial, female carriers can be identified and excluded from transfer according to the family’s experience of mani­ festing carrier females. Tests to achieve spe­ cific diagnosis of hemophilia via PGD have been carried out by direct mutation detec­ tion in conjunction with testing informative linked markers [45] or indirect genetic test­ ing using flanking informative linked mark­ ers only [46]. Ten years of data collection by the ESHRE PGD consortium has seen the use of > 1100 cycles of sex determination for X‐linked disorders while the number of hemophilia disease‐specific tests undertaken is growing, with just under 100 cycles under­ taken [44].

181

182

Inherited Bleeding Disorders in Women

Preimplantation genetic diagnosis can only be performed in conjunction with in  vitro fertilization (IVF) to allow the ­creation of an  embryo in vitro accessible for  biopsy and genetic diagnosis. This involves ovarian stimulation to enable the maturation of ­several oocytes followed by oocyte retrieval, IVF using intracytoplasmic sperm injection (ICSI), and embryo culture. A biopsy sample for genetic testing can be obtained at various developmental stages: (i) polar bodies from the oocyte/zygote stage, (ii) blastomeres from cleavage stage embryos on day 3, and (iii) trophectoderm cells from blastocysts. Each of the methods has its advantages and disadvantages and the choice mainly depends on the clinical situation. Various molecular techniques have been developed for the genetic testing of biopsy materials. The diagnostic protocols are based on PCR to amplify sufficient DNA from the cells followed by the diagnosis of a specific genetic defect or FISH for the analy­ sis of chromosomes, including sexing. Techniques such as multiplex, fluorescent, and/or real‐time PCR and minisequencing are being introduced in an attempt to over­ come the  inherent difficulties of single‐cell PCR (including failure of amplification, potential sample contamination, and allele dropout, ADO) due to the limited amount of DNA template and to improve diagnostic accuracy. Published data from the latest ESHRE PGD consortium show that PGD for sexing in X‐linked diseases was performed with ICSI in 66%, laser drilling in 86%, and biopsy by cleavage‐stage aspiration in 91% of cycles. FISH was still the most frequently used method (78% of cycles); PCR was applied in 15% and whole‐genome amplifica­ tion/arrays in 6% of cycles [43], while for diagnosis of MGD, including X‐linked disor­ ders, ICSI was used in 99% cycles and PCR was still the most widely used first‐line method of DNA amplification (93%). The use of laser was the preferred method for biopsy (81% of cycles). Day 3 cleavage‐stage embryo biopsy was most frequently used

(93%) while blastocyst biopsy was used in only 2%. Genetic testing was carried out on either one blastomere (58% of cycles to PGD with day 3 biopsy) or two blastomeres per embryo (28% of cycles to PGD with day 3 biopsy) [43]. Preimplantation genetic diagnosis is likely to become a realistic option for more cou­ ples at risk of having a child affected by hemophilia or other severe inherited bleed­ ing disorders in the near future. The reliabil­ ity and efficiency of PGD are improving with continuous scientific and technological advancement, but it remains technically challenging and labor intensive, requiring the close collaboration of a team of special­ ists. As the diagnosis depends on a single cell, strict controls are essential to pre­ vent  assay contamination from extraneous DNA  sources. This necessitates dedicated equipment and laboratories with filtered air. Testing at the single cell level is also affected by ADO, where one of the two alleles does not amplify during the initial rounds of PCR. Diagnostic assays must therefore be designed and validated to enable detection of this phenomenon and provide information on the misdiagnosis risk. Preimplantation genetic diagnosis also has  considerable financial implications as it entails the use of IVF in addition to advanced molecular techniques. Apart from being costly, IVF is associated with high levels of stress and anxiety as well as the risks of ­multiple pregnancies and ovarian hyperstim­ ulation. Unlike IVF for infertility, the number of suitable embryos available for transfer is diminished by excluding affected embryos or those with inconclusive results. While con­ current aneuploidy screening has been shown to reduce pregnancy loss and improve live birth rate in PGD for MGD [47], the chance of not having any genetically trans­ ferrable embryos after both tests is high, and usually only 25% of embryos are transferra­ ble [48]. Therefore, the couple should be counseled properly before proceeding to PGD with preimplantation genetic screen­ ing (PGS) treatment cycles.

Antenatal Diagnosis

Preimplantation genetic diagnosis has raised several ethical and safety issues that still need to be addressed. Further data are still required to establish the potential long‐ term effect of embryo biopsy on children born after the procedure. A few studies that  looked at babies born after PGD until they reached two years of age reported early developmental parameters (biometric and health outcomes and mental and psycho­ motor development) to be similar to those of ICSI babies without PGD or naturally conceived babies [48]. The largest study included 581 children born after blastomere biopsy, and revealed similar birth weights and major  malformation rates at birth or at two months  of age in comparison with ICSI babies. However, significantly more perinatal deaths were seen in multiple preg­ nancies than with ICSI multiple pregnan­ cies, while no such difference was observed in singleton pregnancies after PGD [48]. A longer follow‐up study on children from singleton pregnancies after PGD which reached an age of 5–6 years showed compa­ rable cognitive and psychosocial develop­ ment to children born after ICSI or conceived naturally [49].

11.8 ­Views about and Experiences of Prenatal Diagnosis of Women in Families Affected with Inherited Bleeding Disorders Several studies have been conducted on the attitudes of carriers of hemophilia toward PND and their reproductive behavior. There is a lack of data on these aspects for women with type 3 VWD or rare inherited bleeding disorders. The reported experiences of carri­ ers of hemophilia are likely to reflect those of women with rare bleeding disorders since developments in their treatment and PND have followed a similar trend. Nonetheless, further research in these areas is required to confirm this postulation.

11.8.1  Attitudes Toward Prenatal Diagnosis Women’s attitudes toward PND and termi­ nation of pregnancy vary widely between different countries and cultures. Religion is a strong determinant of the individual’s decision [50]. For example, Catholic women are less likely to choose these options than women of other religious belief [50]. Other factors that affect women’s views about PND include their attitudes to termination of pregnancy, the severity of the disorder, their experience of the condition within the fam­ ily, and the nature and availability of the tests involved. In a survey of 105 carriers of severe or mod­ erate hemophilia in Sweden, those who expe­ rienced PND were significantly more positive about selective termination of pregnancy than those who did not opt for PND [51]. This is understandable since selective termination is often the primary purpose for PND. Similarly, in a study of 549 potential and obli­ gate carriers in the Netherlands, one of the main reasons for not choosing PND was a negative attitude to termination of pregnancy [52]. Most women from affected families objected to PND because they did not con­ sider hemophilia to be sufficiently serious to justify an abortion [52]. In a review of the obstetric experience spanning two decades of carriers from a tertiary hemophilia center in the UK, the uptake of invasive prenatal diag­ nostic tests remained relatively low at 35% (17/48) and 20% (13/65) over the periods 1985–1995 and 1995–2005, respectively [53, 54]. A low (14%) uptake of PND was also reported among carriers of severe and mod­ erate hemophilia in Sweden toward the end of the 1990s [51]. In the same study, carriers who had experienced the complications of hemophilia or its treatment were more in favor of PND than women whose affected children received modern treatment without complications. This appears to apply also to the decision to interrupt an affected preg­ nancy. In case series, only carriers from fami­ lies with severe hemophilia opted for PND for

183

184

Inherited Bleeding Disorders in Women

hemophilia and termination of affected preg­ nancy [54]. Timing of the prenatal diagnostic test can also influence the woman’s decision on its uptake. Women are more inclined to object to PND if it is performed at a later stage of gestation. When PND first became possible through fetoscopic fetal blood sampling in the second trimester, only 3% of all pregnant hemophilia carriers in the USA utilized this option [55]. In the Dutch study by Varekamp et al. [52], 30% of the respondents stated that they would opt for PND and potential termi­ nation in the future if this was done early in pregnancy compared with only 15% if it was performed at the 18–20th gestational week. In a study involving 29 carriers of hemophilia and 23 of their partners who had experienced first‐trimester CVS, most women (79%) and all men were positive about having PND by CVS in a future pregnancy [56]. In a more recent nationwide survey in the Netherlands carried out between 1992 and 2004, of 207 carriers of hemophilia A or B, 112 (54%) women opted for PND. Of 22 pregnancies testing positive for an affected male fetus, 18 (82%) opted for termination of pregnancy. A liberal view toward termination of preg­ nancy, severe hemophilia in the family, older maternal age at time of first pregnancy, and having no religion were associated with the choice for PND [57]. A large difference in attitudes to PND and termination of pregnancy for hemophilia has been observed between developed and ­developing countries [58, 59]. The burden of a severe genetic condition is evidently heavy in developing countries where there are lim­ ited facilities and resources. The life expec­ tancy and quality of life of affected individuals are subsequently poorer in these countries. As a result, the prevention of the birth of children with hemophilia is a key objective in these countries [60]. One study compared the views of hemophilic populations in Iran and Italy about PND and termination of pregnancy and found the acceptability of ter­ mination of pregnancy for hemophilia was

four times greater (58% vs 17%) in Iranians than in Italians. Likewise, most families affected by hemophilia in India, where there is poor awareness of the disorder, inadequate diagnostic facilities, and a lack of social sup­ port systems, would consider termination of an affected pregnancy [59]. However, despite high acceptability, the use of PND is limited by multiple factors such as lack of awareness, financial constraints, inadequate diagnostic facilities, and social stigma. There has been a significant advance in the treatment of hemophilia, with life expec­ tancy of affected individuals approaching that for the normal male population [61]. The creation of newer clotting factor con­ centrates that are longer acting and less immunogenic, and/or have enhanced bypass­ ing activity has increased treatment options that are safer, more effective, and more con­ venient for affected individuals. In addition, advances in gene therapy and its recent clinical success are promising and may ­ potentially provide a cure for hemophilia. Undoubtedly, availability of these novel ther­ apies will have an important impact on the attitude to PND in the future. Similarly, avail­ ability of NIPD would also influence decision making for mothers and families with hemo­ philia. Availability of non‐invasive prenatal fetal sex determination increased the uptake for prenatal fetal sex determination among carries of hemophilia by 52–97% [53, 54]. In a recent study including 73 pregnancies among 61 carriers of hemophilia, the uptake for non‐invasive PND was 100% compared to only 23% for invasive testing [11]. 11.8.2  Experiences and Psychological Effects of Prenatal Diagnosis Prenatal diagnosis is associated with a con­ siderable amount of anxiety and distress, particularly when performed because of a high genetic risk [62]. Anxiety arises from the risk of having an affected child with long‐term morbidity, having potentially

Antenatal Diagnosis

painful and invasive diagnostic tests, having a miscarriage due to these tests, receiving an abnormal result, making the decision whether to continue or have a termination of the pregnancy, and undergoing the pro­ cess of termination and its complications. The negative psychological effect associ­ ated with PND has been found to be more prominent when PND is carried out at a more advanced gestation [56]. Equally, ter­ mination of pregnancy is associated with greater mood disturbances following the procedure when carried out after second‐ trimester compared with first‐trimester PND. Among carriers of hemophilia, sig­ nificantly fewer signs of depressive mood were reported following CVS than fetal blood sampling in the second trimester [56,  63]. Termination of an affected preg­ nancy was found to be emotionally painful for all carriers despite the gestation, but the negative experience was more profound after second‐trimester termination [63]. A number of characteristics have been identified among carriers at high risk of hav­ ing negative psychological reactions in asso­ ciation with PND of hemophilia by fetal blood sampling [64]. They include a negative view of oneself in general and of being a car­ rier, a planned pregnancy, high education, good general knowledge of hemophilia, and a guiding philosophy of life. These women also reported signs of depressive mood sig­ nificantly more often at follow‐up. Although fetal blood sampling has largely been super­ seded by CVS and amniocentesis for PND, the results of the study may help to identify women who are likely to have difficulty in coping emotionally with the process of PND and therefore at particular need for psycho­ logical support. The long‐term psychological effects of PND have been evaluated using a symp­ tom checklist (SCL) questionnaire (SCL‐90) among 50 carriers of hemophilia who underwent PND at a median of 5.5 (range 1–17) years since their last PND test [65]. These carriers had a lower tendency for

somatization (psychological distress that arises from perceptions of bodily dysfunc­ tion) than carriers who did not opt for PND. It is reassuring that PND of hemo­ philia did not appear to have negative long‐ term psychological effects. Nevertheless, psychosocial support is recommended and crucial for all couples undergoing PND and those opting for subsequent termination of affected pregnancy.

11.9 ­Termination of Pregnancy When termination of pregnancy is chosen, close collaboration between hemophilia center, fetal medicine, and gynecology unit is required for the appropriate choice of method of termination and avoidance of bleeding complications. Termination of pregnancy can be performed by surgical uterine evacua­ tion before 15 weeks after cervical preparation with  misoprostol or gemeprost. However, most units do not offer a surgical option after 13 weeks of gestation. Medical termina­ tion is performed by mifepristone f­ollowed 36–48 hours later by either misoprostol or gemeprost. Surgical termination of pregnancy is considered to be preferable for women with inherited bleeding disorders as it obviates the risk of unpredicted bleeding with medical ­termination. For a termination after 21 weeks and six days, feticide using an i­ntracardiac injection of KCl is mandatory to  ­prevent the possibility of a live birth. These procedures should be carried out and co‐ordinated by a fetal medicine unit. Counseling and support should be provided by the hemophilia center as well as the gynecology unit. Anti‐D prophy­ laxis for Rh‐negative mothers is admi­nistered as appropriate. Women with bleeding disorders are at a higher risk of bleeding complications during and after termination of pregnancy [50, 53]. Coagulation factors should be assessed prior to the procedure and appropriate hemostatic cover should be provided to minimize the risk of bleeding.

185

186

Inherited Bleeding Disorders in Women

­References 1 Ng, C., Motto, D. G., and di Paola, J.

(2015). Diagnostic approach to von Willebrand disease. Blood 125 (13): 2029–2037. 2 Asakai, R., Chung, D. W., Davie, E. W., and Seligsohn, U. (1991). Factor XI deficiency in Ashkenazi Jews in Israel. N. Engl. J. Med. 325: 153–158. 3 Belvini, D., Salviato, R., Acquila, M. et al. (2013). Prenatal diagnosis of haemophilia B: the Italian experience. Haemophilia 19 (6): 898–903. 4 Royal College of Obstetricians and Gynaecologists (2010). Amniocentesis and CVS. Green‐Top Guideline No. 8. London: RCOG. 5 Alfirevic, Z., Navaratnam, K., and Mujezinovic, F. (2017). Amniocentesis and chorionic villus sampling for prenatal diagnosis (review). Cochrane Database Syst. Rev. (9): CD003252. 6 Canadian Early and Mid‐trimester Amniocentesis Trial (CEMATGroup) (1998). Randomised trial to assess safety and fetal outcome of early and midtrimester amniocentesis. Lancet 351: 242–247. 7 O’Donoghue, K., Giorgi, L., Pontello, V. et al. (2007). Amniocentesis in the third trimester of pregnancy. Prenat. Diagn. 27: 1000–1004. 8 Cutler, J., Chappell, L. C., Kyle, P., and Madan, B. (2013). Third trimester amniocentesis for diagnosis of inherited bleeding disorders prior to delivery. Haemophilia 19 (6): 904–907. 9 Scarr, C., Conner, C., and Collins, P. (2104). Third trimester amniocentesis for the diagnosis of haemophilia A and B. Poster presentations BMFMS: Fetal Medicine PFM.32. BMJ ADC Fetal and Neonatal Ed. 99 (Suppl 1). 10 United Kingdom Haemophilia Centre Doctors’ Organisation (2015). Clinical Genetics Services for Haemophilia. United Kingdom Haemophilia Centre Doctors’ Organisation. Available at: www.ukhcdo.

11

12

13

14

15

16

17

18

19

org/wp‐content/uploads/2015/12/ Guidelines_on_genetics_services_for_ haemophilia_v5‐3_1_final.pdf. Davies, J. (2016). Women with Inherited Bleeding Disorders and their Offspring – The Unresolved Issues, in EGA Institute for Women’s Health. London: University College London. Liao, C., Wei, J., Li, Q. et al. (2006). Efficacy and safety of cordocentesis for prenatal diagnosis. Int. J. Gynaecol. Obstet. 93: 13–17. Deepika, D., Vatsla, D., Kallol, R. K. et al. (2012). Indications of 1342 Fetal cord blood sampling procedures performed as an integral part of high risk pregnancy care. J. Obstet. Gynecol. 62 (1): 20–24. Elejalde, B. R., de Elejalde, M. M., and Heitman, T. (1985). Visualization of the fetal genitalia by ultrasonography: a review of the literature and analysis of its accuracy and ethical implications. J. Ultrasound Med. 4: 633–639. Whitlow, B. J., Lazanakis, M. S., and Economides, D. L. (1999). The sonographic identification of fetal gender from 11 to 14 weeks of gestation. Ultrasound Obstet. Gynecol. 13: 301–304. Mazza, V., di Monte, I., Pati, M. et al. (2004). Sonographic biometrical range of external genitalia differentiation in the first trimester of pregnancy: analysis of 2593 cases. Prenat. Diagn. 24: 677–684. Lo, Y. M., Corbetta, N., Chamberlain, P. F. et al. (1997). Presence of fetal DNA in maternal plasma and serum. Lancet 350: 485–487. Hudecova, I., Sahota, D., Heung, M. M. et al. (2014). Maternal plasma fetal DNA fractions in pregnancies with high and low risks for fetal chromosome aneuploidies. PLoS One 9: e88484. Yu, S. C., Lee, S. W., Jiang, P. et al. (2013). High‐resolution profiling of fetal DNA clearance from maternal plasma by massively parallel sequencing. Clin. Chem. 59 (8): 1228–1237.

Antenatal Diagnosis

20 Mibashan, R. S., Rodeck, C. H., Thumpston,

21

22

23

24

25

26

27

28

29

J. K. et al. (1979). Plasma assay of fetal factors VIIIC and IX for prenatal diagnosis of haemophilia. Lancet 1: 1309–1311. Daffos, F., Capella‐Pavlovsky, M., and Forestier, F. (1983). A new procedure for fetal blood sampling in utero: preliminary results of fifty‐three cases. Am. J. Obstet. Gynecol. 146: 985–987. Sacchi, E., Randi, A. M., Tagliavacca, L. et al. (1992). Carrier detection and prenatal diagnosis of hemophilia A: 5 years experience at a hemophilia center. Int. J. Clin. Lab. Res. 21: 310–313. Caprino, D., Acquila, M., and Mori, P. G. (1993). Carrier detection and prenatal diagnosis of hemophilia B with more advanced techniques. Ann. Hematol. 67: 289–293. Acquila, M., Bottini, F., Valetto, A. et al. (2001). A new strategy for prenatal diagnosis in a sporadic haemophilia B family. Hemophilia 7: 416–418. Salviato, R., Belvini, D., Zanotto, D. et al. (2007). Prenatal diagnosis of haemophilia A by using intron 1 inversion detection. Hemophilia 13: 772–774. Hill, M., Lewis, C., Jenkins, L. et al. (2012). Implementing noninvasive prenatal fetal sex determination using cell‐free fetal DNA in the United Kingdom. Expert Opin. Biol. Ther. 12: S119–S126. Chi, C., Hyett, J. A., Finning, K. M. et al. (2006). Non‐invasive first trimester determination of fetal gender: a new approach for prenatal diagnosis of haemophilia. Br. J. Obstet. Gynaecol. 113: 239–242. Pavord, S., Rayment, R., Madan, B. et al., on behalf of the Royal College of Obstetricians and Gynaecologists(2017). Management of inherited bleeding disorders in pregnancy. Green‐Top Guideline No. 71. Br. J. Obstet. Gynaecol. 124: e193–e263. Tsui, N. B., Kadir, R. A., Chan, K. C. et al. (2011). Noninvasive prenatal diagnosis of hemophilia by microfluidics digital PCR analysis of maternal plasma DNA. Blood 117: 3684–3691.

30 Hudecova, I., Jiang, P., Davies, J. et al.

31

32

33

34

35

36

37

38

(2017). Noninvasive detection of F8 int22h‐related inversions and sequence variants in maternal plasma of hemophilia carriers. Blood 130: 340–347. Lun, F. M., Tsui, N. B., Chan, K. C. et al. (2008). Noninvasive prenatal diagnosis of monogenic diseases by digital size selection and relative mutation dosage on DNA in maternal plasma. Proc. Natl. Acad. Sci. U.S.A. 105: 19920–19925. Lo, Y. M. D., Chan, K. C., Sun, H. et al. (2010). Maternal plasma DNA sequencing reveals the genome‐wide genetic and mutational profile of the fetus. Sci. Transl. Med. 2: 61ra91. Hui, W. W. I., Jiang, P., Tong, Y. K. et al. (2017). Universal haplotype‐based noninvasive prenatal testing for single gene diseases. Clin. Chem. 63: 513–524. Laffan, M. A., Lester, W., O’Donnell, J. S. et al. (2014). The diagnosis and management of von Willebrand disease: a United Kingdom haemophilia centre doctors organization guideline approved by the British Committee for Standards in haematology. Br. J. Haematol. 167 (4): 453–465. Ash, K. M., Mibashan, R. S., and Nicolaides, K. H. (1988). Diagnosis and treatment of feto‐maternal hemorrhage in a fetus with homozygous von Willebrand’s disease. Fetal Ther. 3: 189–191. Peyvandi, F., Jayandharan, G., Chandy, M. et al. (2006). Genetic diagnosis of haemophilia and other inherited bleeding disorders. Hemophilia 12 (Suppl 3): 82–89. Ariffin, H., Millar, D. S., Cooper, D. N. et al. (2003). Prenatal exclusion of severe factor VII deficiency. J. Pediatr. Hematol. Oncol. 25: 418–420. Ingerslev, J., Herlin, T., Sorensen, B. et al. (2007). Severe factor X deficiency in a pair of siblings: clinical presentation, phenotypic and genotypic features, prenatal diagnosis and treatment. Haemophilia 13: 334–336.

187

188

Inherited Bleeding Disorders in Women

39 Killick, C. J., Barton, C. J., Aslam, S.,

40

41

42

43

44

45

46

47

and Standen, G. (1999). Prenatal diagnosis in factor XIII‐A deficiency. Arch. Dis. Child. Fetal Neonatal Ed. 80: F238–F239. Mota, L., Ghosh, K., and Shetty, S. (2007). Second trimester antenatal diagnosis in rare coagulation factor deficiencies. J. Pediatr. Hematol. Oncol. 29: 137–139. Tabibian, S., Shams, M., Naderi, M., and Dorgalaleh, A. (2018). Prenatal diagnosis in rare bleeding disorders—An unresolved issue? Int J Lab Hem. 40 (3): 241–250. Harper, J. C., Wilton, L., Traeger‐ Synodinos, J. et al. (2012). The ESHRE PGD consortium: 10 years of data collection. Hum. Reprod. Update 18 (3): 234–247. De Rycke, M., Goossens, V., Kokkali, G. et al. (2017). ESHRE PGD consortium data collection XIV‐XV: cycles from January 2011 to December 2012 with pregnancy follow‐up to October 2013. Hum. Reprod. 32 (10): 1974–1994. Harper, J. C., Coonen, E., de Rycke, M. et al. (2010). ESHRE PGD consortium data collection X: cycles from January to December 2007 with pregnancy follow‐up to October 2008. Hum. Reprod. 25 (11): 2685–2707. Laurie, A. D., Hill, A. M., Harraway, J. R. et al. (2010). Preimplantation genetic diagnosis for hemophilia A using indirect linkage analysis and direct genotyping approaches. J. Thromb. Haemost. 8 (4): 783–789. Renwick, P., Trussler, J., Lashwood, A. et al. (2010). Preimplantation genetic haplotyping: 127 diagnostic cycles demonstrating a robust, efficient alternative to direct mutation testing on single cells. Reprod. Biomed. 20 (4): 470–476. Goldman, K. N., Nazem, T., Berkeley, A. et al. (2016). Preimplantation genetic diagnosis (PGD) for monogenic disorders: the value of concurrent aneuploidy screening. Am. J. Med. Genet. 25: 1327e37.

48 Desmyttere, S., Bonduelle, M.,

49

50

51

52

53

54

55

56

57

58

Nekkebroeck, J. et al. (2009). Growth and health outcome of 102 2‐year‐old children conceived after preimplantation genetic diagnosis or screening. Early Hum. Dev. 85 (12): 755e9. Liebaers, I., Desmyttere, S., Verpoest, W. et al. (2010). Report on a consecutive series of 581 children born after blastomere biopsy for preimplantation genetic diagnosis. Hum. Reprod. 25 (1): 275e82. Kadir, R. A., Sabin, C. A., Goldman, E. et al. (2000). Reproductive choices of women in families with haemophilia. Hemophilia 6: 33–40. Tedgard, U., Ljung, R., and McNeil, T. F. (1999). Reproductive choices of haemophilia carriers. Br. J. Hematol. 106: 421–426. Varekamp, I., Suurmeijer, T. P., Brocker‐ Vriends, A. H. et al. (1990). Carrier testing and prenatal diagnosis for hemophilia: experiences and attitudes of 549 potential and obligate carriers. Am. J. Med. Genet. 37: 147–154. Kadir, R. A., Economides, D. L., Braithwaite, J. et al. (1997). The obstetric experience of carriers of haemophilia. Br. J. Obstet. Gynaecol. 104: 803–810. Chi, C., Lee, C. A., Shiltagh, N. et al. (2008). Pregnancy in carriers of haemophilia. Hemophilia 14: 56–64. Hoyer, L. W., Carta, C. A., Golbus, M. S. et al. (1985). Prenatal diagnosis of classic hemophilia (hemophilia A) by immunoradiometric assays. Blood 65: 1312–1317. Tedgard, U., Ljung, R., and McNeil, T. F. (1999). How do carriers of hemophilia and their spouses experience prenatal diagnosis by chorionic villus sampling? Clin. Genet. 55: 26–33. Balak, D. M., Gouw, S. C., Plug, I. et al. (2012). Prenatal diagnosis for haemophilia: a nationwide survey among female carriers in the Netherlands. Haemophilia 18 (4): 584–592. Karimi, M., Peyvandi, F., Siboni, S. et al. (2004). Comparison of attitudes towards

Antenatal Diagnosis

59

60

61

62

prenatal diagnosis and termination of pregnancy for haemophilia in Iran and Italy. Hemophilia 10: 367–369. Pandey, G. S., Panigrahi, I., Phadke, S. R., and Mittal, B. (2003). Knowledge and attitudes towards haemophilia: the family side and role of haemophilia societies. Community Genet. 6: 120–122. Peyvandi, F. (2005). Carrier detection and prenatal diagnosis of hemophilia in developing countries. Semin. Thromb. Hemost. 31: 544–554. Darby, S. C., Khan, S. W., and Spooner, R. J. (2007). Mortality rates, life expectancy and cause of death in people with haemophilia A and B in the UK who were not affected with HIV. Blood 110: 815–825. Sjogren, B. and Uddenberg, N. (1990). Prenatal diagnosis for psychological reasons: comparison with other

indications, advanced maternal age and known genetic risk. Prenat. Diagn. 10: 111–120. 63 Tedgard, U., Ljung, R., McNeil, T. et al. (1989). How do carriers of hemophilia experience prenatal diagnosis (PND)? Carriers’ immediate and later reactions to amniocentesis and fetal blood sampling. Acta Paediatr. Scand. 78: 692–700. 4 Tedgard, U., Ljung, R., McNeil, T. F., and 6 Tedgard, E. (1997). Identifying carriers at high risk for negative reactions when performing prenatal diagnosis of haemophilia. Hemophilia 3: 123–130. 5 Tedgard, U., Ljung, R., and McNeil, T. F. 6 (1999). Long‐term psychological effects of carrier testing and prenatal diagnosis of haemophilia: comparison with a control group. Prenat. Diagn. 19: 411–417.

189

191

12 Analgesia and Anesthesia for Pregnant Women with Inherited Bleeding Disorders Anne‐Sophie Bouthors1, Adrian England2, and Rezan A. Kadir3,4 1

Maternité Jeanne de Flandre, Academic Hospital, Department of Anesthesia and Intensive Care, Lille, France Royal Free London NHS Foundation Trust, Department of Anaesthesia, London, UK 3 Department of Obstetrics and Gynaecology and Katharine Dormandy Haemophilia and Thrombosis Centre, The Royal Free Foundation Hospital, London, UK 4 Institute for Women’s Health, University College London, London, UK 2

12.1 ­Introduction Labor, for many women, is one of life’s most intense and painful events. Pain in the first stage of labor is caused by uterine contractions and cervical dilation. Painful stimuli from the uterine contractions are transmitted to the posterior nerve roots from the 10th thoracic through to the first lumbar levels. Contraction pain is predominantly visceral, hence it is poorly defined and frequently referred to the lower abdomen and back. In the late first and second stages of labor, the pain predominantly results from stretching of the pelvic floor, vagina, and perineum. These painful stimuli are transmitted via the pudendal nerve derived from the second to the fourth sacral nerves. In contrast, delivery pain is somatic and is generally well localized and sharp. The experience of pain is multifactorial and cannot be accounted for solely by neurophysiological explanation. The International Association for the Study of Pain defines pain  as “an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in

terms of such damage” [1]. Therefore, the experience of pain, including labor pain, is highly individualized and multidimensional. The response to labor pain is modified by physiological as well as psychosocial and environmental factors. Although satisfaction  with childbirth is not dependent on the absence of pain, the level of pain and the effectiveness of pain relief ­ provided can influence women’s birth experience. Management of pain is an integral part of the care provided to all women in labor. It entails the provision of appropriate pain relief options with detailed information on their benefits and risks. A wide range of non‐­pharmacological and pharmacological options is available to assist women in c­ oping with labor pain. The decision to use a particular method is based on the woman’s preferences and her obstetric, fetal, anesthetic, and hemostatic risks. In women with inherited bleeding disorders, hemostatic risk is of particular concern. However, this risk is “physiologically” minimized by the pregnancy‐induced increase in some coagulation factors and can be further reduced by appropriate ­ prophylactic treatments. A significant rise in level and activity

Inherited Bleeding Disorders in Women, Second Edition. Edited by Rezan A. Kadir, Paula D. James, and Christine A. Lee. © 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.

192

Inherited Bleeding Disorders in Women

of factor VIII (FVIII:C) and von Willebrand factor antigen (VWF:Ag. VWF:Ac) is observed in all women during pregnancy, particularly in the third trimester [2, 3]. Consequently, most carriers of hemophilia A and women with type 1 von Willebrand disease (VWD) develop normal clotting factor levels by term. In contrast, levels of factors IX (FIX) and XI (FXI) do not alter significantly during pregnancy [2, 3]; thus, the coagulation defect in carriers of hemophilia B and in women with FXI deficiency usually persists throughout pregnancy. The hemostatic defects in women with platelet function disorders and severe clotting factor deficiencies, such as type 3 VWD, also remain unchanged in pregnancy. Therefore, it is important to assess coagulation during the third trimester, in order to create an individualized management plan for each woman. In addition to the assessment of coagulation status and factor levels, considerations of personal and family bleeding history are important in the risk assessment and decision for the need for hemostatic treatments. The aim of this chapter is to explore the options of obstetric analgesia and anesthesia specifically for women with inherited bleeding disorders and discuss strategies for a safe delivery of these options.

12.2 ­Non‐Pharmacological Methods A wide variety of non‐pharmacological methods have been used to help women cope with the pain of childbirth. Commonly used techniques include controlled breathing, massage, warm water bath, aromatherapy, acupuncture, acupressure, transcutaneous electrical nerve stimulation (TENS), hypnosis, music, and audio‐analgesia. However, there are only a few published scientific studies which have assessed their efficacy and adverse effects. Women may choose non‐pharmacological or complementary pain relief methods to avoid invasive or ­pharmacological measures and they are frequently used at the first

instance of pain and in the early stages of  labor. A  Cochrane systematic review of non‐pharmacological methods used by the general obstetric population concluded that acupuncture and hypnosis may help to relieve  labor pain, but there is currently ­insufficient evidence to support the effectiveness of ­massage and other complementary therapies [4]. There is limited published scientific data on the efficacy or safety of these non‐­ pharmacological methods in women with inherited bleeding disorders. However, non‐ invasive techniques such as controlled breathing and massage are generally regarded as safe for these women. However, techniques that involve disruption to maternal tissue, such as acupuncture, may be contraindicated because of the potential risk of bleeding or bruising.

12.3 ­Pharmacological Methods Pharmacological methods of controlling pain during labor and delivery have been used for  centuries. Techniques commonly used today include Entonox® (50% nitrous oxide premixed with 50% oxygen), opioid drugs, and regional block techniques such as epidural or combined spinal‐epidural. The use and popularity of these measures evolved concurrently with changes in social and cultural views about the pain of childbirth. Over time, some techniques have gained popularity while others, such as the Cardiff Penthrane inhaler delivering methoxyflurane, have been discontinued because of maternal or neonatal adverse effects. The primary concern when using any pharmacological method in women with inherited bleeding disorders lies in the potential risks of bleeding complications in the mother and newborn. 12.3.1  Inhalation Analgesia Volatile agents have been used to provide sedation and analgesia during childbirth for over

Analgesia and Anesthesia for Pregnant Women with IBDs

160 years. Inhalation analgesia became widely accepted when Queen Victoria received chloroform during the birth of her eighth and ninth children in 1853 and 1857. Nitrous oxide, in a hypoxic mixture of 80% nitrous oxide premixed in air, was first used for pain relief in labor in 1881. In 1934, Minnitt [5] introduced an apparatus allowing self‐administration of nitrous oxide. Other volatile agents have also been used, including enflurane, isoflurane, and sevoflurane, but nitrous oxide is the only volatile agent widely used in current obstetric practice for labor pain. This is due to its ease of administration, minimal toxicity and cardiovascular and respiratory effects, lack of flammability, odor, and effect on uterine contractions, and low cost. Entonox is delivered premixed as 50% nitrous oxide and 50% oxygen. It is self‐ administered and can be used at any stage of labor, depending on the woman’s needs and preferences. It has a fast onset of effect, approximately 50 seconds, and is reversed rapidly when inhalation is ceased. Although Entonox is not a potent analgesic, it can provide substantial pain relief for women in labor when applied properly. Clearance is predominantly by exhalation rather than metabolism and it does not cause prolonged sedation in the mother or the baby after delivery. Thus, it does not affect labor physiology or neonatal outcome and does not require additional maternal or fetal monitoring [6]. Side‐effects include nausea, vomiting, dizziness, and a dry mouth, but no major side‐effects are associated with its short‐term use. The non‐invasive mode of administration, low incidence of significant side‐effects, and absence of effect on bleeding tendency make Entonox a suitable option to help women with inherited bleeding disorders cope with the pain of labor. 12.3.2  Opioid Analgesia The use of systemic opioid analgesia to cope with labor pain is documented in ancient Chinese writings and it remains a common  option for women today. Meperidine

(­pethidine) is currently the opioid most ­frequently used for the relief of labor pain in  the UK. Other opioid drugs that have been assessed include morphine, tramadol, meptazinol, butorphanol, fentanyl, and remifentanil. Opioids provide some pain relief but are associated with maternal sedation, nausea, vomiting, dizziness, and delayed stomach emptying. They can be administered as an intramuscular or intravenous injection, or, in the case of remifentanil, as an intravenous patient‐controlled analgesia (PCA) technique. In women with bleeding disorders, especially those with severe coagulation defects, intramuscular injections can cause bleeding, bruising or a painful hematoma at the injection site.  Intravenous opioid administration is more likely to cause severe complications, such as respiratory depression, in both the mother and the newborn baby. However, any opioid given to the mother in the hours before delivery has the potential to cause respiratory depression in the neonate after delivery [7], irrespective of its mode of administration because it will cross the placenta. An opioid antagonist, such as naloxone, may then be required to reverse respiratory depression in the newborn. Naloxone is usually administered intramuscularly, which can cause intramuscular bleeding/hematoma in neonates with an inherited bleeding disorder. Other neonatal adverse effects of maternal opioid use include decreased alertness, inhibition of suckling and delay in feeding, and lower neurobehavioral scores [7]. Therefore, intramuscular opioid administration should be avoided in women with inherited bleeding disorders when the clotting factor levels have failed to normalize and have not been corrected to the normal range by hemostatic support because of complications at the injection site. And for a neonate with a bleeding disorder, maternal opioid analgesia by any route should avoided in the hours before delivery because of the potential for neonatal respiratory depression and the subsequent need for an intramuscular injection of the antidote naloxone.

193

194

Inherited Bleeding Disorders in Women

Remifentanil, an ultrashort‐acting synthetic opioid, is also used to provide pain relief in labor. It is most commonly given through a PCA device during labor in the UK [8]. It has an effect‐site half‐life for analgesia of 1.3 minutes and a context‐sensitive half‐life (the estimated time required for a 50% reduction in blood concentration after stopping an infusion that reached a steady state) of approximately three minutes in the general population. It is rapidly hydrolyzed by non‐specific tissue and blood esterases to inactive metabolites. It crosses the placenta, but appears to be rapidly metabolized or redistributed in the fetus [9] and its rapid onset and offset of action potentially make it an ideal opioid for PCA during labor, and provide better analgesia compared to meperidine given either intramuscularly or via intravenous PCA apparatus [10]. The side‐ effects are minor and include mild sedation, itching, and nausea, with no evidence of respiratory depression, cardiotocograph abnormalities requiring intervention in laboring women, or adverse neonatal outcome [11]. Remifentanil administered as PCA has been used effectively as an alternative labor analgesic for women with a contraindication for regional analgesia, including parturients with coagulopathies [12]. 12.3.3  Regional (Neuraxial) Analgesia and Anesthesia Regional blockade for pain relief in labor has been used since early 1900, but the “loss of resistance technique” for cannulation of the extradural space was not developed until 1921. Ureteric silk catheters and Tuohy’s curved bevel needle were developed in the 1940s, allowing repeated top‐ups into the extradural space without the need for recannulation and making continuous extradural analgesia for labor possible. These techniques require instrumentation of the spinal canal so drugs can be injected into the intrathecal or epidural spaces, using a single‐ shot injection, repeated boluses or continuous infusion through a catheter. Quality of analgesia and anesthesia has been enhanced

by the addition of opiate drugs to the local anesthetic solutions, and the introduction of patient‐controlled epidural analgesia (PCEA) techniques. By the early part of the twenty‐first century, approximately 21% of women in the UK and 58% of women in the USA used a regional analgesia technique during labor and delivery [13, 14]. Regional analgesia is currently the most effective form of analgesia for labor pain and does not cause the sedation and neonatal respiratory depression associated with opioid analgesia. Regional anesthesia has also become a preferred option during cesarean deliveries. There has been a marked shift from general anesthesia to regional anesthesia for operative deliveries because of its better safety record, minimizing the risk of failed intubation, ventilation, and aspiration. Regional anesthesia is also associated with earlier postoperative mobilization, establishment of breastfeeding, and gastrointestinal function. The proportion of cesarean sections performed under general anesthesia in the UK fell from over 50% in 1989–1990 to less than 10% in 2003–2004 [13]. 12.3.3.1  Techniques of Regional (Neuraxial) Blockade

Regional block techniques for delivery include epidural, spinal, or a combination of both. Epidural blockade involves the administration of repeated doses of anesthetic drugs through a catheter into the extradural space where they act on the nerve roots. Obstetric epidurals are inserted in the lumbar region, usually below the conus medullaris at a point where the spinal cord divides to form the filum terminale, and typically at the interspace between the second and third or the third and fourth lumbar vertebrae. The intention is to block electrical transmission along neurons that run from the uterus and the birth canal and, for cesarean sections, neurons running to the muscles of the anterior abdominal wall as they cross the epidural space in the nerve roots. Blockade of the dermatomes supplying the uterus (T10 to L1) is required for labor pain relief and

Analgesia and Anesthesia for Pregnant Women with IBDs

blockade of sacral nerve roots that s­upply the vagina and perineum is required for delivery. For a cesarean section, the block needs to be more intense and spread from the sacral area, which provides sensation from the bladder, continuously to the level of the fourth thoracic vertebrae to inhibit nerve fibers from the peritoneum. The volume and concentration of local anesthetic solution can be titrated to achieve an appropriate level of analgesia for each procedure. Recently, there has been a trend to use mixtures of opiates and less concentrated local anesthetics for labor analgesia. These provide analgesia while preserving maternal motor function and proprioception, thus enabling the women to mobilize during labor [15]. A spinal block involves the injection of anesthetic drugs into the subarachnoid space and takes effect more quickly than an epidural. It is not usually used on its own to provide pain relief in labor because it is a single‐shot technique, but is commonly used for procedures such as cesarean section and manual removal of a retained placenta. A combined spinal epidural (CSE) technique involves inserting both a spinal and an epidural catheter. The initial administration of an intrathecal dose of local anesthetic/opiate mixture through a spinal needle gives a rapid effect and the simultaneous insertion of a catheter into the epidural space through a Tuohy needle allows repeated epidural doses to be given. This technique is particularly useful in late labor and for instrumental or complex cesarean deliveries when further doses through the epidural catheter may be required. 12.3.3.2  Risks and Complications of Regional (Neuraxial) Blockade

Hypotension may develop soon after the administration of the local anesthetic agent and occurs secondary to blood vessel dilation as the local anesthetic causes the sympathetic vascular tone to fall, resulting in pooling of blood in venous capacitance vessels and arteriolar dilation leading to a reduced

afterload. There may also be a contribution to hypotension from obstruction of venous return from the vena cava secondary to uterine compression. To mitigate these, it is important to ensure adequate hydration, monitor blood pressure, and avoid maternal supine position in women receiving a regional block. Nausea occurs and usually responds to correction of hypotension, and pruritus is related to opioid use. An epidural can also influence the course of labor. A recent Cochrane review of epidural versus non‐epidural or no analgesia in labor showed that women randomized to epidural had a longer second stage of labor, an increased need for oxytocin, and a higher rate of instrumental deliveries [16]. Shivering, fever, drowsiness, and urinary retention have also been reported with regional block. Other potential complications of regional block include inadequate or failure of block, inadvertent puncture of a blood vessel, and accidental dural puncture. Dural puncture can cause headache, which may be self‐limiting, but when it is severe and persistent the ­injection of autologous blood into the epidural space to form a patch over the puncture can improve the symptoms. Serious but rare complications include local anesthetic toxicity, inadvertent high epidural or total spinal block, infection, spinal epidural hematoma, and neurological complications ranging from temporary nerve trauma to permanent spinal cord paralysis. 12.3.3.3  Coagulopathy and the Risk of Neurological Complications

The arterial supply to the lower spinal cord is from the artery of Adamkievicz, usually as a branch of a left intercostal artery or lumbar aorta, with little or no collateral supply, leaving it vulnerable to ischemia if the artery is compromised. Venous drainage is through a venous plexus, which drains into the vertebral and ascending lumbar veins and then to  the azygous and hemiazygous veins, so obstruction of the epidural veins leads to venous congestion in the spinal cord, resulting in ischemia. Therefore, the spinal canal is

195

196

Inherited Bleeding Disorders in Women

a closed, highly vascular and inaccessible area and any vascular injury inside it has the potential to cause a hematoma that cannot be mechanically decompressed from outside. In pregnancy, epidural vessels appear dilated because they offer an easy anastomosis between the azygos venous systems, so late pregnancy is potentially a time of increased risk for vascular trauma in the spinal vessels. Epidural hematoma may occur if instrumentation of the epidural space or catheter insertion through a Tuohy needle is complicated by puncture of an epidural vessel. When epidural vessel puncture occurs, clot formation is crucial to stop ongoing bleeding in this closed space. If epidural bleeding is not stopped, an extensive epidural hematoma may occur, leading to compression of the spinal cord and nerve roots. This compression may be either mechanical, due to the volume of the hematoma in a closed space, or ischemic, due to the compression of the venous plexus or artery supplying the medulla and roots. In the presence of normal coagulation, inadvertent traumatic epidural vascular bleeding sustained during the administration and removal of a regional block is usually self limiting and causes no complications. In a woman with defective coagulation, bleeding is more likely to continue and result in formation of an epidural hematoma. An extensive hematoma in the spinal canal has the potential to cause an increase in pressure, locally obstructing the venous drainage from the spinal cord and disrupting the arterial supply. This exposes the spinal cord to the risk of ischemia, which in its most severe form leads to nerve cell death and permanent paralysis. The risk of spinal and epidural hematomas after neuraxial block is rare in both the ­general and obstetric populations. The third National Audit Project reports the incidence of vertebral canal hematoma after central neuraxial blockade in all patients as 0.85/100 000 (95% CI 0.0–1.8 per 100 000) [17]. A multicenter review of over 500 000 obstetric

regional blocks in the UK reported only one epidural hematoma requiring s­ urgical decompression with an “improving” neurological recovery [18]. There was also one epidural abscess requiring surgical decompression, also with “improving” neurological recovery, and two women who suffered irreversible quadraplegia. One was caused by “anterior spinal artery syndrome” 12 hours after her epidural had worn off. The other occurred 10 days after delivery due to thrombosis of a congenital hemangioma. In a 10‐year review of neuraxial blocks in Sweden, 1 260 000 spinal and 450 000 epidurals were performed, including 200 000 epidurals for labor analgesia; the review identified 33 spinal hematomas [19]. The authors calculate an incidence of epidural hematoma of 1 per 156 000 after spinal anesthesia and 1 per 18 000 after epidural anesthesia. The risk of hematoma was much lower following epidural analgesia in the obstetric population (1 per 200 000) compared with female orthopedic patients undergoing knee arthroplasty (1 per 3600). In 2013, a nationwide survey from 11 American hospitals also reported a lower risk of spinal hematoma and neurological complications with obstetric neuraxial blockade. The authors identified seven cases of epidural hematoma requiring laminectomy out of 62 450 orthopedic procedures, but none out of 79 837 epidurals for obstetric indications [20]. Out of the seven orthopedic cases requiring laminectomy, four received anticoagulation/ antiplatelet therapy that deviated from American Society of Regional Anesthesia guidelines. In a review by Vandermeulen et al. [21], 42 (69%) of the 61 cases of spinal hematoma related to regional blocks identified in the general population occurred in patients with evidence of hemostatic abnormality, mostly attributable to anticoagulants. Five of these were pregnant women. Spinal/epidural hematoma in obstetric patients has also been reported in women with coagulation abnormalities associated with pre‐eclampsia or obstetric cholestasis [22].

Analgesia and Anesthesia for Pregnant Women with IBDs

12.3.3.4  Regional Blockade in Women with Inherited Bleeding Disorders

The above studies suggest that a coagulation abnormality increases the risk of epidural or spinal hematoma. Thus, there is increased concern about this risk in women with inherited bleeding disorders due to their coagulation defect. However, it is difficult to make accurate estimates of the incidence because of the rarity of inherited bleeding disorders and infrequent nature of the complication. Published literature is limited to case series with ill‐defined denominators. A review of 12 case reports and small case series in the English literature of the use of regional block during labor and delivery in 60 women with inherited bleeding disorders (34 with VWD, 19 FXI deficiency, six carries of hemophilia, and one woman with hemophilia) reported no complications [23]. In most cases, women with mild VWD or carriers of hemophilia A did not require prophylactic cover, whereas those with moderate or severe VWD received VWF‐containing FVIII concentrate. Dhar et al. [24] reported a case of uncomplicated regional block in a woman with hemophilia A (FVIII:C  1.0 g L−1 throughout pregnancy. c) Aim to maintain > 3 IU dL−1 throughout pregnancy.

local anesthetic around the cervix. Repeated administration may be required because of its relatively short duration of effect. Although it is easy to perform, concerns over its safety in the fetus and the availability of other effective alternatives have diminished its popularity. Many studies have reported an association between its use and fetal bradycardia and acidosis. The cause of these effects is uncertain, but the proposed mechanisms include vasoconstriction of the uterine vessels, leading to a

reduction in uterine perfusion and fetal hypoxia, or a direct toxic effect via placental transfer of local anesthetic [33]. 12.3.4.2  Pudendal Block

This is primarily used to relieve pain during the second stage of labor. This technique involves the injection of local anesthetic solution into the bilateral pudendal nerves in the pelvis, through a transvaginal or, less commonly, a transperineal approach. It can

Analgesia and Anesthesia for Pregnant Women with IBDs

provide pain relief for instrumental and spontaneous vaginal deliveries, and for episiotomy. Potential complications include systemic toxicity as a result of intravascular administration, vaginal and pelvic hematoma, and infection [34, 35]. In women with inherited bleeding disorders, paracervical and pudendal blocks are not advisable if the clotting defects have not been corrected. However, if the mother’s clotting defect has been corrected, she could be offered these options. Perineal infiltration with a local anesthetic is a suitable alternative for some procedures, such as episiotomy and its repair, where the greatest risk of a bleeding tendency may be due to the tissue trauma already present rather than the concomitant use of local anesthetic infiltration into the area. 12.3.5  General Anesthesia When a cesarean section is required for delivery, usually regional or general anesthesia is administered. In the last few decades, the number of cesarean sections performed in the UK has risen, but anesthetic mortality related to pregnancy has fallen significantly. In the triennium between 1970 and 1972, there were 37 anesthetic‐related maternal deaths, but regional blocks were performed in only two [36]. In the triennium between 2000 and 2002, there were six anesthetic‐ related deaths, all of which were associated with the use of general anesthesia [37], and between 2009 and 2013 there were two anesthetic‐related maternal deaths associated with general anesthesia [38]. General anesthesia in the obstetric population is associated with a higher incidence of difficult and failed intubation, and aspiration of gastric contents is of greater concern because of slower gastric emptying in pregnancy. Historically, these complications were the predominant cause of maternal mortality related to anesthesia. The reduction in anesthetic‐related maternal mortality over the last few decades is associated with an increase in use of regional anesthesia for cesarean

s­ ection alongside an increased awareness of the important anesthetic issues and better organization and resourcing of obstetric services. National UK recommendations are that more than 95% of category 4 (planned) cesarean sections and more than 85% of category 1 cesarean sections (performed where there is an immediate threat to the life of the mother or baby) be performed under central neuraxial blockade [27]. Although regional techniques are the preferred methods for providing anesthesia for cesarean sections, there are circumstances when general anesthesia is indicated. These include emergency procedures for which there is insufficient time to establish regional blockade, inadequate or failed regional anesthesia, and an uncorrected maternal bleeding diathesis for which an epidural is contraindicated. Postoperative pain is greater following general anesthesia because the analgesic benefits of an epidural are not available, and non‐steroidal drugs are contraindicated in women with bleeding disorders as they interfere with platelet function. A regimen containing regular paracetamol, codeine, and other opiate drugs can be given to reduce postoperative pain following general anesthesia. Opiate drugs can be administered intravenously through a PCA device and this is a suitable option for women with inherited bleeding disorders for whom intramuscular injections are contraindicated and postoperative ileus can make absorption via the oral route unreliable.

12.4 ­Conclusion Management of obstetric analgesia and anesthesia in women with inherited bleeding ­ disorders requires planning and risk assessment. Non‐pharmacological and non‐ invasive methods of pain relief are usually not contraindicated. Intramuscular opioids should be avoided in women with uncorrected coagulopathy because of the risk of bleeding and bruising in the mother. If the

201

202

Inherited Bleeding Disorders in Women

neonate has an uncorrected coagulopathy then maternal opioid use, administered by any route, is better avoided as reversing respiratory depression in the newborn with  the opioid antagonist (naloxone) involves an intramuscular injection in the neonate. Remifentanil PCA is an alternative opioid‐based labor analgesic technique, but further efficacy and safety data are required before its use can be recommended. It is possible to safely offer women with inherited ­bleeding disorders regional block techniques provided their coagulation defects have been corrected, either secondary to a spontaneous rise in the level of the deficient

clotting factor during pregnancy or because of prophylactic treatment. Regional techniques are contraindicated in women with uncorrected coagulopathy because of the increased risk of bleeding complications, notably the subsequent neurological sequelae. The use of a regional block in women with inherited bleeding disorders should be  preceded by discussions about its risks and benefits, assessment of coagulation ­status, and planning for any prophylactic treatment required. Close collaboration between hematologists, anesthetists, and obstetricians is important to ensuring optimal outcomes.

­References 1 Mersky, H. (1979). Pain terms: a list with

2

3

4

5 6

7

definitions and a note on usage. Recommended by the International Association for the Study of Pain (IASP) Subcommittee on Taxonomy. Pain 6: 249–252. Kadir, R.A., Economides, D.L., Braithwaite, J. et al. (1997). The obstetric experience of carriers of haemophilia. Br. J. Obstet. Gynaecol. 104: 803–810. Kadir, R.A., Lee, C.A., Sabin, C.A. et al. (1998). Pregnancy in women with von Willebrand’s disease or factor XI deficiency. Br. J. Obstet. Gynaecol. 105: 314–321. Smith, C.A., Collins, C.T., Cyna, A.M., and Crowther, C.A. (2006). Complementary and alternative therapies for pain management in labour. Cochrane Database Syst. Rev. (4): CD003521. Minnitt, R. (1934). Self‐administered anaesthesia in childbirth. BMJ 1: 501–503. Stefani, S., Hughes, S., Shnider, S.M. et al. (1982). Neonatal neurobehavioral effects of inhalation analgesia for vaginal delivery. Anesthesiology 56: 351–355. Wiener, P.C., Hogg, M.I., and Rosen, M. (1979). Neonatal respiration, feeding and neurobehavioural state. Effects of intrapartum bupivacaine, pethidine and pethidine reversed by naloxone. Anaesthesia 34: 996–1004.

8 Saravanakumar, K., Garstang, J.S., and

Hasan, K. (2007). Intravenous patient‐ controlled analgesia for labour: a survey of UK practice. Int. J. Obstet. Anesth. 16: 221–225. 9 Kan, R.E., Hughes, S.C., Rosen, M.A. et al. (1998). Intravenous remifentanil: placental transfer, maternal and neonatal effects. Anesthesiology 88: 1467–1471. 10 Blair, J.M., Dobson, G.T., Hill, D.A. et al. (2005). Patient controlled analgesia for labour: a comparison of remifentanil with pethidine. Anaesthesia 60: 22–27. 11 Volikas, I., Butwick, A., Wilkinson, C. et al. (2005). Maternal and neonatal side‐effects of remifentanil patient‐controlled analgesia in labour. Br. J. Anaesth. 95: 504–509. 12 Novoa, L., Navarro, E.M., Vieito, A.M. et al. (2003). Obstetric analgesia and anesthesia with remifentanyl in a patient with von Willebrand disease. Rev. Esp. Anestesiol. Reanim. 50: 242–244. 13 Department of Health (2005). Statistical Bulletin – NHS Maternity Statistics, England: 2003–2004. London: Department of Health. 14 Declerq, E., Sakala, C., Corry, M. et al. (2002). Listening to Mothers: Report of the First National Survey of Women’s Childbearing Experiences. New York:

Analgesia and Anesthesia for Pregnant Women with IBDs

15

16

17

18

19

20

21

22

23

24

Maternity Center Association/Harris Interactive. Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK (2001). Effect of low dose mobile versus traditional epidural techniques on mode of delivery: a randomised control trial. Lancet 358: 19–23. Anim‐Somuah, M., Smyth, R., and Howell, C. (2005). Epidural versus non‐epidural or no analgesia in labour. Cochrane Database Syst. Rev. (4): CD000331. Cook, T.M., Counsell, D., and Wildsmith, J.A. Royal College of Anaesthetists Third National Audit Project(2009). Major complications of central neuraxial block: report on the Third National Audit Project of the Royal College of Anaesthetists. Br. J. Anaesth. 102: 179–190. Scott, D.B. and Hibbard, B.M. (1990). Serious non‐fatal complications associated with extradural block in obstetric practice. Br. J. Anaesth. 64: 537–541. Moen, V., Dahlgren, N., and Irestedt, L. (2004). Severe neurological complications after central neuraxial blockades in Sweden 1990–1999. Anesthesiology 101 (4): 950–959. Bateman, B.T., Mhyre, J.M., Ehrenfeld, J. et al. (2013). The risk and outcomes of epidural hematomas after perioperative and obstetric epidural catheterization: a report from the Multicenter Perioperative Outcomes Group Research Consortium. Anesth. Analg. 116 (6): 1380–1385. Vandermeulen, E.P., van Aken, H., and Vermylen, J. (1994). Anticoagulants and spinal‐epidural anesthesia. Anesth. Analg. 79: 1165–1177. Abramovitz, S. and Beilin, Y. (2003). Thrombocytopenia, low molecular weight heparin, and obstetric anesthesia. Anesthesiol. Clin. North Am. 21: 99–109. Chi, C., Lee, C.A., England, A. et al. (2009). Obstetric analgesia and anaesthesia in women with inherited bleeding disorders. Thromb. Haemost. 101: 1104–1111. Dhar, P., Abramovitz, S., DiMichele, D. et al. (2003). Management of pregnancy in

25

26

27

28

29

30

31

32

33

34

a patient with severe haemophilia A. Br. J. Anaesth. 91: 432–435. Chi, C., Kulkarni, A., Lee, C.A., and Kadir, R.A. (2009). The obstetric experience of women with factor XI deficiency. Acta Obstet. Gynecol. Scand. 88: 1095–1100. Verghese, Y., Tingi, E., Thachil, J. et al. (2017). Management of parturients with factor XI deficiency – 10 year case series and review of literature. Eur. J. Obstet. Gynecol. Reprod. Biol. 215: 85–92. Working Party, Association of Anaesthetists of Great Britain and Ireland; Obstetric Anaesthetists’ Association; Regional Anaesthesia UK (2013). Regional anaesthesia and patients with abnormalities of coagulation: the Association of Anaesthetists of Great Britain and Ireland, the Obstetric Anaesthetists’ Association, Regional Anaesthesia UK. Anaesthesia 68: 966–972. Pavord, S., Rayment, R., Madan, B. et al. on behalf of the Royal College of Obstetricians and Gynaecologists(2017). Management of inherited bleeding disorders in pregnancy. Green‐Top Guideline No. 71. Br. J. Obstet. Gynaecol. 124: e193–e263. Kadir, R.A. and Davies, J. (2013). Hemostatic disorders in women. J. Thromb. Haemost. (11): 170–179. Haljamae, H. (1996). Thromboprophylaxis, coagulation disorders, and regional anaesthesia. Acta Anaesthesiol. Scand. 40: 1024–1040. Verniquet, A.J.W. (1980). Vessel puncture with epidural catheters. Experience in obstetric patients. Anaesthesia 35: 660–662. Wulf, H. (1996). Epidural anaesthesia and spinal haematoma. Can. J. Anaesth. 43: 1260–1271. Thiery, M. and Vroman, S. (1973). Fetal bradycardia after paracervical block analgesia in labor. Acta Anaesthesiol. Belg. 24: 288–292. Bozynski, M.E., Rubarth, L.B., and Patel, J.A. (1987). Lidocaine toxicity after maternal pudendal anesthesia in a term infant with fetal distress. Am. J. Perinatol. 4: 164–166.

203

204

Inherited Bleeding Disorders in Women

35 Kurzel, R.B., Au, A.H., and Rooholamini,

S.A. (1996). Retroperitoneal hematoma as a complication of pudendal block – diagnosis made by computed tomography. West. J. Med. 164: 523–525. 36 Department of Health and Social Security (1975). Report on Health and Social Subjects, No. 11. Report on Confidential Enquiries into Maternal Deaths in England and Wales 1970–1972. London: HMSO. 7 (2004). Why Mothers Die. Report on 3 Confidential Enquiries into Maternal

Deaths in the United Kingdom 2002–2004. London: RCOG Press. 8 Knight, M., Nair, M., Tuffnell, D. et al. 3 (eds.) on behalf of MBRRACE‐UK(2017). Saving Lives, Improving Mothers’ Care – Lessons Learned to Inform Maternity Care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. Oxford: National Perinatal Epidemiology Unit, University of Oxford.

205

13 The Newborn Manuel Carcao 1 and Vanessa Bouskill 2 1

Division of Haematology/Oncology, Department of Paediatrics and Child Health Evaluative Sciences, Research Institute, Hospital for Sick Children, Toronto, Ontario, Canada 2 Department of Nursing, Hospital for Sick Children, Toronto, Ontario, Canada

13.1 ­Introduction Bleeding problems in the newborn can have severe consequences. The newborn is in a very vulnerable state from a bleeding per­ spective and has to survive myriad bleeding challenges stemming from the birth process itself as well as from blood sampling, intra­ muscular injections, vaccinations, and in some cases elective procedures such as cir­ cumcisions. All newborns are in a precarious balance of bleeding and clotting arising from  having reduced levels of procoagulant and anticoagulant factors. Any alteration of this balance can predispose to bleeding. Furthermore, newborns can be born with congenital bleeding disorders, which indi­ vidually are rare but collectively are fairly common and can place the child at further risk of bleeding. Fortunately, most congenital bleeding disorders are mild in nature and by themselves are not likely to place the new­ born at much increased risk; however, severe subtypes of bleeding disorders can present with life‐threatening hemorrhage at birth. Therefore, it is very important to have a good understanding of hemostasis in the newborn, and for clinicians to suspect an underlying

congenital bleeding disorder in every child with unexpected or abnormal bleeding, even when there is a negative family history of bleeding (Figure 13.1). In this chapter, normal hemostasis in full‐ term and preterm newborns will be dis­ cussed. This knowledge is important not only to be able to diagnose congenital bleed­ ing disorders in the newborn but also to pre­ vent misdiagnosing normal healthy children as having a bleeding disorder due to lack of awareness of normal neonatal hemostasis. In addition, the challenges of delivery/birth and early postpartum bleeding complications will be outlined. Lastly, the presentation of con­ genital bleeding disorders in neonates (first 28 days of life) and infants and their manage­ ment will be reviewed.

13.2 ­Developmental Hemostasis Hemostasis is a dynamic system beginning in utero and evolving throughout the neo­natal period. Studies have provided age‐dependent reference values that delineate developmen­ tal  features of hemostasis, both primary

Inherited Bleeding Disorders in Women, Second Edition. Edited by Rezan A. Kadir, Paula D. James, and Christine A. Lee. © 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.

206

Inherited Bleeding Disorders in Women Screening Tests: CBC (peripheral smear), PT/INR, aPTT, fibrinogen activity

Low platelet count PT/INR normal aPTT normal

CBC normal PT/INR abnormal aPTT normal

CBC normal PT/INR normal aPTT abnormal

CBC normal PT/INR normal aPTT normal

CBC normal, PT/INR abnormal aPTT abnormal

Assess platelet size Large Normal Small

Deficiency of FVII

Specific deficiencies of FXII*, FXI, FIX, FVIII

Disorders of platelet function Deficiency of FXIII

Specific deficiencies of FX, FV, FII, FI Liver disease Vitamin K deficiency

Test for platelet dysfunction (PFA-100/200, platelet aggregations) FXIII testing

Large platelets

Normal size platelets

Small platelets

1. 2. 3. 4.

1. Maternal ITP 2. Neonatal allo-immune thrombocytopenia 3. Congenital amegakaryocytic thrombocytopenia (CAMT) 4. Thrombocytopenia with absent radii (TAR)

1. 2.

5. 6. 7. 8. 9.

Bernard-Soulier syndrome Gray platelet syndrome MYH9-related thrombocytopenia X-linked macrothrombocytopenia with dyserythropoiesis Mediterranean macrothrombocytopenia DiGeorge syndrome Platelet-type (pseudo) von Willebrand disease Paris-Trousseau thrombocytopenia & Jacobsen syndrome Macrothrombocytopenia with platelet expression of glycophorin A

Wiskott-Aldrich syndrome X-linked thrombocytopenia

Figure 13.1  (a) Diagnostic algorithm for use in neonates with suspected congenital bleeding disorders. (b) Platelet disorders characterized by platelet size. *FXII deficiency does not result in a bleeding disorder. aPTT, activated partial thromboplastin time; CBC, complete blood count; INR, international normalized ratio; ITP, immune thrombocytopenic purpura; PFA, platelet function analyzer; PT, prothrombin time.

(platelet‐related hemostasis characterized by adhesion of platelets to s­ ub­endothelial matrix followed by platelet aggregation) and second­ ary (production of fibrin by the coagulation cascade) [1, 2]. The reference values of hemostatic param­ eters in newborns are very different from those of older children and adults. Interpre­ tation of laboratory values in a newborn can be a challenge, since physiological levels of many coagulation proteins are low and this makes it difficult to differentiate a mild bleeding disorder from appropriately low coagulation protein values for age. However, severe deficiencies of most coagulation

f­actors may be correctly diagnosed immedi­ ately after birth. 13.2.1  Primary Hemostasis in the Newborn Platelets are present at five weeks of gesta­ tion and reach adult values (≥ 150 × 109/L) by the end of the first trimester. By 10 weeks’ gestation, megakaryocytes can be detected in the liver and spleen; by the second trimes­ ter, they are present in the bone marrow. The number, size, and ultrastructure of platelets of both full‐term and premature neonates are similar to those of adults. Functionally,

The Newborn

platelets in newborns show mild impair­ ments of aggregation and secretion [3]. In contrast, newborns show higher hema­ tocrits, higher levels of von Willebrand factor (VWF), and a greater proportion of active ultra‐large (ULVWF) VWF multimers than older children/adults. This may help to com­ pensate for the mild impairments in platelet function. Consequently, primary hemostasis is overall not that different in newborns than it is in older children and adults. 13.2.2  Secondary Hemostasis in the Newborn In contrast to primary hemostasis, second­ ary hemostasis is substantially different in newborns versus older children and adults. Coagulation proteins do not cross the pla­ centa but are synthesized by the fetus start­ ing at week 10 of gestation. Most clotting factors and factors involved in the fibrino­ lytic system are produced in the liver. The exceptions are factors (F) VIII, VWF, tissue factor pathway inhibitor (TFPI), and throm­ bomodulin. These are all produced in differ­ ent types of endothelial cells; additionally, VWF and TFPI are also produced in mega­ karyocytes. FV is mainly produced in the liver but additionally some is produced in megakaryocytes. While the B subunit of FXIII is produced in the liver, the A subunit is produced by marrow‐derived cells (mono­ cytes/macrophages). Liver immaturity, a feature of neonatal life, results in lower levels of most coagula­ tion factors; with the exception of FV, FVIII, FXIII, VWF, and fibrinogen, the levels of most coagulation factors (including contact factors) are reduced (in most cases to approximately 50% of normal adult levels) at birth. This is particularly the case with the vitamin K‐dependent factors (II, VII, IX, and X), which are markedly affected due to immaturity of the hepatic γ‐carboxylation process. For these, normal neonatal levels can be as low as 20% of adult norms. For ref­ erence ranges of coagulation factors in healthy full‐term neonates, please refer to

the publications of Andrews et al. (1987) [1] and Monagle et  al. (2006) [2]. Coagulation factor levels are even further reduced in preterm neonates [4]. This has implications in diagnosing coagulation factor deficien­ cies. For example, young children may be  falsely diagnosed as having mild hemo­ philia B (FIX deficiency) when in fact their “low” FIX levels are simply physiological and appropriate for age. The same applies to FVII deficiency. Also, children who do have hemophilia B may be initially declared to have moderate hemophilia B but, with age, as their FIX levels rise they might be reclas­ sified as having mild hemophilia B. As liver immaturity does not affect FVIII and VWF levels, both are slightly elevated in the neonate due to the stress of delivery and remain elevated in comparison to adult val­ ues until approximately three months of age. Neonates have decreased fibrinolytic capac­ ity as a result of reduced levels of plasmino­ gen, leading to reduced ability to generate plasmin. This contributes to a reduced ability of neonates to fibrinolyze (i.e. break down clots); it is because of this that giving plasma as a source of plasminogen to newborns may assist with fibrinolysis. The consequence of all of these differen­ tial levels of coagulation factors is that both the prothrombin time (PT)/international normalized ratio (INR) and especially the activated partial thromboplastin time (aPTT) of newborns are prolonged in rela­ tion to adult norms; both are even more pro­ longed in premature newborns. However, this ­ physiological prolongation of INR/ aPTT is not associated with higher risk of bleeding in neonates. Also, the slightly reduced platelet function is offset by high levels of VWF and high hematocrits to an extent that bleeding times and platelet func­ tion analyzer (PFA‐100/200) closure times are significantly shorter in healthy neonates than in older children and adults. Functional levels of coagulation proteins gradually increase during gestation and continue to do so at a rapid pace in the first six months of life, almost attaining normal

207

208

Inherited Bleeding Disorders in Women

adult levels by six months of age. True adult levels of the vitamin K‐dependent factors are, however, not attained until a child reaches puberty. If it were not for the fact that levels of natural anticoagulants (inhibi­ tors of coagulation: protein C, protein S, and antithrombin) are similarly reduced at birth, the newborn would be at significant risk of bleeding. These balanced physiologi­ cal age‐related changes in the coagulation system provide effective protection to the healthy neonate. However, this balanced state of hemostasis can be easily disturbed, making the neonate susceptible to both bleeding and thrombosis. Furthermore, the lower levels of procoagulants and anticoag­ ulants in newborns can affect the manage­ ment of both bleeding and clotting. For example, low antithrombin activity can lead to a relative resistance to therapeutic ­heparin in neonates, resulting in the need for much higher heparin dosing in newborns. Understanding these age‐related differ­ ences in concentration and function of coag­ ulation proteins is essential when evaluating a newborn with abnormal bleeding or throm­ bosis (clot).

13.3 ­Laboratory Hemostatic Evaluation of the Neonate Laboratory evaluation and interpretation in neonates is challenging due to difficulty in obtaining blood samples and adequate vol­ umes for testing. Additionally, variability in ranges of neonatal coagulation parameters needs to be considered when interpreting the results of laboratory tests. As coagulation factors do not cross the placental barrier, levels of coagulation fac­ tors in cord blood are representative of neo­ natal levels and can be used to check the coagulation status of newborns at birth and diagnose disorders (e.g. hemophilia or FXIII deficiency). Specimens from cord blood are of great value in neonates known to be at

risk of a congenital bleeding disorder and obtaining them should be planned prior to  delivery whenever possible. A particu­ lar  benefit of cord blood testing is that it avoids the need to perform venepunctures on n ­ ewborns  – a  procedure that is often quite difficult. Although cord blood can be an important source for diagnostic material in neonates, unfortunately, in 50% of new­ borns with severe hemophilia [5] and in most patients with autosomal recessive bleeding disorders, the lack of a known family history of a bleeding disorder results in the cord blood not being obtained and tested. As noted above, cord blood testing when a bleeding disorder is suspected at birth avoids obtaining a blood sample through peripheral venepuncture from a newborn [6]. It should be noted that difficult venepuncture in new­ borns may give misleading laboratory coagu­ lation results, especially for platelet function, and repeat samples may be required if there is evidence of hemolysis or a clot in the sam­ ple. A 21 gauge needle is ideal for venepunc­ ture (in preterm neonates, a 23 gauge needle may be used) to minimize stasis so as to avoid activation of the hemostatic system. Underfilling of the sodium citrate blood col­ lection tube is a common occurrence and should be avoided, as this leads to insuffi­ cient plasma levels on which to perform test­ ing and inaccurate results. Capillary samples are unacceptable for any coagulation testing due to potential activation of the hemostatic system. The absolute values of reference ranges for coagulation assays can vary with ana­ lyzer and reagent systems in neonates and children. In newborns presenting with bleeding without a specific known bleeding disorder within the family, initial laboratory testing to rule out a congenital bleeding disorder should include a complete blood count (CBC) with peripheral smear (to evaluate platelet number and morphology), meas­ urements of clotting capacity including PT/ INR, aPTT, and fibrinogen activity. When

The Newborn

the family history is suggestive of a parti­ cular disorder, then specific factor assays should be included in the testing. Genetic testing at the onset of a bleeding episode is seldom helpful since clinical deci­ sions and therapeutic interventions may be urgently required and consequently cannot wait on results of mutational analysis.

13.4 ­When to Suspect a Congenital Bleeding Disorder in a Newborn The clinical presentation of a congenital bleeding disorder presenting in neonates is different from bleeding in older children or adults and typically includes at least one of the following symptoms: extracranial hemor­ rhage (ECH) (e.g. cephalohematoma, sub­ galeal hematoma); intracranial hemorrhage (ICH); bleeding related to trauma from deliv­ ery (e.g. from use of forceps and/or vacuum); mucocutaneous type bleeding (e.g. pete­ chiae, purpura); post circumcision bleeding; and bleeding from the umbilicus. Musculos­ keletal bleeding (e.g. joint and muscle bleed­ ing) and epistaxis, which are commonly seen in older children with bleeding disorders, are unusual in neonates. There are two scenarios in which new­ borns may be suspected of having a bleeding disorder: those in whom there is a positive family history of a bleeding disorder and those without a family history but in whom bleeding prompts investigations. 13.4.1  Newborns with a Positive Family History of a Bleeding Disorder The most important thing in such cases is the anticipation of the newborn having a bleeding disorder and taking appropriate precautions to avoid the neonate bleeding. Prenatal diagnosis can be undertaken for most bleeding disorders to determine if the  fetus is affected. However, just because

prenatal testing is possible does not neces­ sarily imply that it should be done. This is a very individualized decision for the preg­ nant woman/family, taking into conside­ ration the risks and benefits of prenatal diagnosis for both the mother and the fetus. With respect to taking appropriate precau­ tions to avoid bleeding, often this involves making decisions regarding how best to deliver the child. 13.4.1.1  Prenatal Diagnosis of Congenital Bleeding Disorders

Prenatal diagnosis is possible when a family is known to be at risk of having a child affected with a specific bleeding disorder. For example, this can occur when a woman is a known/suspected hemophilia carrier and as such, has a 50% chance of having an affected son (if she is carrying a male fetus), or when a couple are known to be carriers of recessive mutations of coagulation proteins (e.g. factor VII, X, or XIII) and as such, have a 25% chance of having an affected child. In the for­ mer setting, it is critical to determine the fetal sex as, in general, female newborn carri­ ers are at a significantly lower risk of suffer­ ing from bleeding as neonates in comparison to newborn males with hemophilia. Prenatal diagnosis permits family coun­ seling/planning, allows for family education about the diagnosis, and allows for better planning for delivery. The early diagnosis of an affected fetus allows the parents to be pre­ sented with options: early termination of the pregnancy or proceeding to plan a safe deliv­ ery of the fetus. Prenatal diagnosis is usually performed by obtaining “fetal” material through amniocen­ tesis or chorionic villous sampling (CVS), sent for genetic testing to test for the known mutation(s) within the family. Prenatal test­ ing is associated with certain risks (e.g. fetal demise). If parents are not planning termina­ tion of the affected fetus, then the value of prenatal testing may not be justifiable to many parents given the risks of prenatal test­ ing. For such families, it is best to assume that the fetus is affected and plan the labor

209

210

Inherited Bleeding Disorders in Women

and delivery accordingly. Prenatal diagnosis is discussed in detail in Chapter 11. Another option for mothers who are known carriers of hemophilia or parents who are known carriers of autosomal reces­ sive disorders is to have a preimplantation embryo selection (PIES) in conjunction with in vitro fertilization (IVF). This allows for the selection of an unaffected embryo at very early stages prior to actual pregnancy. 13.4.1.2  Optimal Mode of Delivery of the Neonate with a Bleeding Disorder

The optimal mode of delivery for carriers of hemophilia (with an affected or possibly affected fetus) has typically been controver­ sial. Different modes of delivery carry differ­ ent risks for the mother and the newborn. It has been found that the incidence of ECH and ICH is lowest in planned cesarean sec­ tion deliveries; however, it must be noted that even planned cesarean sections do not completely eliminate the risk of ECH/ICH. The magnitude of the reduction in risk of ICH was the topic of a recent meta‐analysis which demonstrated that a planned cesarean section conveyed a lower risk of ICH (odds ratio 0.34; 95% confidence interval (CI) 0.14–0.83; P = 0.018) in comparison to vagi­ nal delivery [7]. In the same study, it was found that in newborns with hemophilia, the odds ratio of experiencing an ICH is 4.4‐ fold higher (95% CI 1.46–13.7; P  =  0.008) following an assisted vaginal delivery than an unassisted vaginal delivery. Yet cesarean section is associated with more maternal morbidity and may present hemostatic challenges for the mother. Consequently, ­ guidelines do not consistently recommend cesarean section over vaginal delivery. The mode of delivery and management plan should take into consideration both the mother and the fetal risk at the time of pres­ entation, prior labor experiences of the mother, if any, and the mother’s reproduc­ tive expectations after delivery. For optimal outcomes, the critical decisions and plans of care should include a multidisciplinary team

of hemophilia specialists, an obstetrician, and a pediatrician. There is universal agreement that all ­vaginal deliveries should avoid instrument‐ assisted delivery (forceps and vacuum), as well as fetal scalp monitoring and fetal scalp blood sampling. Vacuum extraction and high forceps seem to lead to the highest inci­ dence of ICH and ECH, with low forceps and unassisted vaginal delivery being the least problematic. Any neonate with a known family history of a bleeding disorder presenting with bleed­ ing symptoms should be assumed to be affected; diagnosis should be made promptly and treatment based upon the most likely diagnosis should be administered. In the case of severe bleeding (e.g. ICH/ECH), treatment should be instituted even prior to laboratory confirmation of the bleeding disorder. In newborns with a family history of a severe bleeding disorder, prophylactic clot­ ting factor concentrate administration at birth is controversial. In general, without evi­ dence of a bleed, most clinicians avoid such exposure although there is no consensus on this. Of course, in the setting of a traumatic delivery or of the child being premature, administration should certainly be consid­ ered since the risk of ICH in such a setting is higher. Any possible signs of an ICH in the newborn should be investigated and treated promptly with appropriate therapy. This should be undertaken before confirmation of an ICH by imaging. Significant diversity in practice exists with respect to whether imaging should be done for all asymptomatic newborns suspected or known to have significant bleeding disorders such as moderate or severe hemophilia. If imaging is undertaken, cranial ultrasonogra­ phy is typically the modality used as it is readily available and non‐invasive. However, due to the location of the fontanelles and their size, ultrasound is not sufficiently sensi­ tive to detect posterior bleeding or certain types of bleeds (e.g. subdural bleeding). Consequently, many organizations, including the United Kingdom Haemophilia Centre

The Newborn

Doctors’ Organisation, recommend that if a newborn is presenting with concerning signs for an ICH, that a magnetic resonance imag­ ing (MRI)/computed tomography (CT) scan be  performed if the cranial ultrasound is deemed to be within normal limits. Of note, female newborn carriers of hemo­ philia B are likely at higher risk of bleeding compared to newborn carriers of hemo­ philia A, due to FIX being a vitamin K‐depend­ ent factor (being low at birth) whereas FVIII is an acute‐phase reactant and as such is likely to be increased at birth. This needs to be taken into consideration when managing newborn females who are suspected to be carriers of hemophilia B. 13.4.2  Newborn Presenting with Bleeding Symptoms Clinicians should be aware that severe or unusual bleeding in otherwise well children can be a manifestation of an underlying bleed­ ing disorder and should be investigated. Although a bleeding disorder may occur in any child, there are certain ethnic groups in  which particular bleeding disorders are more  common and consequently should be considered (e.g. FXI deficiency in Ashkenazi Jews, Glanzmann thrombasthenia (GT) in French gypsies, and hemophilia B in some Amish communities) [8–10]. Additionally, in families in which there is consanguinity, the risk of autosomal recessive bleeding disor­ ders becomes much more common. It is important that in children demon­ strating bleeding or suspected to have a bleeding disorder, vitamin K be administered in order to prevent them having both a ­congenital bleeding disorder and an acquired bleeding disorder caused by vitamin K ­deficiency. The route of administration of ­vitamin K at birth is also controversial if a severe bleeding disorder is diagnosed antena­ tally. Most of the literature suggests that ­vitamin K should be given subcutaneously with pressure applied afterwards; however, it can be given orally. In the latter case, differ­ ent oral regimens have been recommended;

in  some cases as few as two doses are needed, one immediately at birth and a sec­ ond sometime in the first week of life [11]. The most common types of bleeding in neo­ nates with bleeding disorders are described below. 13.4.2.1  Extracranial and Intracranial Bleeding

Extracranial bleeding such as cephalohema­ toma and subgaleal hematoma are relatively common in a healthy newborn that has undergone a traumatic birth or an assisted delivery requiring high forceps and/or vac­ uum. In newborns without bleeding disor­ ders, these tend to not be very large. However, in newborns with severe congenital bleeding disorders such as hemophilia or FXIII defi­ ciency, these ECHs can not only be much more common but additionally can be quite large and may result in significant blood loss and consequent anemia. The most severe bleeding manifestation that a neonate can present with is an ICH. Although all bleeding disorders can poten­ tially increase the risk of ICH at birth, those most likely to cause an ICH are severe defi­ ciencies of FXIII, FX, FVII, FVIII, and FIX. In patients with FXIII deficiency, the inci­ dence of ICH in neonates is 25% or higher, while in those with severe hemophilia the incidence of ICH is approximately 3–4%, which is still 30–80 times higher than expected in the healthy neonatal popula­ tion [12–14]. Children with an ICH generally have vague signs consisting of some or all of the follow­ ing: pallor, jaundice, poor latching and/or feeding, increased lethargy and inconsola­ bility. In addition to the most common signs listed, neurological deficits may also be pre­ sent. Neonates may have apneic and/or bradycardic episodes, paresis, bulging fon­ tanelles, and seizures. If there is a high degree of clinical suspicion of an ICH, prompt treatment should be initiated and any invasive procedure, such as a lumbar puncture (LP), should be avoided until hemostasis is corrected.

211

212

Inherited Bleeding Disorders in Women

For treatment of ICH in bleeding disor­ ders, refer to section 13.6. 13.4.2.2  Umbilical Stump Bleeding

Umbilical stump bleeding is rare in neonates with hemophilia, but more common in neo­ nates with severe deficiencies of FX, FXIII, or fibrinogen (i.e. afibrinogenemia). In these disorders, almost 100% of severely affected neonates manifest abnormal umbilical stump bleeding and consequently if this is seen, these disorders should be strongly suspected [16]. Occasionally in these disorders, umbili­ cal stump bleeding may result in significant blood loss, leading to profound anemia and the need for a red blood cell transfusion. 13.4.2.3  Venepuncture, Vaccines, and Vitamin K Injections

Figure 13.2  Left‐sided ICH (arrows) causing occlusion of left ventricle and midline shift to the right in a male with severe hemophilia A.

If not promptly treated, ICH (Figure 13.2) in a neonate with a severe bleeding disorder may lead to the child’s death. Even when such a child survives, the occurrence of an ICH is associated with a high risk of severe long‐term neurological damage, with nearly half of survivors experiencing serious long‐ term neurological sequelae including sei­ zures, psychomotor retardation, and cerebral palsy [15]. Therefore, it is important that ICH is suspected in every child born after a traumatic delivery.

Serious bleeding may also be encountered follow­ing venepuncture (Figure 13.3) (includ­ ing heel sticks) as well as following intramus­ cular injections for vaccines or for vitamin K administration. Surprisingly, most neonates, even those with severe bleeding disorders, seldom bleed excessively following intra­ muscular injections. Yet attention should be given to those who do and such children should be investigated for disorders of coagulation. Routine immunization schedules should be followed for all newborns regardless of  congenital bleeding disorder. However, where possible, it is recommended that in patients with proven severe bleeding disor­ ders, that all immunizations be given subcu­ taneously (not intramuscularly due to the risk

Figure 13.3  Bleeding in the antecubital space of both arms after venepuncture in a male with severe hemophilia A.

The Newborn

of muscular hematomas) and with the small­ est gauge needle (preferably 27 gauge). After all immunizations, pressure should be applied for 5–10 minutes post administration. In addition to routine immunizations, all chil­ dren with bleeding disorders should receive hepatitis A (can be given from one year of age onwards) and B vaccinations to minimize the risk of acquiring hepatitis from exposure to non‐recombinant blood products. 13.4.2.4  Circumcision in Newborns

Abnormal bleeding can occur in any male newborn who undergoes circumcision; how­ ever, severe bleeding with circumcision is rare unless an underlying bleeding disorder is present. Most males who bleed excessively with circumcision should be suspected of having hemophilia and tested appropriately. It should be noted that the lack of bleeding after circumcision does not exclude hemo­ philia. Of course, other bleeding disorders such as type 3 von Willebrand disease (VWD) and severe deficiencies of FVII, X, XIII, or fibrinogen can present with severe bleeding following circumcision. Given the autosomal recessive nature of these disor­ ders, they are substantially less common than hemophilia other than in the setting of consanguinity. In children suspected of having a bleeding disorder, the diagnosis should be confirmed prior to circumcision. Once a particular bleeding disorder is confirmed on laboratory evaluation and the family still wishes to undertake a circumcision, the appropriate hemostatic correction should be undertaken before the procedure. For neonates with hemophilia, generally a single dose of factor is sufficient, although some clinicians may elect to give postoperative doses [17]. 13.4.2.5  Hemarthroses, Hematomas, and Oral/Gastrointestinal/Genitourinary Bleeding

Joint hemarthroses and muscular hemato­ mas are hallmark bleeds in severe h ­ emophilia, but they rarely occur before 6–8 months of age. It is only at this age that a child becomes more mobile, and as such begins to manifest musculoskeletal bleeding.

Oral bleeding does not typically occur within the first six months of a child’s life. Eruption of teeth generally begins at around six months of age although this can begin as early as four months or as late as 12 months of age. By 14 months of age, the front eight incisors (four upper incisors and four lower incisors) will usually have erupted. Most chil­ dren, even those with bleeding disorders, do not experience bleeding with eruption of teeth, but if this does occur, it characteristi­ cally presents as oozing over a period of time. In addition to the eruption of teeth, at six months of age, parents are advised to start their infants on solid foods, and frenulum tears leading to bleeding can occur secondary to trauma from utensils used to feed the child. Also, tears to the frenulum can occur with oral trauma and might ensue if a child falls forward and sustains trauma to the lower jaw. Hemoptysis, hematemesis, gastrointestinal tract bleeding or hematuria do not typically occur in the neonatal period.

13.5 ­Congenital Bleeding Disorders and Their Presentation in Newborns 13.5.1  Inherited Platelet Disorders 13.5.1.1 Thrombocytopenias

Thrombocytopenia in neonates, as in older children and adults, has classically been defined as an absolute platelet count below 150 × 109/L. According to the platelet count, thrombocyto­ penia can be classified as mild (platelet count between 100 and 150 × 109/L), moderate (­platelet count between 50 and 100 × 109/L), and severe (platelet count < 50  ×  109/L). Thrombocytopenia is relatively infrequent in the general neonatal population, occurring in 0.5–2% of unselected healthy newborns as detected by cord blood sampling [18]. In contrast, thrombocytopenia (including severe thrombocytopenia) is very common in the neonatal intensive care unit, affecting approximately one‐third of all neonates during the course of their hospitalization. It  is especially common in premature and extremely low birth‐weight infants [19].

213

214

Inherited Bleeding Disorders in Women

Thrombocytopenias in unwell children are often part of chronic fetal hypoxic states (e.g. maternal hypertensive disorders, maternal diabetes, and fetal growth restriction) or are a result of sepsis or disseminated intravascular coagulation (DIC), defined as a coagulopathic disorder which consumes procoagulant, anti­ coagulant, and fibrinolytic proteins as well as platelets, leading to a high risk of bleeding. DIC is often seen in the setting of sepsis. Thrombocytopenia is frequently reported in many syndromes, including Turner, trisomy 13, trisomy 18, and triploidy states [18]. Thrombocytopenias in healthy, other­ wise well full‐term neonates may be indica­ tive of either an immune (allo‐ or auto‐) thrombocytopenia or an inherited throm­ bocytopenia. Clues to immune thrombocy­ topenias (ITPs) are the lack of dysmorphic or non‐hematological abnormalities, gen­ erally normal‐sized platelets with normal morphology, a maternal history of ITP and lastly, lack of response to platelet transfusions. Inherited thrombocytopenias are often cate­ gorized according to platelet size and mor­ phology. Macrothrombocytopenias include myosin heavy chain (MYH)9‐related disor­ ders, Bernard–Soulier syndrome (BSS), inher­ ited thrombocytopenias related to GATA‐1 mutations, type 2B VWD, and platelet‐type VWD. For a more thorough list of thrombocy­ topenias distinguished on the basis of platelet size, see Figure 13.1b. Normal‐sized platelets are seen in con­ genital amegakaryocytic thrombocytopenia (CAMT) and thrombocytopenia with absent radii (TAR), while small platelets are seen in  Wiskott–Aldrich syndrome (WAS) and X-linked microthrombocytopenia. In many of these disorders, there are associated non‐ hematological abnormalities which are often a clue to the diagnosis. In general, children with inherited throm­ bocytopenias show more bleeding than expected for the platelet count, show abnor­ malities in size or morphology (e.g. neutrophil Döhle‐like inclusion bodies in MYH9‐related disorders) of platelets and, unlike in ITP, will usually respond to platelet transfusions.

13.5.1.2  Disorders of Platelet Function

Although mild platelet function disorders (e.g. platelet storage pool deficiency and secretion defects) are common, severe platelet function disorders are not. Of the latter, the most severe are GT and BSS. Typically, such patients present with bruis­ ing and extended bleeding after minor trauma and venepunctures in the setting of a normal platelet count (GT) (Figure  13.4) or mild/moderate macrothrombocytopenia (BSS) and a normal PT/INR and aPTT. Platelet function disorders, due to their rar­ ity and the fact that most do not typically present with severe bleeding, are not typi­ cally diagnosed at birth. Platelet function disorders can overlap with thrombocytopenias as some platelet function disorders also manifest with a low platelet count;  this includes BSS, MYH9‐ related disorder, and gray platelet syndrome. For thrombocytopenic patients, neonatal testing is hampered by difficulties in obtain­ ing a sufficient number of platelets for aggregation studies. It should be noted that PFA‐100/200 testing and flow cytometry, both of which require much less blood than is required for platelet aggregation testing, can be used to make the diagnosis of GT and BSS. Patients with GT will have decreased glycoprotein (GP) IIb‐IIIa on their surface membranes while patients with BSS will have decreased GPIb‐IX‐V. PFA‐100/200

Figure 13.4  Newborn with GT after venepuncture and holding of the child’s left arm.

The Newborn

testing in both GT and BSS shows absence of closure (e.g. closure times > 300 s) with both collagen/epinephrine (Col/Epi) and collagen/adenosine diphosphate (Col/ADP) cartridges. However, PFA‐100/200 testing may be compromised if the child has suf­ fered significant bleeding and consequently is anemic, as anemia causes prolongation of closure times [20]. If platelet aggregation testing is done patients with GT will demon­ strate absent aggregation to all agonists apart from ristocetin, while patients with BSS will demonstrate the converse. It should be noted that a neonate born to a parent with a platelet function disorder is not usually at risk of bleeding due to the autosomal recessive nature of most platelet function disorders. Mild platelet function disorders are rarely diagnosed in the neona­ tal period given their milder phenotype. 13.5.2  Disorders of Coagulation 13.5.2.1  Hemophilia A and B

Hemophilia is an X‐linked deficiency in clot­ ting factor VIII or IX. Severity is dependent upon the plasma level of factor: severe < 1%; moderate 1–5%; and mild > 5%. Hemophilia affects one in 5000 males, and of these diag­ noses, 80% are hemophilia A [21]. Hemophilia affects all ethnicities equally but the reported prevalence of diagnosed hemophilia in coun­ tries is variable, reflecting physician awareness, laboratory expertise, and socioeconomic con­ text. In less affluent countries, it is mostly patients with severe hemophilia who are diag­ nosed. Because of a lack of treatment in these countries, a substantial percentage of patients will die before they reach reproductive age. In contrast, in countries where hemophilia treat­ ment is widely available, patients now have an almost normal life expectancy and increas­ ingly are demonstrating normal fecundity. As well, in such countries, a greater proportion of milder forms of hemophilia are diagnosed. Any newborn male with unexpected ­bleeding from venepuncture sites, surgical interventions, such as circumcision, or easy bruising or bleeding should be suspected of

having hemophilia and FVIII and FIX levels should be measured. The aPTT is typically much more prolonged in affected males than is typically seen in normal healthy newborns, although borderline normal results are also possible in milder deficiencies and thus spe­ cific factor level measurements are required. Accurate and quick diagnosis of affected neonates allows for early management and counseling of the family. 13.5.2.2  Von Willebrand Disease

Von Willebrand disease is a congenital bleeding disorder that affects both primary, platelet‐related hemostasis and secondary hemostasis. It is caused by mutations in the  gene for VWF, resulting in a quantita­ tive  (mildly reduced  –  type 1 or severely reduced  –  type 3) or qualitative (type 2) abnormality of VWF. As VWF is the carrier protein for FVIII, levels of FVIII are reduced in the setting of more severe forms of VWD, thereby leading to a defect in secondary hemostasis. The clinical picture of VWD in the neona­ tal period is, for the most part, less pro­ nounced than that of severe hemophilia. The exceptions are those patients with type  3 VWD who will have both very low VWF levels and very low FVIII levels. These children have an increased risk for ICH and other bleeding complications. The low FVIII level in type 3 VWD highlights the fact that a low FVIII level is not pathogno­ monic of hemophilia A and that other con­ genital bleeding disorders can present with a low FVIII level (e.g. type 3 VWD, severe type 1 VWD, type 2N VWD, combined defi­ ciency of FV and FVIII). Hence, in patients with a low FVIII level, it is important that the clinician consider these other disorders and test for them. This would include per­ forming tests of VWF (antigen and activity testing – to rule out type 3 VWD and severe type 1 VWD), doing a PT/INR as a screen for combined FV and FVIII deficiency and either genetic testing or FVIII‐VWF bind­ ing studies to rule out type 2N VWD. This is particularly important when hemophilia A

215

216

Inherited Bleeding Disorders in Women

is suspected but no mutation is found in the FVIII gene. Prenatal diagnosis is rarely performed for fetuses suspected of having VWD except for those at risk of type 3 VWD. In newborns at risk for VWD, levels of VWF:antigen, VWF:activity, and FVIII can be measured in cord blood. Normal values may be sufficient to exclude severe forms of VWD in at‐risk cases but milder forms of VWD may be missed as VWF levels are physiologically ele­ vated for the first few months of life, thereby potentially masking mild forms of VWD. Type 2B VWD should also be suspected in newborns with a family history of VWD and a low platelet count, which may decrease fur­ ther with stress. A lack of thrombocytopenia does not completely exclude type 2B VWD, however, as some patients with type 2B VWD do not show thrombocytopenia. 13.5.2.3  Vitamin K Deficiency Bleeding (VKDB) in Infancy

Factors II, VII, IX, and X are vitamin K‐dependent coagulation proteins and are typically decreased in the newborn due to liver immaturity. Upon laboratory evalua­ tion, the PT/INR is most prolonged; the aPTT is usually prolonged but to a lesser extent. All neonates at risk should have fac­ tor levels measured on cord blood and for those with borderline results, levels should be retested after a course of vitamin K ther­ apy to rule out congenital genetic deficien­ cies of these coagulation factors. Vitamin K deficiency will show reduced levels of all vitamin K‐dependent factors, although due to its short half‐life FVII is typically the most reduced. Vitamin K deficiency should be differentiated from liver failure and from a congenital genetic deficiency of a single vitamin K‐dependent factor (particularly FVII as this is the most common) [16]. Severe deficiencies of vitamin K‐dependent factors because of deficiency in vitamin K in  newborns may lead to a hemorrhagic diathesis in the newborn called vitamin K deficiency bleeding (VKDB). Typical pres­ entation of VKBD includes bruising and

internal bleeding such as an ICH. Failure to administer vitamin K at birth, prolonged jaundice, and failure to thrive are risk factors for VKDB [22]. Vitamin K deficiency bleeding is classi­ fied according to when it presents – early, classic, and late. Early VKDB is defined as bleeding within the first 24 hours of life and is usually associated with maternal medications (e.g. antiepileptics, warfarin, cephalosporins, rifampin, and isoniazid) that interfere with the transfer of vitamin K from  the mother to the child through the placenta. Classic VKDB typically presents between days 2 and 7 of life and is seen in exclusively breastfed newborns not given vitamin K at birth. Late VKDB occurs in the setting of intestinal malabsorption; this may be secondary to systemic antibiotic therapies [22]. Vitamin K replacement in the newborn should be instituted to prevent bleeding in VKDB. With the standard practice of vita­ min K prophylaxis immediately after birth, bleeding as a result of classic VKDB is rare. Those newborns who present with suspected VKDB should be treated immediately with intravenous vitamin K. If the newborn is pre­ senting with a life‐threatening bleed, addi­ tional immediate infusion of fresh frozen plasma (FFP) or inactivated prothrombin complex concentrate (PCC) should be given. 13.5.2.4  Rare Coagulation Factor Deficiencies

Rare congenital bleeding disorders include fibrinogen (FI), FII, FV, FVII, FX, FXI, and FXIII deficiencies. The bleeding pattern in these deficiencies is variable and, in some of these disorders, there is a poor correla­ tion between the severity of the disease phenotype and the factor level. In all of these dis­orders, severe deficiencies are only seen in a  child who has inherited two abnormal alleles  –  one from each parent. Hence, these autosomal recessive disorders are in general rare but more common if there is parental consanguinity. The genet­ ics and clinical presentation of these rare

The Newborn

coagulation factor deficiencies (in new­ borns) are described in Table  13.1. Note that FXII deficiency is not discussed as it does not cause bleeding Factor I (fibrinogen) deficiency may be divided into (i) quantitative or type I fibrino­ gen abnormalities and (ii) qualitative or type II fibrinogen abnormalities (these are referred to as dysfibrinogenemias). Quantitative defects may be further subdivided into afibrinogenemia (complete absence of fib­ ­ rinogen) and hypofibrinogenemia (partial deficiency of fibrinogen). Inherited disorders of fibrinogen may be more common than pre­ viously appreciated; fortunately, most are mild and unlikely to present with significant bleeding in neonates. Afibrinogenemia is an autosomal recessive disorder in that dele­ terious mutations exist on both fibrinogen genes.  In contrast, hypofibrinogenemia and dysfibrinogenemia are autosomal dominant in inheritance and caused by the patient possessing one mutated fibrinogen gene. ­ Typically, the PT/INR and aPTT are pro­ longed, while fibrinogen activity levels are reduced in all of these disorders. Most fibrin­ ogen testing done involves measuring fibrino­

gen activity and not fibrinogen antigen levels. Consequently, most fibrinogen assays do not distinguish between hypofibrinogenemia and dysfibrinogenemia as in both cases the fibrin­ ogen activity is reduced. To differentiate the two requires that a quantitative fibrinogen antigen assay be performed – this is generally only available in a few specialized laborato­ ries. Of note, dysfibrinogenemia can  present with either a hemorrhagic or a thrombotic phenotype. Data from epidemiological studies indicate that about 55% of patients with dysfi­ brinogenemia are asymptomatic, 25% may present with a hemorrhagic phenotype, and 20% with a thrombotic phenotype [23, 24]. Factor II deficiency is the rarest congenital bleeding disorder. Newborn bleeding is unu­ sual. The PT/INR and aPTT may both be prolonged. Factor II levels may be measured in at‐risk neonates but levels must be inter­ preted in light of age‐corrected reference values. Factor V deficiency is extremely rare. The  PT/INR and aPTT may be prolonged. Factor V levels should be measured in at‐risk babies since ICH is reported in severe cases. Levels increase during the first month of life

Table 13.1  Genetics of rare coagulation factor deficiencies, listed in descending order of prevalence. Clinical bleeding in newborns Deficiency chromosome

Incidence

CNS

Umbilical

Other

FXI

4

1/100 000 (overall); 1/1000 (Ashkenazi Jews)

−

−

Postsurgical bleeding

FVII

13

1/500 000

++

+

Mucosal bleeding

FI (Fibrinogen)

4

1/1 000 000

+

+++

Mucosal bleeding Poor wound healing

FV

1

1/1 000 000

+

−

Mucosal bleeding

FV + FVIII

18

1/1 000 000

−

−

Mucosal bleeding

FXIII

6 (A subunit) 1 (B subunit)