The studies of the Belgian Obstetric Surveillance System described in this book were possible thanks to the voluntary participation of the following Belgian maternity units: GZA Sint-Jozef, Mortsel; GZA Sint-Augustinus, Wilrijk; GZA Sint-Vincentius, Antwerpen; ZNA, SintErasmus, Borgerhout; ZNA, Jan Palfijn, Merksem; ZNA, Middelheim, Antwerpen; ASZ, campus Aalst; ASZ, campus ; AZ Lokeren; Imelda Ziekenhuis, Bonheiden; Sint-Jozef ziekenhuis, Bornem; AZ Klina, Brasschaat; AZ Sint-Jan, Brugge; AZ Sint-Lucas, Assebroek; Centre de Santé des Fagnes; Centre Hospitalier de l’Ardenne; Centre Hospitalier EpiCura, site Hornu; Centre Hospitalier de Mouscron; Centre Hospitalier du Bois de l’Abbaye et de Hesbay; Centre Hospitalier Peltzer, La Tourelle; CHC Saint-Vincent; CHC Saint-Joseph; CHC Sainte-Elisabeth; CHIREC, Clinique Sainte-Anne, Saint-Remi; CHIREC Braine-l’Alleud-Waterloo; CHR de la Citadelle; CHR de Namur; CHR du Val de Sambre; CHR Haute Senne-Le Tilleriau; CHR Mons Clinique Saint-Joseph; CHU Ambroise Paré; CHU Brugmann; CHU Charleroi, André Vésale; CHU Notre-Dame des Bruyères; CHU Saint-Pierre; CHU Tivoli; CHwapi-Site Notre-Dame; Clinique Edith Cavell; Clinique et Maternité Sainte Elisabeth; Clinique Notre dame de Grâce; Clinique Reine Astrid; Clinique SaintePiere; Cliniques de l’Europe, Saint-Michel; Cliniques de l’Europe, Sainte-Elisabeth; Cliniques du Sud ; Cliniques Universitaires Saint-Luc; AZ SintVincentius, Deinze; AZ Sint-Blasius, Dendermonde; AZ Monica, Deurne; AZ Diest; AZ Sint- Maarten, Duffel; AZ Alma, Eeklo; Dimpna Ziekenhuis, Geel; AZ Jan Palfijn, Gent; AZ Maria Middelares, Gent; AZ Sint-Lucas, Gent; GHDC - Notre Dame, Charleroi; AZ Maria ziekenhuis, Halle; Jessaziekenhuis, ; AZ Sint Elisabeth, Herentals; CAZ Midden-, Heusden- Zolder; Hôpital Civil Marie Curie; Hôpital Erasme, Bruxelles; Hôpitaux Iris Sud - Site Ixelles; Jan Yperman ziekenhuis, Ieper; IFAC Hôpital Princesse Paola; INDC Entité Jolimontoise; Sint-Jozef ziekenhuis, Izegem; Kliniek Sint-Jan; Klinik Saint-Josef; AZ Zeno, campus Knokke-Heist; AZ Groeninge, Kortrijk; Heilig Hart ziekenhuis, Leuven; UZ Leuven; Heilig Hart ziekenhuis, Lier; AZ Sint-Jozef, Malle; AZ Sint-Maarten, Mechelen; AZ Delta, campus Menen; Heilig Hart ziekenhuis, Mol; Onze-Lieve-Vrouw ziekenhuis, campus Aalst; Onze-LieveVrouw ziekenhuis, campus Asse; AZ Damiaan, campus Sint-Jozef, Oostende; AZ Sint-Jan, campus Oostende; AZ Oudenaarde; Mariaziekenhuis Noord-Limburg, Overpelt; AZ Heilige Familie, Reet AZ Delta, campus Brugsesteenweg, Roeselare; AZ Delta, campus Wilgenstraat, Roeselare; AZ Glorieux, Ronse; AZ Nikolaas, Sint-Niklaas; Sint-Trudo ziekenhuis, Sint-Truiden; Saint-Nikolaus Hospital; SintAndries ziekenhuis, Tielt; Heilig Hartziekenhuis, Tienen; AZ Vesalius, Tongeren; Sint-Rembertziekenhuis, Torhout; AZ Turnhout; UZ Antwerpen; UZ Brussel; AZ SintAugustinus, Veurne; AZ Jan Portaels, Vilvoorde; Onze-Lieve-Vrouw van Lourdes ziekenhuis, Waregem; Ziekenhuis Oost-Limburg, campus Sint-Jan, Genk; AZ SintElisabeth, .

© Griet Vandenberghe, 2018. All rights reserved. No part of this book may be reproduced or transmitted, in any form or by any means, without written permission from the author.

The Belgian Obstetric Surveillance Study was funded by the College for Mother and Newborn, a consultative body of the Belgian Public Health Service.

Griet Vandenberghe received financial support by the Flemish Research Foundation (FWO): a Clinical PhD fellowship in 2015 - 2016, grant number FWO14/KDB/003.

Cover and layout by Fran Vandenberghe. Printed by Nevelland Grafics.

GRIET VANDENBERGHE PROMOTORS AND DOCTORAL GUIDANCE COMMITTEE

MEMBERS OF THE EXAMINATION COMMITTEE

TABLE OF CONTENTS

TABLE OF CONTENTS ...... 7

LIST OF ABBREVIATIONS ...... 11

INTRODUCTION ...... 17

REGISTERS AND SURVEYS ...... 17 ASSESSING QUALITY IN OBSTETRIC HEALTH CARE ...... 17 HEALTH CARE SYSTEM AND OBSTETRIC HEALTH CARE IN ...... 22 OBSTETRIC REGISTERS AND SURVEYS IN BELGIUM...... 25 REFERENCES ...... 27

OBJECTIVES ...... 33

IN SUMMARY: ...... 33 IN TARGETS: ...... 33

METHODOLOGY...... 37

INSTITUTE...... 37 DATA COLLECTION...... 37 WEBSITE WWW.B-OSS.BE ...... 37 OBSTETRIC COMPLICATIONS UNDER SURVEILLANCE ...... 38 REGISTERED VARIABLES ...... 39 DATA ANALYSIS ...... 39 ETHICS APPROVAL ...... 40 REFERENCES ...... 41

RESULTS ...... 45

NATIONWIDE POPULATION-BASED COHORT STUDY OF UTERINE RUPTURE IN BELGIUM: RESULTS FROM THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM ... 49

ABSTRACT ...... 50 BACKGROUND ...... 51 METHODS ...... 51 RESULTS ...... 54 DISCUSSION...... 61 CONCLUSION ...... 63 ACKNOWLEDGMENTS ...... 63 REFERENCES ...... 64 FOOTNOTES...... 67 NATIONWIDE POPULATION-BASED COHORT STUDY OF PERIPARTUM HYSTERECTOMY AND ARTERIAL EMBOLISATION IN BELGIUM: RESULTS FROM THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM ...... 71

ABSTRACT ...... 72 BACKGROUND ...... 73 METHOD ...... 74 RESULTS ...... 76 DISCUSSION ...... 90 CONCLUSION ...... 93 ACKNOWLEDGMENTS ...... 93 REFERENCES ...... 95 FOOTNOTES ...... 97

THE INOSS STUDY OF UTERINE RUPTURE: A DESCRIPTIVE MULTI-COUNTRY POPULATION BASED STUDY...... 101

ABSTRACT ...... 102 INTRODUCTION ...... 103 METHODS ...... 104 RESULTS ...... 107 DISCUSSION ...... 113 CONCLUSION ...... 116 ACKNOWLEDGEMENTS ...... 117 DISCLOSURE OF INTEREST ...... 118 CONTRIBUTION TO AUTHORSHIP ...... 118 DETAILS OF ETHICAL APPROVAL ...... 118 FUNDING ...... 118 REFERENCES ...... 119

THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM TO MONITOR SEVERE MATERNAL MORBIDITY...... 125 ABSTRACT ...... 126 BACKGROUND: WHY DID WE DEVELOP THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM? ... 127 METHODOLOGY: HOW DOES THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM WORK? ...... 128 RESULTS: WHAT IS THE OUTPUT OF THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM? ...... 130 DISCUSSION: HOW CAN WE FURTHER IMPROVE THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM? ...... 134 CONCLUSION ...... 136 ACKNOWLEDGEMENTS ...... 136 FUNDING ...... 136 DISCLOSURE OF INTEREST ...... 136

REFERENCES ...... 137 LEGEND OF TABLES ...... 138 LEGEND OF FIGURES ...... 138

DISCUSSION ...... 143

MAIN FINDINGS: INFOGRAPHIC...... 143 STRENGTHS AND WEAKNESSES...... 149 RECOMMENDATIONS TO IMPROVE THE PERFORMANCE OF B.OSS...... 150 RECOMMENDATIONS FOR FURTHER DEVELOPMENT AND FUTURE STUDIES OF B.OSS...... 152 RECOMMENDATIONS TO IMPROVE OBSTETRIC CARE IN BELGIUM...... 156 REFERENCES ...... 163

SUMMARY ...... 169

SAMENVATTING ...... 173

CURRICULUM VITAE ...... 177

ADDENDA ...... 191

NATIONWIDE POPULATION-BASED COHORT STUDY OF UTERINE RUPTURE IN BELGIUM: RESULTS FROM THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM. SUPPLEMENTARY FILES...... 191 NATIONWIDE POPULATION-BASED COHORT STUDY OF PERIPARTUM HYSTERECTOMY AND ARTERIAL EMBOLISATION IN BELGIUM: RESULTS FROM THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM. SUPPLEMENTARY FILES...... 195 THE INOSS STUDY OF UTERINE RUPTURE: A DESCRIPTIVE MULTI-COUNTRY POPULATION BASED STUDY. SUPPLEMENTARY FILES...... 196

THANK YOU ...... 217

LIST OF ABBREVIATIONS

ACOG American College of Obstetricans and Gynecologists AFSfC Association Francophone des Sages-femmes Catholiques AIM Agence InterMutualiste (in dutch: IMA) AIP Abnormal Invasive Placenta AMS Auditcommissie Maternale Sterfte APE Antenatal Pulmonary Embolism ART Artificial Reproductive Technologies AuOSS Austrian Obstetric Surveillance System BMA Belgian Midwives Association BMI Body Mass Index (weight in kg/ (length in m)2) B.OSS Belgian Obstetric Surveillance System BPSU British Paediatric Surveillance Unit CD Caesarean Delivery CDC Center for Disease Control and prevention CEMD Confidential Enquiries into Maternal Deaths CEMM Confidential Enquiries into Maternal Morbidity CEpiP Centre d’épidémiologie périnatale CI Confidence Interval CNGOF Collège National des Gynécologues et Obstétriciens Français COD Cause of Death COS Core Outcome Set CS Caesarean section CTG Cardio-Toco Graphy D&C Dilatation and Curettage DDI Diagnosis-to-Delivery Interval DIC Disseminated Intravascular Coagulation EC Ethics Committee EHCI Euro Health Consumer Index EmCS Emergency Caesarean Section ENCMM Enquête Nationale Confidentielle Sur les Morts Maternelles EPIMOMS EPIdémiologie de la MOrbidité Maternalle Sévère ERCS Elective Repeat Caesarean Section FFS Fee for service FOD Federale Overheids Dienst FWO Fonds voor Wetenschappelijk Onderzoek Vlaanderen. GerOSS German Obstetric Surveillance System GGOLFB Groupement des Gynécologues Obstétriciens de Langue Française de Belgique GLOSS Global Maternal Sepsis Study HIE Hypoxic Ischemic Encephalopathy HSR Health Services Research ICD-9-CM International Classification of Diseases, Ninth Revision, Clinical Modification ICoD INOSS Core Dataset ICSI Intra-Cytoplasmic Sperm Injection ICU Intensive Care Unit IMA InterMutualistisch Agentschap (In french: AIM) INOSS International Network of Obstetric Survey Systems IR Interventional radiology IUFD Intra-uterine Fetal Death IVF In-Vitro Fertilisation KCE Belgian Health Care Knowledge Centre LEMMoN Landelijke studie naar Etnische Determinanten van Maternale Morbiditeit in Nederland. MBRRACE-UK Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries accross the UK MDSR Maternal Death Surveillance and Response system MIC Maternal Intensive Care MNM Maternal Near Miss MMR Maternal Mortality Ratio MRI Magnetic Resonance Imaging MZG Minimale Ziekenhuis Gegevens. In French: RHM NethOSS Netherlands Obstetric Surveillance System NIC Neonatal Intensive Care NOSS Nordic Obstetric Surveillance System NVOG Nederlandse Vereniging voor Obstetrie & Gynaecologie OB-GYN Obstetrician-Gynaecologist OECD Organisation for Economic Cooperation and Development OR Odds Ratio PC Packed Cells PG Prostaglandin PMR Perinatal Mortality Rate PPH Postpartum haemorrhage PPSB Prothrombin-Proconvertin/Stuart Factor-Antihemophilic Factor B PPV Positive Predictive Value P4Q Pay for Quality RBC Red Blood Cell RBSOG Royal Belgian Society for Obstetrics & Gynaecology RCOG Royal College of Obstetricians and Gynaecologists RHM Résumé Hospitalier Minimum (In dutch: MZG) RR Relative Risk SAMM Severe Acute Maternal Morbidity SHIP Spontaneous Hemoperitoneum In Pregnancy SMFM Society for Maternal Fetal Medicine SMM Severe Maternal Morbidity SPE Studiecentrum voor Perinatale Epidemiologie TOLAC Trial of Labour after Caesarean Section UKOSS United Kingdom Obstetric Surveillance System UPSfB Union Professionnelle des Sages-femmes Belges UR Uterine Rupture VBAC Vaginal Birth after Caesarean Section VBOV Vlaamse Beroepsorganisatie van Vroedvrouwen VVOG Vlaamse Vereniging voor Obstetrie en Gynecologie WHO World Health Organisation WIV-ISP Wetenschappelijk Instituut Volksgezondheid. Institut Scientifique de Santé Publique. Scientific Institute of Public Health

INTRODUCTION

Registers and surveys

The key to improve is to measure. Measurements can reveal weaknesses and weaknesses are opportunities to improve. Measurements allow to discover trends and to benchmark results within a group. Measurements support policy-makers in their decisions to change. Subsequent measurements then monitor the results when interventions were implemented. This wisdom is generally true and widely used: in pursuit of a long tradition in the business world surveys and registers have become common practice in health care.

Assessing quality in obstetric health care

Obstetric health care will always get special attention, because it concerns our most delicate group: young, mostly healthy women who become mothers and vulnerable fetuses and newborns. Maternal and perinatal mortality and morbidity are core indicators in the assessment of the quality of obstetric and perinatal care and health care in general. While the definition of maternal death seems most obvious (see Box 1), the registration of maternal deaths has proven not to be that straightforward. Routine death statistic systems, even in high-income countries and even when linked to birth and other registers, are known to under-estimate the true maternal mortality ratio (MMR).1, 2 The most accurate way to monitor maternal mortality is an enhanced system that actively identifies maternal deaths and assesses the quality of care in each case: the so-called confidential enquiries into maternal deaths (CEMD). The oldest and most praised system is the United Kingdom’s CEMD, established in 1952 and currently run by MBRRACE-UK: Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK.3 Its methodology is considered as the global standard and served as a template for many other high-income, even middle- and low-income countries. Confidentiality for the women, health care providers and institutions is a key feature, beside the direct notification of the maternal deaths to the CEMD office and the assessment of the quality of care by a team of experts based on the full case notes. Despite the high-quality health care in the UK, the CEMD identifies quality of care concerns in more than 50% of cases.4 Lessons to be learned to improve future care are reported triennially and had significant impact on maternal deaths throughout the years.5 Similar CEMD systems are organised in other countries surrounding Belgium: the Netherlands (Auditcommissie Maternale Sterfte, AMS)6 and France (Enquête Nationale Confidentielle Sur les Morts Maternelles (ENCMM)7).

Box 1. Definitions of maternal death according to the rules of the International Statistical Classification of Diseases (ICD-10).8

Maternal death A maternal death is the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of duration and the site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management, but not from accidental of incidental causes.

Maternal deaths are subdivided into two groups: Direct obstetric deaths: maternal deaths resulting from obstetric complications of the pregnancy state (pregnancy, labour and the puerperium), from interventions, omissions, incorrect treatment, or from a chain of events resulting from any of the above.

Indirect obstetric deaths: maternal deaths resulting from previous existing disease or disease that developed during pregnancy and which was not due to direct obstetric causes, but which was aggravated by physiologic effects of pregnancy.

Late maternal death A late maternal death is the death of a woman from direct or indirect causes more than 42 days but less than one year after termination of pregnancy.

In high-income countries maternal deaths have become extremely rare events occurring in exceptional situations. For every woman who dies from a severe obstetric complication, a larger number of women will survive but suffer severe, possibly permanent morbidity. Since two decades the focus has shifted from maternal mortality, as major indicator in measuring the quality of obstetric health care, to severe maternal morbidity (SMM) and maternal near-miss (MNM)9, 10: the last-but- one level in a continuum from uncomplicated pregnancy to maternal death (Figure 1). The study of women who experienced and survived a severe obstetric complication has advantages compared to the study of maternal deaths: an adequate number of cases for analysis is achieved more rapidly because they are more common, the women can be interrogated about the provided care, the clinicians may provide information more easily because they feel less threatened, the comparison with cases of maternal death can reveal risk factors associated with mortality and the resulting recommendations will be more generalisable. 9, 11 Implementing the entity severe maternal morbidity in surveys encounters the difficulty of formulating a uniform definition and a set of usable criteria to identify cases.12-14 Disease-specific criteria (e.g. eclampsia, haemorrhage), intervention-based criteria (e.g. ICU admission, hysterectomy) and organ-system-dysfunction-based criteria (e.g. oliguria, shock) are three main approaches (Figure 2) and all of them have disadvantages. 9, 12 Different groups, including the World Health Organisation (WHO)9, the Centers for Disease Control and Prevention15 and Euro-Peristat 16, proposed a set of identification criteria, but a widely accepted consensus has not been reached so far. However, to quote Professor Marian Knight, “actions are more important than definitions”. “Focusing on how to define severe maternal or near-miss morbidity is largely irrelevant. The most important aspect of any work is to develop an enthusiastic network of clinicians and hospitals who agree on what is for them the most important aspect of morbidity in their setting”.17

Figure 1. The continuum from uncomplicated pregnancy to maternal mortality. Adapted from The Pyramid of Disease. PowerPoint Presentation. Introduction to UKOSS. Retrieved from https://www.npeu.ox.ac.uk/ukoss And The spectrum of morbidity: from noncomplicated pregnancies to maternal death. Say et al., (2009).8

Figure 2. World Health Organization Maternal Near-Miss Tool.18 Reprinted from: Evaluating the quality of care for severe pregnancy complications: the WHO near-miss approach for maternal health. WHO (2011), page 21. Available from: http://www.who.int/reproductivehealth/topics/maternal_perinatal/nmconcept/en/. Copyright (2017). Rather than aiming to survey all cases of severe maternal morbidity as a proper entity, severe obstetric complications can be studied in a topic-based approach. This is done by the Obstetric Surveillance Systems: prospective population-based surveys that identify and study rare diseases, severe complications and near-miss events in pregnancy. The UK was again a pioneer with the well-established United Kingdom Obstetric Surveillance System (UKOSS).19 UKOSS completed around three dozen studies since its initiation in 2006: mostly one-year cohort, case-control and descriptive epidemiological studies. The efforts of UKOSS have resulted in improved patient safety, e.g. UKOSS data were used to inform and change guidelines or enabled benchmarking on a local (hospital), national and international level.19, 20 The methodology of UKOSS served as a template to several high-income countries. The International Network of Obstetric Survey Systems (INOSS) was constituted in 2010 and unites the obstetric surveillance systems of Australia, Austria, Belgium, Denmark, Finland, France, Germany, Italy, the Netherlands, New Zealand, Norway, Portugal, Slovakia, Spain, Sweden and the United Kingdom. The INOSS has the mission to co-operate, share information and enable cross-national comparisons and collaborative studies of very rare conditions.21 To this end, INOSS established a set of uniform definitions for eight severe obstetric complications, using the Delphi- method.22

Health care system and obstetric health care in Belgium.

Belgium is a small country, 30.528 square kilometres and almost 11.5 million inhabitants, known for its complex federal state. Health care in Belgium is considered of high quality, ranking 4th in the Euro Health Consumer Index (EHCI) 2016, with apparently the best accessibility to healthcare services anywhere in Europe.23 The main features of the Belgian health care system are a compulsory health insurance, free choice of health care provider, institution and insurance for patients, diagnostic and therapeutic freedom for physicians and a predominantly fee-for-service (FFS) payment.24 There are concerns about the financial sustainability of the Belgian health care system, because of the high - eight highest health expenditure per capita among the EU countries in 201524, 25 - and rising expenditures. Major reforms to the organisation and financing of the Belgian health care system are pending, in order to guarantee the quality and accessibility in the long term at a sustainable cost. The obstetric health care in Belgium is mainly ob/gyn specialist-led and hospital- based, with approximately 1% of home births. There are around 1600 accredited ob/gyn specialists under the age of 65,26 with an increasing number of female specialists. This trend of feminisation resulted in more ob/gyn specialists organised in groups with a scheduled on-call system. Most ob/gyn specialists combine a private ambulatory practice with their work in a public or private non-profit hospital, where self-employed doctors can charge supplemental fees for treatment in private rooms. Belgium counts over 10,400 accredited midwives,27 with the highest ratio of midwives per 100,000 inhabitants in the EU-28 in 2014;28 however only part of them are working in obstetric health care.27, 29 The great majority of practising midwives work in hospitals, providing care during pregnancy, labour, delivery and postpartum under supervision of ob/gyn specialists. An increasing number of midwives is working autonomously outside the hospitals providing antenatal and postnatal care, with a very small number of midwives organised in group practices offering home births and/or birth centres.27, 29 Belgium has a high number of small volume maternity units (57% have less than 1000 annual births) (Table1), besides 17 tertiary referral centres providing both Neonatal Intensive Care (NIC) and Maternal Intensive Care (MIC) (Figure 3).30, 31 The MIC concept, introduced by law in 1996, is defined as ‘a division dedicated to the intensive observation of high-risk pregnancies, patients with a pregnancy at high risk for neonatal observation at a NIC service and patients who will need highly specialized postpartum care…”.32 However, the statute, the function and terms of reference of the MIC-service has not been precisely defined.32

Table 1. Distribution of Belgian maternity units according to annual number of deliveries. Situation in 2012. 30, 31 Annual number of deliveries Number of Belgian maternity units < 500 14 500 - 999 49 1000 - 1999 35 ≥2000 15

Figure 3. Geographic distribution of Maternal Intensive Care centres in Belgium. Reprinted from: Van Parys A-S, Lucet C, Remacle A, Di Zinno T, Verstraelen H, Mambourg F, et al. Intensieve maternele verzorging (Maternal Intensive Care) in België. Health Services Research (HSR). Brussel: Federaal Kenniscentrum voor de Gezondheidszorg (KCE); 2008. KCE reports 94A (D/2008/10.273/77). 32 Source: Ministry of Public Health and phone calls with hospital workers. There are 2 maternities with NIC service without MIC service.

The annual number of deliveries in Belgium is around 124,000.30, 31 Noteworthy is a relatively high rate of induction of labour (27.1% in 2015)16 and a steadily increasing caesarean section (CS) rate (20.7% in 2015)33-35, still below the European median rate (25.2% in 2010).16 Large differences are observed between regions (e.g. induction of labour was 33% in compared to 21.4% in Flanders) and between hospitals (e.g. CS rates range from 11.8% to 32.9%).

Belgium has only a few national guidelines for obstetric care, provided by the Belgian Health Care Knowledge Centre (KCE): - KCE Report 248 (2015): What are the recommended clinical assessment and screening tests during pregnancy?36 - KCE Report 228Cs (2014): Prevention of preterm birth in women at risk: selected topics.37 - KCE Reports 139C (2010): Guideline relative to low risk birth.38 There are of course renowned and international guidelines, easily accessible online for Belgian ob/gyn specialists to rely on in their daily practice; such as the guidelines provided by the Royal College of Obstetricians and Gynaecologists (RCOG), the American College of Obstetricians and Gynecologists (ACOG), de Nederlandse Vereniging voor Obstetrie & Gynaecologie (NVOG), the French College of Gynecologists and Obstetricians (CNGOF)). The lack of Belgian guidelines may be partly blamed on the complexity of the Belgian state, reflected in the existence of various professional associations for ob/gyn specialists (Vlaamse Vereniging voor Obstetrie en Gynecologie (VVOG) and Groupement des Gynécologues Obstétriciens de Langue Française de Belgique (GGOLFB), united in the Royal Belgian Society for Obstetrics and Gynaecology (RBSOG)) and for midwifes (Vlaamse Beroepsorganisatie van Vroedvrouwen (VBOV), Union Professionnelle des Sages-femmes Belges (UPSfB), Association Francophone des Sages-femmes Catholiques (AFSfC), united in the Belgian Midwives Association (BMA)).

Obstetric registers and surveys in Belgium.

Maternal and perinatal data registration in Belgium is organised in two parallel institutes for Perinatal Registry: Studiecentrum voor Perinatale Epidemiologie (SPE) in Flanders and Centre d’Epidémiologie Périnatale (CEpiP) in Brussels and Wallonia.30, 31 Both register on a mandatory basis, covering 100% of births in Belgian maternity units and home births. A selected set of maternal and perinatal data are recorded by the ob/gyn specialist, midwife and neonatologist immediately after birth. In Brussels and Wallonia (CEpiP) births are registered as such, in case of a live birth or in case of a stillbirth at a birth weight of 500 grams or more and/or after 22 completed weeks of gestation, whereas in Flanders (SPE) live births or stillbirths with a birth weight below 500 grams, irrespective of gestational age, are not registered. They make it even more complex: birth certificates for vital statistics must be completed in case of a live birth, death certificates in case of a stillbirth at a birth weight of 500 grams or more, or, when birth weight is unavailable, after 22 completed weeks of gestation or with a crown-heel length of 25 cm or more, according to the WHO guidelines. Civil registration to the National Register is only required for stillbirths more than 180 days.

The perinatal registries (CEpiP, SPE) do not include data on maternal complications, severe maternal morbidity or near-miss cases. However, SPE - but not CEpiP - registers the main cause of maternal deaths related to births according to the aforementioned SPE definition of birth. Within the national Statistics Belgium (Statbel) maternal deaths are analysed by a COD (Causes of Death) working group.39 Possible maternal deaths are identified in the Cause of Death database based on the death certificates. Cases are then checked by using three other sources: the National Register, the vital statistics Birth Register and the Cause of Death database for stillbirths and infant deaths. Deaths are then classified into the following group: direct maternal death, indirect maternal death, late maternal death or no maternal death. The maternal mortality ratio (MMR) is calculated based on data of 5 consecutive years due to extremely low and variable numbers: MMR was 5.2 per 100,000 livebirths in 2012 (2009-2014) (mean MMR in Europe is 6.3). At this point Belgium has no enhanced system to survey cases of maternal deaths.

Belgian ob/gyn specialists and pregnant women can rely on general numbers in reports of Euro-peristat, OECD (Organisation for Economic Cooperation and Development) or WHO and trust that obstetric health care in Belgium is of high quality and safety. It cannot be ignored however that quality of care concerns are identified in more than 50% of maternal deaths in our neighbouring countries or that in high-income countries maternal mortality and severe maternal morbidity are considered ‘sentinel- events’ that require further investigation.40 Therefore, we should embrace the opportunity and acknowledge that the primary aim of surveys and enquiries is to improve patient safety and quality of care. In 2011 the College of Physicians for Mother and Newborn, a consultative body of the Federal Public Service of Health, decided to set up a Belgian Obstetric Surveillance System to register and analyse severe maternal morbidity in Belgium. References

1. Bouvier-Colle MH, Mohangoo AD, Gissler M, Novak-Antolic Z, Vutuc C, Szamotulska K, et al. What about the mothers? An analysis of maternal mortality and morbidity in perinatal health surveillance systems in Europe. BJOG: an international journal of obstetrics and gynaecology. 2012 Jun;119(7):880- 9; discussion 90. 2. Saucedo M, Bouvier-Colle MH, Chantry AA, Lamarche-Vadel A, Rey G, Deneux-Tharaux C. Pitfalls of national routine death statistics for maternal mortality study. Paediatric and perinatal epidemiology. 2014 Nov;28(6):479-88. 3. Kurinczuk JJ, Draper ES, Field DJ, Bevan C, Brocklehurst P, Gray R, et al. Experiences with maternal and perinatal death reviews in the UK--the MBRRACE-UK programme. BJOG: an international journal of obstetrics and gynaecology. 2014 Sep;121 Suppl 4:41-6. 4. Cantwell R, Clutton-Brock T, Cooper G, Dawson A, Drife J, Garrod D, et al. Saving Mothers' Lives: Reviewing maternal deaths to make motherhood safer: 2006-2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG: an international journal of obstetrics and gynaecology. 2011 Mar;118 Suppl 1:1-203. 5. Lewis G. Saving Mothers' Lives: the continuing benefits for maternal health from the United Kingdom (UK) Confidential Enquires into Maternal Deaths. Seminars in perinatology. 2012 Feb;36(1):19-26. 6. Schutte JM, Steegers EA, Schuitemaker NW, Santema JG, de Boer K, Pel M, et al. Rise in maternal mortality in the Netherlands. BJOG: an international journal of obstetrics and gynaecology. 2010 Mar;117(4):399-406. 7. Saucedo M, Deneux-Tharaux C, Bouvier-Colle MH. Ten years of confidential inquiries into maternal deaths in France, 1998-2007. Obstetrics and gynecology. 2013 Oct;122(4):752-60. 8. World Health Organisation. The WHO Application of ICD-10 to deaths during pregnancy, childbirth and the puerperium: ICD-MM; 2012. Retrieved March 2018 from: http://www.who.int/reproductivehealth/publications/monitoring/9789241548458/en/ 9. Say L, Souza JP, Pattinson RC. Maternal near miss--towards a standard tool for monitoring quality of maternal health care. Best practice & research Clinical obstetrics & gynaecology. 2009 Jun;23(3):287-96. 10. Stones W, Lim W, Al-Azzawi F, Kelly M. An investigation of maternal morbidity with identification of life-threatening 'near miss' episodes. Health trends. 1991;23(1):13-5. 11. Knight M, Lewis G, Acosta CD, Kurinczuk JJ. Maternal near-miss case reviews: the UK approach. BJOG: an international journal of obstetrics and gynaecology. 2014 Sep;121 Suppl 4:112-6. 12. Say L, Pattinson RC, Gulmezoglu AM. WHO systematic review of maternal morbidity and mortality: the prevalence of severe acute maternal morbidity (near miss). Reproductive health. 2004 Aug 17;1(1):3. 13. Tuncalp O, Hindin MJ, Souza JP, Chou D, Say L. The prevalence of maternal near miss: a systematic review. BJOG : an international journal of obstetrics and gynaecology. 2012 May;119(6):653-61. 14. Kilpatrick SK, Ecker JL. Severe maternal morbidity: screening and review. American journal of obstetrics and gynecology. 2016 Sep;215(3):B17-22. 15. Centers for Disease Control and Prevention. Severe Maternal Morbidity in the United States. 2017. Retrieved 2017 November from: https://www.cdc.gov/reproductivehealth/maternalinfanthealth/ severematernalmorbidity.html#anchor_SMM 16. Euro-Peristat project with SCPE and Eurocat. European Health Report. The health and care of pregnant women and babies in Europe in 2010. 2013. Retrieved 2017 October, from: http://www.europeristat.com/images/doc/EPHR2010_w_disclaimer.pdf 17. Knight M. Severe maternal morbidity—actions are more important than definitions. BJOG: An International Journal of Obstetrics & Gynaecology. 2016;123(6):954-. 18. World Health Organisation. Evaluating the quality of care for severe pregnancy complications: the WHO near-miss approach for maternal health. 2011. Retrieved 2017 November from: http://www.who.int/reproductivehealth/topics/maternal_perinatal/nmconcept/en/ 19. Knight M, Kurinczuk JJ, Tuffnell D, Brocklehurst P. The UK Obstetric Surveillance System for rare disorders of pregnancy. BJOG: an international journal of obstetrics and gynaecology. 2005 Mar;112(3):263-5. 20. Knight M, Lindquist A. The UK Obstetric Surveillance System: impact on patient safety. Best practice & research Clinical obstetrics & gynaecology. 2013 Aug;27(4):621-30. 21. Knight M. The International Network of Obstetric Survey Systems (INOSS): benefits of multi-country studies of severe and uncommon maternal morbidities. Acta obstetricia et gynecologica Scandinavica. 2014 Feb;93(2):127-31. 22. Schaap T, Bloemenkamp K, Deneux-Tharaux C, Knight M, Langhoff-Roos J, Sullivan E, et al. Defining definitions: a Delphi study to develop a core outcome set for conditions of severe maternal morbidity. BJOG : an international journal of obstetrics and gynaecology. 2017 Jul 29. 23. Björnberg A. Euro Health Consumer Index 2016: Health Consumer Powerhouse Ltd., 2017; 2017. 24. Gerkens S, Merkur S. Belgium: Health system review. Health Syst Transit. 2010;12(5):1-266, xxv. 25. OECD/European Observatory on Health Systems and Policies. Belgium: Country Health Profile 2017: OECD Publishing/European Observatory on Health Systems and Policies. 26. Federal Public Service of Health. Lijst met alle artsen die over een visum beschikken. 2017. Retrieved 2017 November from: https://www.health.belgium.be/en/node/27461#lijst 27. Benahmed N, Hendrickx E, Adriaenssens J, Stordeur S. Planning van gezondheidszorgpersoneel en gegevens over vroedvrouwen - Synthese. Health Services Research (HSR). Brussel: Belgian Health Care Knowledge Centre (KCE). 2016. KCE Report 278As. D/2016/10.273/93.; 2016. 28. Eurostat. Statistics Explained. Healthcare personnel statistics - nursing and caring professionals. Main statistical findings. Healthcare personnel - midwives. 2016. Retrieved 2017 November from: http://ec.europa.eu/eurostat/statistics-explained/index.php/Healthcare_personnel_statistics_- _nursing_and_caring_professionals 29. Van Kelst L, Spitz B, Sermeus W, Thomson AM. A hermeneutic phenomenological study of Belgian midwives' views on ideal and actual maternity care. Midwifery. 2013 Jan;29(1):e9-17. 30. Agentschap Zorg en Gezondheid. Belangrijkste trends in geboorte en bevalling. Retrieved 2017 October from: https://www.zorg-en-gezondheid.be/sites/default/files/atoms/files/ Evaluatierapport%20SPE%202015.pdf 31. Centre d'Épidémiologie Périnatale. Retrieved 2017 October from: http://www.cepip.be/ 32. Van Parijs A-S, Lucet C, Remacle A, Di Zinno T, Verstraelen H, Mambourg F, et al. Maternal Intensive Care in Belgium. Health Services Research (HSR). Brussel: Federaal Kenniscentrum voor de Gezondheidszorg (KCE); 2007. Report No.: (D/2008/10.273/77) 33. Leroy C, Van Leeuw V, Englert Y, Zhang W. Santé périnatale en Wallonie; 2015. 34. Van Leeuw V, Leroy C, Englert Y, Zhang W. Santé périnatale en Région bruxelloise; 2015. 35. Devlieger R, Martens E, Martens G, Van Mol C, Cammu H. Perinatale activiteiten in Vlaanderen 2015; 2015. 36. Gyselaers W, Jonckheer P, Ahmadzai N, Ansari M, S. C, Dworzynski K, et al. What are the recommended clinical assessment and screening tests during pregnancy? Good Clinical Practice (GCP) Brussels: Belgian Health care Knowledge Centre (KCE). 2015. KCE Report 248. D/2015/10.273/58. Retrieved 2017 November from: Https://kce.fgov.be/sites/default/files/atoms/files/ KCE_248_assessments_and_test_during_pregnancy_Report_0.pdf 37. Roelens K, Roberfroid D, Ahmadzai N, Ansari M, Singh K, Gaudet L, et al. KCE Report 228. Prevention of preterm birth in women at risk: selected topics. Good Clinical Practice (GCP) Brussels: Belgian Health Care Knowledge Centre (KCE). 2014. KCE Report 228. D/2014/10.273/63. Retrieved 2017 November from: https://kce.fgov.be/sites/default/files/atoms/files/KCE_228_Preterm%20birth_Report.pdf 38. Mambourg F, Gailly J, Wei-Hong Z. Guideline relative to low risk birth. Good Clinical Practice (GCP). Brussels: Belgian Health Care Knowledge Centre (KCE). 2010. KCE Reports 139C. D/2010/10.273/64. Retrieved 2017 November from: https://kce.fgov.be/sites/default/files/atoms/files/kce_139c_guideline_low_risk_birth.pdf 39. Algemene Directie Statistiek - Statistics Belgium - Statbel. Metagegevens. De statistiek van de moedersterfte. Retrieved 2017 November from: http://statbel.fgov.be/nl/binaries/meta_matmort_nl_tcm325-281108.pdf 40. The American Congress of Obstetricians and Gynecologists. Severe Maternal Morbidity: Clarification of the New Joint Commission Sentinel Event Policy. 2015. Retrieved 2017 November from: https://www.acog.org/About-ACOG/News-Room/Statements/2015/Severe-Maternal-Morbidity- Clarification-of-the-New-Joint-Commission-Sentinel-Event-Policy

OBJECTIVES

In summary:

We aim to develop a nationwide obstetric surveillance system to identify and analyse severe obstetric complications in Belgium. The primary objective of this system is to get an accurate picture of the obstetric complications under surveillance, the secondary objective is to improve the quality and safety of obstetric health care in Belgium by practical recommendations based on the findings of the surveys. Furthermore, the ambition is to develop an obstetric surveillance system that is performing efficiently and of high quality, to become a respectable partner of the International Network of Obstetric Survey System (INOSS), capable to co-operate and compare with other international obstetric surveillance systems.

In targets:

i. To conduct descriptive epidemiological studies on the severe obstetric complications under surveillance: to describe - the prevalence, - the risk-factors, - the management, - the outcome for mother and child. ii. To compare the prevalence, risk-factors, management and outcome of severe obstetric complications in Belgium with comparable international studies. iii. To formulate recommendations for prevention: primary prevention (based on risk-factors) as well as secondary prevention (based on management and observed quality of care concerns). To formulate national guidelines. iv. To evaluate the performance of B.OSS: a. the participation rate of Belgian maternity units (the number of maternity units who formally agree to participate in B.OSS, appointed a contact person and receive monthly mailing, in relation to the total number of maternity units), b. the response rate (the number of monthly reporting forms returned in relation to the total number of reporting forms sent), c. the number of completed data collection forms and accuracy of completion, d. and the performance of B.OSS within the INOSS.

METHODOLOGY

Institute.

The Belgian Obstetric Surveillance System, briefly B.OSS, was set up in 2011 with support of the College of Physicians for Mother and Newborn, a consultative body of the Federal Public Service of Health.

B.OSS is endorsed by the two professional associations for OB-GYNs (VVOG and GGOLF-B) and by the perinatal registries SPE and CEpiP. The College operates as the steering committee of B.OSS. In 2017, a formal scientific board is constituted with representatives of SPE, CEpiP, the Belgian Health Care Knowledge Centre (KCE), the Scientific Institute of Public Health (WIV-ISP) and the College. Daily reporting and data collection tasks are carried out by two cooperating teams: one in Flanders, another in Brussels and Wallonia.

Data collection.

B.OSS has adopted the methodology for case reporting developed by the UK Obstetric Surveillance System (UKOSS).1 An appointed contact person (OB-GYN or senior midwife) in each participating maternity unit is invited by monthly mailing to report a selected number of rare obstetric complications that occurred in the preceding month or alternatively to state that there was ‘nothing to report’, as is mostly the case. In the event of a case being reported, the contact person is asked to complete an extensive data collection form. In case of incomplete reporting, the contact person is encouraged repeatedly by email and phone to provide the missing data.

Website www.b-oss.be

Initially, the data of the reported cases were obtained through use of a standardised form, filled out electronically or on hard copy. Web-based data-collection was

gradually introduced following the launch of the B.OSS website www.b-oss.be in August 2013, facilitating monthly reporting and completion of data collection forms online. Monthly emails calling to report for the previous month are generated automatically, with reminders for missing reporting forms and incomplete data collection forms. Restricted access to the website is provided to the appointed contact person, one per maternity unit, via a personal login. They have access to the reporting forms and data collection forms of their maternity unit. Data protection is secured by the use of anonymous hash codes, replacing person-identifiable information such as the woman’s name, date of birth or hospital number.

Figure 4. B.OSS website www.b-oss.be

Obstetric complications under surveillance

B.OSS started the surveillance with three rather ‘common’ severe complications: namely a two-year study of uterine rupture (2012-2013), a two-year study of peripartum hysterectomy and/or arterial embolisation of the uterine arteries (2012- 2013) and a three-year study of eclampsia (2012-2014). Thereafter, B.OSS initiated a four-year study of antenatal pulmonary embolism (APE) (2015-2018) and participated in the INOSS international studies of spontaneous haemoperitoneum in pregnancy (SHiP) (2015-2017) and anaphylaxis in pregnancy (2016-2018). In november 2017 B.OSS participated in the one-week Global Maternal Sepsis Study (GLOSS) organised by the World Health Organisation.

Figure 5. Poster with current B.OSS studies in 2017.

Registered variables

The data collection forms seek information on maternal characteristics, medical, surgical and obstetric history, details of the index pregnancy, details of the delivery, the circumstances of the adverse event, its management and the outcome for mother and neonate. The data are retrieved in retrospect from the woman’s case notes.

Data analysis

The collected data are coded and exported as comma-separated values (csv) files, then cleaned and processed into operational SPSS data files. Data-analysis is performed using IBM SPSS statistics (IBM Corp. Released 2013. IBM SPSS Statistics for

Windows, Version 22.0. Armonk, NY: IBM Corp), EpiTools epidemiological calculators (Sergeant, ESG, 2017. Epitools epidemiological calculators. Ausvet Pty Ltd. Available at: http://epitools.ausvet.com.au) and MedCalc for Windows, version 15.0 Software (MedCalc Software, , Belgium). The prevalence of the obstetric events targeted is estimated using as denominator the total number of deliveries in Belgium during the survey period, corrected for the maternity units that did not participate. Reference data are obtained from the perinatal registries (SPE and CEpiP) when available. 2, 3

Note: Epidemiologists will debate whether ‘incidence’ or ‘prevalence’ is most appropriate to describe the frequency in this context of obstetric complications.4 The ‘population at risk’ are pregnant women and pregnancies vary in duration lasting 42 weeks at most. Therefore, the frequency of obstetric complications is commonly reported as a proportion, e.g. per 10,000 deliveries or delivery hospitalisations or pregnancies. Many authors refer to the frequency of obstetric complications as the ‘incidence’ (including the United Kingdom of Obstetric Surveillance System (UKOSS)), while others use ‘prevalence’ (including the World Health Organisation (WHO)). ‘Incidence’ seems more appropriate, because an obstetric complication concerns a ‘new’ condition or injury, while prevalence is more often used for chronic conditions including both new and pre-existing cases. Our reservations to use ‘incidence’ were 1) that ‘incidence proportion’ requires the size of the population at start of the study period as denominator and 2) that ‘incidence rate’ is explicitely expressed per units of time (e.g. per year). We decided to use the term ‘prevalence’, defined as the proportion of a population that has a particular disease or injury at any time during the interval, to desribe the burden of these obstetric complications in Belgium.

Ethics approval

The B.OSS methodology was approved by the Medical Ethics Committee of Ghent University Hospital (EC UZG 2012/734; B670201215359) and by the Medical Ethics Committee of the Erasme University Hospital, Brussels (EC ULB 2012/111; B406201213660) at the beginning. The Medical Ethics Committee of the Ghent University Hospital became Central Ethics Committee in 2015 (EC UZG 2015/1470; B670201526875). The women eligible for inclusion are informed by their OB-GYN and offered an information letter enabling them to opt-out. Confidentiality is guaranteed for mother, provider and hospital. Person-identifiable information is eliminated from data-analysis.

References

1. Knight M, Kurinczuk JJ, Tuffnell D, Brocklehurst P. The UK Obstetric Surveillance System for rare disorders of pregnancy. BJOG : an international journal of obstetrics and gynaecology. 2005 Mar;112(3):263-5. 2. Agentschap Zorg en Gezondheid. Belangrijkste trends in geboorte en bevalling. Retrieved 2017 October from: https://www.zorg-en-gezondheid.be/sites/default/files/atoms/files/ Evaluatierapport%20SPE%202015.pdf 3. Centre d'Épidémiologie Périnatale. Retrieved 2017 October from: http://www.cepip.be/ 4. Centers for Disease Control and Prevention. Principles of Epidemiology in Public Health Practice. Third Edition. October 2006, last update November 2011. Retrieved 2018 March from: https://www.cdc.gov/ophss/csels/dsepd/ss1978/SS1978.pdf

RESULTS

The first registration round of B.OSS (2012-2013) resulted in two scientific papers reporting on the prevalence, risk factors, management and outcome of uterine rupture and of peripartum hysterectomy and/or embolisation of the uterine arteries in Belgium.

Within the INOSS network, we conducted an international comparison study of uterine rupture between nine countries, resulting in the largest prospective population-based report of uterine rupture in high-income countries.

We evaluated the performance of B.OSS, its role within the INOSS, the possibilities to improve. We consulted two national data sources that could possibly serve as a control for underreporting for the data recorded by B.OSS.

We drafted a nationally adopted guideline on trial of labour after caesarean section (TOLAC).

NATIONWIDE POPULATION-BASED COHORT STUDY OF UTERINE RUPTURE IN BELGIUM: RESULTS FROM THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM G Vandenberghe1, M De Blaere1, V Van Leeuw2, K Roelens1, Y Englert3, M Hanssens4, H Verstraelen1.

1. Department of Obstetrics & Gynaecology, Ghent University Hospital, Ghent, Belgium 2. Centre d'Epidémiologie Périnatale (CEpiP), Bruxelles, Belgium 3. Faculty of Medicine, Centre d'Epidémiologie Périnatale (CEpiP), Research Laboratory on Human Reproduction, Université Libre de Bruxelles (ULB), Brussels, Belgium 4. Department of Obstetrics & Gynaecology, University Hospital Leuven, Leuven, Belgium

Abstract

Objectives We aimed to assess the prevalence of uterine rupture in Belgium and to evaluate risk factors, management and outcomes for mother and child. Design Nationwide population-based prospective cohort study. Setting Emergency obstetric care. Participation of 97% of maternity units covering 98.6% of the deliveries in Belgium. Participants All women with uterine rupture in Belgium between January 2012 and December 2013. 8 women were excluded because data collection forms were not returned. Results Data on 90 cases of confirmed uterine rupture were obtained, of which 73 had a previous Caesarean section (CS), representing an estimated prevalence of 3.6 (95% CI 2.9 to 4.4) per 10 000 deliveries overall and of 27 (95% CI 21 to 33) and 0.7 (95% CI 0.4 to 1.2) per 10 000 deliveries in women with and without previous CS, respectively. Rupture occurred during trial of labour after caesarean section (TOLAC) in 57 women (81.4%, 95% CI 68% to 88%), with a high rate of augmented (38.5%) and induced (29.8%) labour. All patients who underwent induction of labour had an unfavourable cervix at start of induction (Bishop Score ≤7 in 100%). Other uterine surgery was reported in the history of 22 cases (24%, 95% CI 17% to 34%), including 1 case of myomectomy, 3 cases of salpingectomy and 2 cases of hysteroscopic resection of a uterine septum. 14 cases ruptured in the absence of labour (15.6%, 95% CI 9.5% to 24.7%). No mothers died; 8 required hysterectomy (8.9%, 95% CI 4.6% to 16.6%). There were 10 perinatal deaths (perinatal mortality rate 117/1000 births, 95% CI 60 to 203) and perinatal asphyxia was observed in 29 infants (34.5%, 95% CI 25.2% to 45.1%). Conclusions The prevalence of uterine rupture in Belgium is similar to that in other Western countries. There is scope for improvement through the implementation of nationally adopted guidelines on TOLAC, to prevent use of unsafe procedures, and thereby reduce avoidable morbidity and mortality.

Background

Uterine rupture is a rare, but potentially devastating complication of pregnancy, whereby the uterine wall tears during pregnancy or delivery, extending to the uterine serosa and occasionally involving the bladder or broad ligament.1 ,2 The course is usually milder in case of a uterine dehiscence, when rupture is confined to the uterine muscle without affecting the uterine serosa.1 ,2 Disruption of the uterine wall can be accompanied by displacement of the baby into the abdomen, causing severe asphyxia and perinatal death, and by massive maternal bleeding necessitating massive transfusion or hysterectomy to save the mother's life. Albeit a rare event, complicating an estimated 0.03% of pregnancies in developed countries,3 the severity of this complication has a profound impact on obstetrical practice. Large population-based studies in European countries, including obstetric surveillance systems,4 ,5 have improved our knowledge of risk factors and management of uterine rupture, and have guided strategies for prevention.6–10 In developed countries, uterine rupture most commonly occurs in uteri scarred by previous caesarean section (CS), mainly at term and during labour, and is even more frequent when labour is induced or augmented. Awareness of this risk has contributed to a more cautious policy towards trial of labour after caesarean section (TOLAC), stated in widely used guidelines.1 ,11 ,12 Belgium has a large number of tertiary referral centres providing neonatal intensive care (n=19) along with a high concentration of small-volume maternity units.13 Obstetric care is mainly organised in private practice. The impact of this almost exclusively specialist-led care, the high rate of induction of labour13(26.5% in 2013)14–16 and the steadily increasing CS rate (20.7% in 201314–16) on the prevalence of uterine rupture in Belgium is currently not known. We aimed to investigate the prevalence of uterine rupture in Belgium and to assess risk factors, management and outcomes, for mother and child.

Methods

Uterine rupture in Belgium has been registered as part of a national reporting system to study rare obstetric complications, being the first completed study of this newly developed project. The B.OSS was constituted in 2011 and initiated registration in January 2012, with support of the College for Mother and Newborn, a consultative body of the Federal Public Service of Health in Belgium.

B.OSS has adopted the methodology for case reporting of severe obstetric morbidity developed by the UK Obstetric Surveillance System (UKOSS),4 pioneer of the obstetric surveillance systems united in the International Network of Obstetric Survey Systems (INOSS).5 Participation of 97% of Belgian maternity units (n=113) was obtained, covering 98.6% (248 681) of deliveries in the study period. Briefly, an appointed contact person (obstetrician (OB)/gynaecologist or senior-midwife) in each participating maternity unit (n=110) was invited by monthly mailing to report a selected number of rare obstetric complications (including uterine rupture, as well as eclampsia and peripartum hysterectomy, and/or embolisation for massive obstetric haemorrhage) that had occurred in the preceding month. In the event a case was reported in reply, the contact person was asked to complete an extensive data collection form. Confidentiality was guaranteed for mother, provider and hospital; person-identifiable information was eliminated from data-analysis. In case of incomplete reporting, the appointed contact person was encouraged repeatedly by email and phone to provide missing data, up to 6 months following the period of case reporting.

Initially, data on reported cases were obtained through use of a standardised form, filled out electronically or on hard copy, according to preference. Web-based data- collection was gradually introduced following the launch of the B.OSS website (http://www.b-oss.be) in August 2013, facilitating monthly reporting and completion of data collection forms online. Standardised forms were entered manually into the website forms by two investigators, independently, to assure quality of data entry. Collected data of completed online data collection forms were coded and exported as a comma-separated values file.

Data collection forms requested information on maternal characteristics, medical history and obstetrical history, details on the index pregnancy, circumstances of the event, and the management and outcome for mother and neonate.

B.OSS aimed to collect all cases of uterine rupture in Belgium at any gestational age corresponding to an anatomical definition (delineated by UKOSS7) or to a clinical definition (delineated by the LEMMoN study in the Netherlands8), allowing for the comparison of Belgian data with both population-based studies. Uterine rupture as defined by UKOSS is ‘a complete separation of the wall of the pregnant uterus, with or without expulsion of the fetus, involving rupture of the membranes at the site of the uterine rupture or extension into uterine muscle separate from any previous scar, and endangering the life of the mother or the fetus’. Uterine rupture as defined by the LEMMoN study is ‘the occurrence of clinical symptoms leading to an emergency CS, at which the presumed diagnosis of uterine rupture was confirmed; or a peripartum hysterectomy or laparotomy is required for uterine rupture after a vaginal birth’. Anatomical details were requested in the data collection forms, defining the status of the myometrium, the peritoneum and the fetal membranes. This anatomical distinction enables the identification of cases that correspond to the UKOSS definition and the comparison with other population-based studies that have used anatomical definitions in the same manner. B.OSS has specifically excluded cases of asymptomatic dehiscence (intact uterine peritoneum) noted incidentally at CS, in accordance with UKOSS7 and LEMMoN.8

B.OSS aimed to collect all cases of uterine rupture occurring in Belgium from January 2012 to December 2013.

Eight reported cases were excluded from the present analysis, because data collection forms were not returned. Data collection forms that were returned were not always complete for all variables. Study data were evaluated by the number of cases missing per variable and by the number of variables missing per case. No further cases were excluded because of missing data. Statistical analysis The most recently available data from the Belgian perinatal registry were used as denominator or control group to calculate rates of prevalence and relative risks. The Belgian perinatal registry (Studiecentrum voor Perinatale Epidemiologie (SPE) in Flanders,14 Centre d'Epidémiologie Périnatale (CEpiP) in Brussels15 and Wallonia16) collects data on a mandatory basis, covering nearly 100% of births in Belgian maternity units and homebirths. A selected set of perinatal data are recorded by the OB, midwife and neonatologist immediately after a birth. According to Belgian legislation, births are registered as such, in case of a live birth or stillbirth at or after 22 completed weeks of gestation or, if gestational age is missing, at a birth weight of 500 g or more. The overall prevalence of uterine rupture was calculated using the total number of mothers who had given birth in Belgium in 2012–2013, as denominator, corrected for three hospitals that did not participate in B.OSS. The number of mothers with previous CS, in 2012–2013, was used as denominator to calculate the prevalence of

uterine rupture in women with and without previous CS. Belgium has a high rate of induction of labour, but data of method or medication used for induction are not available. Sociodemographic and obstetric risk factors were obtained from SPE14 and CEpiP (2012–2013)15 ,16; data on marital status and employment were only available from CEpiP. Neonatal asphyxia was defined as Apgar score <7 at 5 min and/or the umbilical artery pH≤7.0, and/or need for resuscitation including endotracheal ventilation, and/or neurological manifestations (seizures), and/or whole body hypothermia treatment initiated after birth.17 Relative risks and 95% CIs were calculated in univariate analyses, comparing frequencies of events between the study group and control group, with ordinary Pearson's χ2 tests using weighted cases when appropriate. p Values of <0.05 were considered statistically significant. Data were analysed using the statistical software package IBM SPSS statistics V.22.0 (IBM Corp, IBM SPSS Statistics for Windows, Version 22.0, Armonk, New York, USA: IBM Corp, 2013).

Results

From January 2012 to December 2013, 90 confirmed cases of uterine rupture were documented. Eight more reported cases needed to be excluded because of lacking data collection forms. The corresponding prevalence of uterine rupture in Belgium is therefore estimated at 3.6 (95% CI 2.9 to 4.4) per 10 000 deliveries. When taking into account only cases of complete rupture (n=74), the estimated prevalence is 2.9 (2.3 to 3.7). Details on reported cases and completeness of data collection forms are shown in figure 1.

Figure 1. Flowchart of case reporting and data collection.

Women's characteristics Relevant sociodemographic, medical and obstetrical characteristics are displayed in table 1, in comparison with national data when available. Of note, patients with uterine rupture included six cases (7.7%) with documented uterine anomaly, of whom two had a septal resection prior to the index pregnancy, and one case of uterine rupture in a patient with osteogenesis imperfecta type 1. Rupture in the latter patient, who had undergone a previous CS, occurred during the latent phase of labour at term (38 weeks) and involved a longitudinal tear towards the fundus.

Table 1. Relevant sociodemographic, medical and obstetrical risk factors in cases of uterine rupture

Cases of Background Unadjusted p-value uterine rupture population RR (95% CI) 2012-2013 2012-2013 (n=90)* (n=252,272) n, % n, %

Sociodemographic factors

Maternal age ≥ 35 years 17 (18.8) 43,256 (17) 1.11 (0.72-1.70) 0.6

BMI at booking ≥ 30 kg/m2 13 (19.6) 29,453 (12) 1.59 (0.9-2.59) 0.05

Ethnicity non-white ** 5 (5.5) - - - Not married or cohabiting 3 (3.8) 20,785 (17) 0.22 (0.07-0.69) 0.008 Unemployed 22 (24.4) 51,296 (43) 0.72 (0.50-1.02) 0.06 Smoking 8 (8.8) - - -

Medical factors

Uterine anomaly (bicornuate, septate) 6 (6.6) - - - Connective tissue disease*** 1 (1.1) - - -

Obstetrical factors

Parity ≥ 3 9 (10) 18,855(7.4) 1.33 (0.71-2.48) 0.3 Nulliparity 6 (6.6) 11,0711(43.9) 0.15 (0.07-0.32) <0.0001

Induction of labour 24 (27.9) 68,703 (26.7) 1.03 (0.7-1.4) 0.8

Previous caesarean delivery (CD) 73 (81.1) 27,007 (10.7) 7.57 (6.85-8.37) <0.0001 Previous CD, cases <24 weeks excluded 70 (81.3) 12,754 / 27,007 Trial of labour after caesarean delivery 57/70 (81.4) 1.39 (1.15-1.69) <0.001 (47.2) Interpregnancy interval ≤ 12 months 12 (16.4) - - - Artifical Reproductive Technology 4 (4.4) 9,233 (3.7) 1.21 (0.46-3.15) 0.6 (IVF/ICSI) Abnormal placentation (accreta, increta, 1 (1.1) - - - percreta)

Birthweight ≥ 4000 g 11 (12.2) 19,967 (7.8) 1.67 (0.96-2.90) 0.06 Breech presentation 4 (4.4) 12,630 (4.9) 0.93 (0.35-2.42) 0.8 * Study data were not complete for all variables; number of data available on BMI (n=66), marital status (n=77), employment (n=71), smoking (n=87), ART (n=89), birth weight (n=84), interpregnancy interval (n=67) ** White ethnicity defined as European, Middle-East, North-African, white South-African. Five women were of African descent defined as African, Caribbean or Afro-American ethnicity. *** Connective tissue disease: the study included one case known with Osteogenesis imperfecta type I. Obstetrical risk factors

Seventy-three cases of uterine rupture (81.1%) occurred in women who had at least one previous CS, representing an estimated prevalence of 27 (95% CI 21 to 33) per 10 000 deliveries in women with previous CS. Seventeen cases of uterine rupture (18.8%) were observed in women who did not have a previous CS, representing an estimated prevalence of 0.7 (95% CI 0.4 to 1.2) per 10 000 deliveries in women without previous CS. Eight women without a previous CS (8/17) had one or more other types of uterine surgery documented in their history (dilation and curettage (D&C) (n=5), salpingectomy (n=2), resection of uterine septum (n=2), myomectomy (n=1), manual removal of the placenta (n=1)). In none of these cases was uterine perforation confirmed. Similarly, 14 women with previous CS also had a history of another type of uterine surgery (D&C (n=12), manual removal of placenta (n=1), salpingectomy for an interstitial pregnancy (n=1)); with documented uterine perforation in the latter case followed by suturing of the uterine cornua. Interestingly, this rupture occurred at the site of the cornual scar. Finally, nine women did not have any surgical risk factor that could explain a weakness in the uterine wall (see online supplementary figure S1). In the cases with previous CS, the interpregnancy interval was beyond 24 months in 58% of the cases, 12–23 months in 24% of cases and <12 months in 12 cases (18%), of whom four became pregnant within the first 6 months after previous CS.

Four cases of uterine rupture occurred before 24 completed weeks of gestation. In one patient, uterine rupture occurred during termination of pregnancy at 18 weeks, by means of prostaglandins, in a patient with two previous CS. The second patient received prostaglandins because of intrauterine fetal death (IUFD) at an estimated 22 weeks in an unrecognised pregnancy; this patient had undergone four previous CS. The third case was a rupture at 18 weeks in a previously unrevealed bicornuate uterus. The fourth case was a fundal rupture at 18 weeks at the site of a placenta percreta. This patient had two previous CS and two D&C's for termination of pregnancy and was managed by a hysterectomy. These four cases were excluded from analysis of neonatal outcome. The median gestational age of the 86 remaining cases was 39 weeks (range 25–41 weeks). In 71 cases (82.5%), uterine rupture occurred during labour. Of these, labour was induced in 22 cases (25.5%) and augmented by use of oxytocin in 24 cases (28%) while another 24 cases laboured spontaneously without the use of uterotonic drugs (28%). Sixteen cases were stated to have ruptured in the absence of labour (18.6%), two of these women having received vaginal prostaglandins in an attempt to induce labour (see online supplementary figure S2). The rate of TOLAC in this study was 81.4% (n=57), including in 17 women where it was induced (29.8%) and in 22 where it was augmented (38.5%). The Bishop score prior to induction, as the most commonly used predictor for the readiness of the

cervix for induction,18 was unfavourable in the majority of these women (<6 in 82%, ≤7 in 100%).

Signs and symptoms that raised suspicion of uterine rupture are demonstrated in table 2. In five cases, there were signs of maternal shock (hypotension, sweating, pallor, maternal collapse); shoulder pain was mentioned in one case. In 19 cases (21.1%), the diagnosis was made after vaginal delivery, with vaginal bleeding occurring as the presenting sign in 47% of the cases. Delivery was assisted with vacuum (n=10) or forceps (n=2) because of fetal distress (n=7) or maternal reasons (n=5). One case of uterine rupture was detected during routine manual examination of the uterus postdelivery, an obsolete procedure that is still practised by a minority of OBs.

Table 2. Signs and symptoms, and combinations of symptoms prior to diagnosis of uterine rupture Presence Combination of two symptoms of symptom N (%) Abdominal Vaginal Hypertonia Acute Haema- pain bleeding absence turia of con- tractions Abnormal CTG 55/74 36/74 16/74 9/73 6/74 4/74 (74.3) (48.6) (21.6) (12.3) (8.1) (5.4) Abdominal 59/89 16/87 8/86 5/87 5/87 pain (66.2) (18.3) (9.3) (5.7) (5.7) Vaginal 27/86 1/86 3/86 1/86

bleeding (31.3) (1.1) (3.4) (1.1) Hypertonia 9/85 2/85 1/85

(10.5) (2.3) (1.1) Acute absence 8/86 3/86 of contractions (9.3) (3.4) Haematuria 6/86

(6.9)

Emergency CS was performed in 67 women (74%). Dilation at the moment of decision was 0–3 cm in nearly half of these women (n=32), while nine women (13%) were fully dilated. Eight women underwent a hysterectomy (9%) (see online supplementary figure S3). Reported reason to perform the hysterectomy was massive haemorrhage in four cases while the other four cases underwent hysterectomy mainly because of major anatomical damage to the uterus. Location of the uterine rupture was in the lower uterine segment in the majority of cases (n=57, 63%). Other reported locations of rupture were fundal (n=13, 14%), longitudinal (n=12, 13%), high transverse (n=4, 4%) and cornual (n=2, 2%). The bladder or the parametrium was involved in 7 (8%) and in 12 women (13%), respectively. Twenty women (22.2%) were admitted to an intensive care unit (ICU), with a mean stay of 2.1 (1–7) days. Three women required intubation and ventilation, one patient suffered from aspiration pneumonia. Thirty-six women received transfusion (40%), 14 women were transfused with >4 units of packed cells (PC) and fresh frozen plasma (FFP). Other reported maternal morbidity included a vesicovaginal fistula (n=1), a nephrostomy (n=1) and an abcedation (n=1). The neonatal outcome is demonstrated in the flowchart in figure 2. There were 10 perinatal deaths in consequence of the uterine rupture: two neonates died during rupture before birth and eight neonates died in the first 7 days of life. Early neonatal deaths occurred at a term gestational age (>39+0 weeks),19with the exception of one at extreme preterm (25 weeks) and two at moderate preterm (34 weeks) gestational age. Perinatal mortality observed with uterine rupture occurred at a rate of 117/1000 (95% CI 60 to 203) compared with the perinatal mortality rate of 6.7/1000 in the background population (RR 17.3, 95% CI 9.6 to 31.0).

Figure 2. Neonatal outcome in 90 cases of uterine rupture. Cong, congenital; IUFD, intrauterine fetal death.

Seventy-four neonates survived (86%), of whom 29 suffered from asphyxia (39%). Twenty-four neonates needed intubation and ventilation (33%), two neonates needed resuscitation at birth, five received cooling therapy (7%) and one neonate required anticonvulsant drugs. Median gestational age at birth was 39 weeks (range 29–41 weeks) in the group of surviving neonates. A higher risk of severe maternal outcome (hysterectomy, transfusion >4 units of PC, admission to ICU) and severe neonatal outcome (asphyxia, intrapartum or neonatal death) was observed in women without previous CS than in women with a previous CS. A more in-depth analysis taking history of other uterine surgery into account reveals a further increased risk of severe maternal outcome in those women with other uterine surgery; however, numbers in this study are too small to permit firm conclusions (see online supplementary table S1).

Discussion

The estimated prevalence (3.6/10 000 deliveries) of uterine rupture in Belgium, based on this study, is similar to that found in other Western countries. Compared with three national population-based studies that have used a similar approach: the Netherlands (LEMMoN-study),8 the UK (UKOSS)7 and the Nordic countries (Nordic Obstetric Surveillance Study, NOSS),10 this prevalence stands roughly midway between the low rate of UKOSS (1.9/10 000) and the higher rates of NOSS (5.5/10 000) and LEMMoN (5.9/10 000). Important is that UKOSS and NOSS have included only cases of complete rupture, while LEMMoN has no information on the anatomical details of the ruptures. The estimated prevalence of complete uterine ruptures in Belgium is 2.9/10 000 deliveries. The strengths of this study are its methodology—collecting cases prospectively on a monthly basis—and its national design with excellent participation and response rates (98.9%) from the Belgian maternity units. The small number of cases, intrinsic to rare but severe obstetric complications, warrants caution, however, in making strong inferences. Furthermore, we were unable to control for several variables not available from the background population through the Belgian perinatal registry. Finally, due to the study design and non-mandatory reporting of severe pregnancy complications, under-reporting may have biased our results to some extent, an issue we will address in a forthcoming study. The majority of cases in this study fit the pattern of known risk factors for uterine rupture: scarring by a previous CS, being beyond 37 weeks of gestation, being in labour and being induced or augmented. Still, 17 cases had no previous CS, of which 8 ruptured before 37 weeks, and of which 2 were not in labour. Insight into circumstances and risk factors of rare cases of uterine rupture, namely occurring in the unscarred and/or preterm, and/or non-labouring uterus, can be improved by the study of a larger number of cases; the INOSS collaborative5 may be the only route to achieve adequate numbers of these rare cases of uterine rupture. This study confirms a higher risk of severe maternal and neonatal outcome when a rupture occurs in a uterus without previous CS, than among women with previous CS, as reported previously by the LEMMoN study,8 among others.3 ,20 This can be explained by the delay in diagnosis because of a lower suspicion of uterine rupture in an unscarred uterus and by the higher risk of massive haemorrhage in the more vascular, unscarred myometrium compared with in uterine scar tissue. A further increased risk of severe maternal outcome was observed when accounting for a

history of other uterine surgery, though our observations are hypothesis-generating and need confirmation in larger study sets. Evidence on the risk of uterine rupture following transabdominal or laparoscopic myomectomy remains inconclusive, because surgeons recommend a primary CS in a high number of these women. Based on the literature review, the risk of uterine rupture following myomectomy appears to be comparable to the risk after previous low transverse CS (around 0.5–0.7%).21 Evidence-based guidelines on the mode of delivery following myomectomy, however, are not available.1 Further, this study includes three cases of fundal uterine rupture following salpingectomy, and two cases following hysteroscopic resection of a uterine septum. Evidence of uterine rupture following salpingectomy22 ,23 or following resection of a uterine septum24 is even more scarce, limited to case reports and series. In line with myomectomy, surgeons should evaluate and document the risk of uterine rupture based on intraoperative findings; next, OBs should be aware of this potential risk during antenatal care. Finally, to what extent can we assign the occurrence of a uterine rupture to an unnoted perforation during routine surgical procedures such as D&C (n=17 in this study), diagnostic or simple operative hysteroscopy, or even the insertion of an intrauterine device? The exceptional case reports in the literature, in view of the frequency of these procedures, suggest that these are merely accidental findings.25 ,26 Information on the prevalence of these routine surgical procedures in the background pregnant population, in relation to the prevalence of these procedures in a uterine rupture cohort, could provide a more accurate estimation of the relative risk. Linkage between national birth and medical registers, as is the case in the Nordic countries,27 may allow for studies of such magnitude. The overall CS rate in Belgium is slowly increasing and has just exceeded the landmark of 20%. Although this rate is still acceptable13 the trend causes concern, awakened by recent reporting of the WHO stating that CS rates higher than 10% are not associated with reductions in maternal and newborn mortality rates.28 Belgium has no national guidelines for TOLAC specifically, hence Belgian OBs will turn to various guidelines (Royal College of Obstetricians and Gynaecologists (RCOG),1 American College of Obstetricians and Gynecologists (ACOG),11 Nederlandse Vereniging voor Obstetrie & Gynaecologie (NVOG),29 French College of Gynecologists and Obstetricians (CNGOF)30 and literature to rely on in their daily practice. We observe a restrictive policy towards TOLAC in Belgium: in 2012–2013, 47.2% of Belgian women with previous CS attempted a vaginal birth in their next pregnancy, with success in 69%.14–16 The estimated TOLAC rate in the Netherlands and the UK was 71.7% and 52.2%, successful in 76% and 63.4%, respectively.31 ,32 The percentage of women with TOLAC that was induced or augmented in this study is similar to that reported in the LEMMoN study (28% induced, 25% augmented),8 yet higher than observed in the UKOSS study (14% induced, 25% augmented).7 Prostaglandins were used in 27% (compared with 13% and 14% in LEMMoN and UKOSS, respectively). The reported doses and regimens were not according to guidelines in a small number of cases (intervals <3 h, maximum 24 h dose exceeded, misoprostol for induction of labour at 38 weeks). Thus, five cases of TOLAC were noted to be induced at 38–39 weeks in the absence of a medical indication. These findings suggest that, despite an overall cautious attitude towards TOLAC in Belgium, a number of uterine rupture cases may have been avoidable as they were preceded by unneeded and unsafe procedures, inconsistent with current widely accepted guidelines.1 ,11 ,12 ,30 ,33 Accordingly, there seems to be scope for improvement in promoting a more liberal policy towards TOLAC in appropriate candidates in Belgium aiming to reduce the repeat and thus overall CS rate. Whether this will effectively result in a significant reduction of associated morbidity (rate of abnormally invasive placenta, massive obstetric haemorrhage, uterine rupture), may be the subject of future B.OSS surveillance.

Conclusion

On the basis of the findings in this study, we consider the need for nationally adopted guidelines on TOLAC in Belgium. These should be developed to improve the awareness of true uterine rupture risk, and thereby prevent unsafe procedures on the one hand, and promote safe TOLAC in appropriate candidates on the other, to enable reduction of the repeat and overall CS rate. There is need for further research on the true risk of uterine rupture following myomectomy and other uterine surgical procedures that similarly injure the myometrium, to develop evidence-based guidelines on the delivery mode in subsequent pregnancies. The INOSS enables the study of rare cases of uterine rupture, namely in the unscarred and/or preterm, and/or non-labouring uterus and recurrent uterine rupture.

Acknowledgments

This study would not have been possible without the voluntary participation of the Belgian maternity units: GZA Sint-Jozef, Mortsel; GZA Sint-Augustinus, Wilrijk; GZA Sint-Vincentius, Antwerpen; ZNA, Sint-Erasmus, Borgerhout; ZNA, Jan Palfijn, Merksem; ZNA, Middelheim, Antwerpen; ASZ, campus Aalst; ASZ, campus Geraardsbergen; AZ Lokeren; Imelda Ziekenhuis, Bonheiden; Sint-Jozef, Bornem; AZ Klina, Brasschaat; AZ Sint-Jan, Brugge; AZ Sint-Lucas, Assebroek; Centre de Santé des Fagnes; Centre Hospitalier de l'Ardenne; Centre Hospitalier EpiCura, site Hornu; Centre Hospitalier de Mouscron; Centre Hospitalier du Bois de l'Abbaye et de Hesbay; Centre Hospitalier Peltzer, La Tourelle; CHC Saint-Vincent; CHC Saint-Joseph; CHC Sainte-Elisabeth; CHIREC, Clinique Sainte-Anne, Saint-Remi; CHIREC Braine-l'Alleud- Waterloo; CHR de la Citadelle; CHR de Namur; CHR du Val de Sambre; CHR Haute Senne—Le Tilleriau; CHR Mons Clinique Saint-Joseph; CHU Ambroise Paré; CHU Brugmann; CHU Charleroi, André Vésale; CHU Notre-Dame des Bruyères; CHU Saint- Pierre; CHU Tivoli; CHwapi—Site Notre-Dame; Clinique Edith Cavell; Clinique et

Maternité Sainte Elisabeth; Clinique Notre dame de Grâce; Clinique Reine Astrid; Clinique Sainte-Piere; Cliniques de l'Europe, Saint-Michel; Cliniques de l'Europe, Sainte-Elisabeth; Cliniques du Sud Luxembourg; Cliniques Universitaires Saint-Luc; AZ Sint-Vincentius, Deinze; AZ Sint-Blasius, Dendermonde; AZ Monica, Deurne; AZ Diest; AZ Sint-Maarten, Duffel; AZ Alma, Eeklo; Dimpna Ziekenhuis, Geel; AZ Jan Palfijn, Gent; AZ Maria Middelares, Gent; AZ Sint-Lucas, Gent; GHDC asbl—Notre Dame; AZ Maria ziekenhuis, Halle; Jessaziekenhuis, Hasselt; AZ Sint Elisabeth, Herentals; CAZ Midden-Limburg, Heusden-Zolder; Hôpital Civil Marie Curie; Hôpital Erasme, Bruxelles; Hôpitaux Iris Sud-Site Ixelles; Jan Yperman ziekenhuis, Ieper; IFAC Hôpital Princesse Paola; INDC Entité Jolimontoise; Sint-Jozef ziekenhuis, Izegem; Kliniek Sint- Jan; Klinik Saint-Josef; AZ Zeno, campus Knokke-Heist; AZ Groeninge, Kortrijk; Heilig Hart ziekenhuis, Leuven; UZ Leuven; Heilig Hart ziekenhuis, Lier; AZ Sint-Jozef, Malle; AZ Sint-Maarten, Mechelen; AZ Delta, campus Menen; Heilig Hart ziekenhuis, Mol; Onze-Lieve-Vrouw ziekenhuis, campus Aalst; Onze-Lieve-Vrouw ziekenhuis, campus Asse; AZ Damiaan, campus Sint-Jozef, Oostende; AZ Sint-Jan, campus Oostende; Christine Dallequin AZ Oudenaarde; Mariaziekenhuis Noord-Limburg, Overpelt; AZ Heilige Familie, Reet AZ Delta, campus Brugsesteenweg, Roeselare; AZ Delta, campus Wilgenstraat, Roeselare; AZ Glorieux, Ronse; AZ Nikolaas, Sint-Niklaas; Sint-Trudo ziekenhuis, Sint-Truiden; Saint-Nikolaus Hospital; Sint-Andries ziekenhuis, Tielt; Heilig Hartziekenhuis, Tienen; AZ Vesalius, Tongeren; Sint-Rembertziekenhuis, Torhout; AZ Turnhout; UZ Antwerpen; UZ Brussel; AZ Sint-Augustinus, Veurne; AZ Jan Portaels, Vilvoorde; Onze-Lieve-Vrouw van Lourdes ziekenhuis, Waregem; Ziekenhuis Oost- Limburg, campus Sint-Jan Genk; AZ Sint-Elisabeth, Zottegem. The authors would like to thank Fatima Bercha, Marine Guisset, Ann Langedock, Marlies Maly, Emilie Vandevoorde and Iris Janssens for their contribution in collecting and processing the data.

References

1. Royal College of Obstetricians and Gynaecologists (RCOG). Birth after previous caesarean birth. Green- top guideline no. 45. Published 1 October 2015. www.rcog.org.uk/globalassets/documents/guidelines/gtg_45.pdf. (Accessed 21 Oct 2015). 2. Landon MB, Hauth JC, Leveno KJ et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med 2004;351:2581-9. 3. Hofmeyr GJ, Say L, lmezoglu AM. WHO systematic review of maternal mortality and morbidity: the prevalence of uterine rupture. BJOG 2005;112: 1221-8. 4. Knight M, Kurinczuk JJ, Tuffnell D et al. The UK Obstetric Surveillance System for rare disorders of pregnancy. BJOG 2005;112: 263-5. 5. Knight M, INOSS. The International Network of Obstetric Survey Systems (INOSS): benefits of multi- country studies of severe and uncommom maternal morbidities. Acta Obstet Gynecol Scand 2014;93:127-31 doi: 10.1111/aogs.12316 [Published Online First: 30 Dec 2013]. 6. Kaczmarczyk M, Sparen P, Terry P et al. Cnattingius S (2007) Risk factors for uterine rupture and neonatal consequences of uterine rupture: a population-based study of successive pregnancies in Sweden. BJOG 2007;114:1208-14. 7. Fitzpatrick EF, Kurinczuk JJ, Alfirevic Z et al. Uterine Rupture by Intended Mode of Delivery in the UK: A National Case-Control Study. PLoS Med 2012;9(3): e1001184 doi:10.1371/journal.pmed.1001184 [Published Online First: 13 Mar 2012]. 8. Zwart JJ, Richters JM, Ory F et al. Uterine rupture in The Netherlands: a nationwide population-based cohort study. BJOG 2009;116:1069-78. 9. Al-Zirqi I, Stray-Pedersen B, Forsén L et al. Uterine rupture: trends over 40 years. BJOG Published Online First: 2 April 2015. doi:10.1111/1471-0528.13394 10. Colmorn LB, Petersen KB, Jakobsson M et al.). The Nordic Obstetric Surveillance Study: a study of complete uterine rupture, abnormally invasive placenta, peripartum hysterectomy and severe blood loss at delivery. Acta Obstet Gynecol Scand 2015;94:734-44 doi: 10.1111/aogs.12639 [Published Online First: 17 Apr 2015]. 11. American College of Obstetricians and Gynecologists. ACOG Practice bulletin no. 115: Vaginal birth after previous cesarean delivery. Obstet Gynaecol 2010;116:450-63 doi: 10.1097/AOG.0b013e3181eeb251. 12. Society of Obsetricians and Gynaecologists of Canada. SOGT Clinical Practice Guidelines. Guidelines for vaginal birth after previous caesarean birth. Number 155 (Replaces guideline Number 147). Int J Gynaecol Obstet 2005;89:319-31. 13. EUROPEAN PERINATAL HEALTH REPORT. Health and Care of Pregnant Women and Babies in Europe in 2010. www.europeristat.com/images/doc/EPHR2010_w_disclaimer.pdf (Accessed 21 Oct 2015). 14. Perinatale Activiteiten in Vlaanderen 2013 H. Cammu, E. Martens, G. Martens, C. Van Mol, Y. Jacquemyn SPE, Brussel, 2014. www.zorg-en- gezondheid.be/uploadedFiles/NLsite_v2/Cijfers/Cijfers_over_geboorte_en_bevalling/SPE_jaarrapport %202013.pdf (Accessed 21 Oct 2015). 15. Van Leeuw V, Leroy Ch, Englert Y. Données périnatales en Région bruxelloise – Année 2013. Centre d’Épidémiologie Périnatale, 2015. www.cepip.be/pdf/rapport_CEPIP_Bxl2013_tma-FR.pdf (Accessed 21 Oct 2015). 16. Leroy Ch, Van Leeuw V, Englert Y. Données périnatales en Wallonie – Année 2013. Centre d’Épidémiologie Périnatale, 2015. http://www.cepip.be/pdf/rapport_CEPIP_wallonie2013_tmab.pdf (Accessed 21 Oct 2015). 17. Executive summary: Neonatal encephalopathy and neurologic Outcome, second Edition. Report of the American College of Obstetricians and Gynecologists’ Task force on Neonatal Encephalopathy. Obstet Gynecol 2014;123:896-901 doi: 10.1097/01.AOG.0000445580.65983.d2 [Published Online First April 2014]. 18. Bishop EH. Pelvic scoring for elective induction. J Obstet Gynecol 1964;24:266-8. 19. American College of Obstetricans and Gynecologists. Definition of term pregnancy. Committee Opinion No. 579. Obstet Gynecol 2013;122:1139-40 doi: 10.1097/01.AOG.0000437385.88715.4a [Published Online First: Nov 2013]. 20. Walsh CA, Baxi LV. Rupture of the primigravid uterus: a review of the literature. Obstet Gynecol Surv 2007;62:327-34. 21. Landon MB, Lynch CD. Optimal timing and mode of delivery after cesarean with previous classical incision of myomectomy: a review of the data. Semin Perinatol 2011;35:257-61 doi: 10.1053/j.semperi.2011.05.008 [Published Online First: Oct 2011]. 22. Nishijima Y, Suzuki T, Kashiwagi H et al. Uterine rupture at 26 weeks of pregnancy following laparoscopic salpingectomy with resection of interstitital portion: a case report. Tokai J Exp Clin Med 2014;39:167-71. 23. Inovay J, Marton T, Urbancsek J et al. Spontaneous bilateral cornual uterine dehiscence early in the second trimester after bilateral laparoscopic salpingectomy and in-vitro fertilization: a case report. Hum Reprod 1999;14:2471-3. 24. Sentilhes L, Sergent F, Roman H et al. Late complications of operative hysteroscopy: predicting patients at risk of uterine rupture during subsequent pregnancy. Eur J Obstet Gynecol Reprod Bio 2005;120: 134-8. 25. Zwart JJ, Van Huisseling HC, Schuttevaer HM et al. Nearly fatal uterine rupture during manual removal of the placenta: a case report. J Reprod Med 2007;52:974-6.

26. Misra M, Roychowdhury R, Sarkar NC et al. The spontaneous prelabour rupture of an unscarred uterus at 34 weeks of pregnancy. J Clin Diagn Res 2013;7:548-9 doi: 10.7860/JCDR/2013/4496.2820 [Published Online First: 1 Mar 2013]. 27. Langhoff-Roos J, Krebs L, Klungsøyr K et al. The Nordic medical birth registers – a potential goldmine for clinical research. Acta Abstet Gynecol Scand 2014;93:132-7 doi: 10.1111/aogs.12302 [Published Online First: Feb 2014]. 28. Ye J, Zhang J, Mikolajczyk R, Torloni MR et al. Association between rates of caesarean section and maternal and neonatal mortality in the 21st century: a worldwide population-based ecological study with longitudinal data. BJOG 2015; doi: 10.1111/1471-0528.13592 29. Nederlandse Vereniging voor Obstetrie & Gynaecologie. Richtlijnen. Zwangerschap en bevalling na een voorgaande sectio caesarea. Goedkeuring 2010-06-04. http://nvog-documenten.nl/ index.php?pagina=/richtlijn/pagina.php&fSelectTG_62=75&fSelectedSub=62&fSelectedParent=75 (Accessed 21 Oct 2015). 30. Sentilhes L, Vayssière C, Beucher G et al. Delivery for women with a previous cesarean: guidelines for clinical practice from the French College of Gynecologists and Obstetricians (CNGOF). Eur J Obstet Gynecol Reprod Biol 2013;170:25-32 doi: 10.1016/j.ejogrb.2013.05.015 [Published Online First: 28 Jun 2013]. 31. Kwee A, Bots ML, Visser GH et al. Obstetric management and outcome of pregnancy in women with a history of caesarean section in the Netherlands. Eur J Obstet Gynaecol Reprod Biol 2007; 132:171-6. 32. Knight HE, Gurol-Urganci I, van der Meulen JH et al. Vaginal birth after caesarean section: a cohort study investigating factors associated with its uptake and success. BJOG 2014;121:183–193 doi: 10.1111/1471-0528.12508. 33. Jozwiak M, Dodd JM. Methods of term labour induction for women with a previous caesarean section. Cochrane Database Syst Rev 2013;3:CD009792 doi: 10.1002/14651858.CD009792.pub2 [Published Online First:28 Mar 2013]. Footnotes

Contributors MH and YE conceived the project. MH, YE, GV, MDB and VVL designed the study. GV, MDB, VVL and MH collected the data. GV, MDB, KR and HV analysed the data. GV and HV wrote the first draft of the manuscript. MDB, KR and HV contributed to the writing of the manuscript. MDB, VVL, KR, HV, MH and YE critically revised the manuscript. GV, MDB, VVL, KR, MH, YE and HV approved the final version. Funding The study was funded by the College for Mother and Newborn, a consultative body of the Federal Public Service of Health in Belgium. GV was funded by the Research Foundation Flanders (FWO) (grant number FWO14/KDB/003). Competing interests None declared. Ethics approval The collection of patient information by B.OSS has been approved by the Medical Ethics Committee of the Ghent University Hospital (EC UZG 2012/734; B670201215359) and by the Medical Ethics Committee of the University Hospital Erasme, Brussels (EC ULB 2012/111; B406201213660). Data sharing statement No additional data are available.

NATIONWIDE POPULATION-BASED COHORT STUDY OF PERIPARTUM HYSTERECTOMY AND ARTERIAL EMBOLISATION IN BELGIUM: RESULTS FROM THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM Griet Vandenberghe1, Marine Guisset2, Iris Janssens3, Virginie Van Leeuw4, Kristien Roelens1, Myriam Hanssens2, Erika Russo6, Joachim Van Keirsbilck7, Yvon Englert4,8, Hans Verstraelen1

1. Department of Obstetrics and Gynaecology, Ghent University Hospital, Ghent, Belgium 2. Department of Obstetrics and Gynaecology, Leuven University Hospital, Leuven, Belgium 3. Faculty of Medicine, Ghent University, Ghent, Belgium 4. Perinatal Epidemiology Center (Centre d'Épidémiologie Périnatale, CEpiP), School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium 5. Université Libre de Bruxelles (ULB), School of Public Health, Brussels, Belgium 6. Department of Obstetrics and Gynaecology, Intercommunale de Santé Publique du Pays de Charleroi (ISPPC), Hôpital Civil Marie Curie, Charleroi, Belgium 7. Department of Obstetrics and Gynaecology, St Jan’s Hospital, , Belgium 8. Faculty of Medicine, Research Laboratory on Human Reproduction, Université Libre de Bruxelles (ULB), Brussels, Belgium

Abstract

Objectives To assess the prevalence of major obstetric haemorrhage managed with peripartum hysterectomy and/or interventional radiology (IR) in Belgium. To describe women characteristics, the circumstances in which the interventions took place, the management of the obstetric haemorrhage, the outcome and additional morbidity of these women. Design Nationwide population-based prospective cohort study. Setting Emergency obstetric care. Participation of 97% of maternities covering 98.6% of deliveries in Belgium. Participants All women who underwent peripartum hysterectomy and/or IR procedures in Belgium between January 2012 and December 2013. Results We obtained data on 166 women who underwent peripartum hysterectomy (n=84) and/or IR procedures (n=102), corresponding to 1 in 3030 women undergoing a peripartum hysterectomy and another 1 in 3030 women being managed by IR, thereby preserving the uterus. Seventeen women underwent hysterectomy following IR and three women needed further IR despite hysterectomy. Abnormal placentation and/or uterine atony were the reported causes of haemorrhage in 83.7%. Abnormally invasive placenta was not detected antenatally in 34% of cases. The interventions were planned in 15 women. Three women were transferred antenatally and 17 women postnatally to a hospital with emergency IR service. Urgent peripartum hysterectomy was averted in 72% of the women who were transferred, with no significant difference in need for transfusion. IR procedures were able to stop the bleeding in 87.8% of the attempts. Disseminated intravascular coagulation secondary to major haemorrhage was reported in 32 women (19%). Conclusion The prevalence in Belgium of major obstetric haemorrhage requiring peripartum hysterectomy and/or IR is estimated at 6.6 (95% CI 5.7 to 7.7) per 10 000 deliveries. Increased clinician awareness of the risk factors of abnormal placentation could further improve the management and outcome of major obstetric haemorrhage. A case-by-case in-depth analysis is necessary to reveal whether the hysterectomies and arterial embolisations performed in this study were appropriate or preventable.

Background

Emergency peripartum hysterectomy is the single most dramatic procedure in modern obstetrics. The procedure is stressful and surgically challenging, inevitably causing additional maternal morbidity and, not least, infertility. It is listed by the WHO as an identification criterion for maternal near-miss,1 a concept that, in view of the very low maternal mortality rates in developed country settings,2 was introduced as an analytical tool to address health system failures with the overall goal of improving obstetric care.3Substandard care in the management of major obstetric haemorrhage continues to be a critical cause of maternal death and severe maternal morbidity in developed countries.4 5 Every obstetric practitioner needs to be trained in the management of postpartum haemorrhage in order to guarantee an immediate response and a multidisciplinary team approach. Internationally recognised guidelines6–8 indicate that one or more second-line measures, including intrauterine (balloon) tamponade, haemostatic brace suturing, ligation of the uterine arteries and interventional radiology (IR), should be available in hospitals with delivery units and that obstetric practitioners should be familiar with these procedures. Notably, the early involvement of an experienced senior obstetrician, hence appropriate clinical judgement, can save lives: the timely decision to turn to a hysterectomy can avoid further blood loss, risk of disseminated intravascular coagulation (DIC), additional morbidity and finally death.6 While emergency peripartum hysterectomy serves as the ultimate response to major obstetric haemorrhage, resorting to IR procedures on the pathway towards hysterectomy can be both life-saving and fertility-preserving. IR in this context basically involves intra-arterial balloon occlusion or arterial embolisation of the uterine or the internal iliac arteries. Research publications on the effectiveness of IR procedures in obstetric settings reporting high success rates of 85%–95% are accumulating.9–12 Success rates should be interpreted with caution, however, as they may be driven by ascertainment bias. IR is cautiously suggested in guidelines and recommendations6–8 13 14 describing its possible role in obstetric emergency settings, but further evidence is awaited before firm recommendations can be made. Yet in the UK a national recommendation was made to provide a network of emergency IR services available to all trusts with delivery units in response to a Healthcare Commission Investigation into maternal deaths.13 15 The aims of this study are to investigate the population-based prevalence of peripartum hysterectomy and the indications to proceed to this intervention, likewise the prevalence of major obstetric haemorrhage managed with IR in Belgium. Furthermore, we aim to describe the characteristics of the women in Belgium who required these interventions in order to treat or prevent major obstetric haemorrhage, the circumstances in which the interventions took place, the management of the obstetric haemorrhage and the outcome and additional morbidity of these women.

Method

Peripartum hysterectomy and IR in Belgium were registered as part of a national reporting system for the study of rare obstetric complications. The methodology of the Belgian Obstetric Surveillance System (B.OSS) has been described previously.16 Briefly, an appointed contact person (obstetrician or senior midwife) in each participating maternity unit was invited by monthly mailing to report a selected number of rare obstetric complications that had occurred in the preceding month or, alternatively, to state that there was ‘nothing to report’. In the event of such a case being reported, the contact person was asked to complete an extensive data collection form. Confidentiality was guaranteed for mother, provider and hospital, and person-identifiable information was eliminated from data analysis. In cases of incomplete reporting, the appointed contact person was encouraged repeatedly by email and phone to provide the missing data, up to 6 months following the date of the original report. The participation of 97% of Belgian maternity units (n=110/113) was obtained, covering 98.6% of deliveries (n=248 681) in the 2-year study period. Belgium has a relatively large number of tertiary referral centres providing neonatal intensive care (n=19) along with a high concentration of small-volume maternity units (more than half of all units) in which there are fewer than 1000 annual births. All tertiary referral centres, as well as some of the larger non-tertiary maternity services, have an emergency IR service. In answer to an enquiry of B.OSS contact persons, 40 maternity units (38%) declared they had an IR service available in the hospital, and a further seven units declared that they had an IR service available but not 24/7 or that they could call on vascular surgeons for similar procedures.

The study was conducted between January 2012 and December 2013.

B.OSS aimed to collect all cases of peripartum hysterectomy and IR at any gestational age corresponding to the following definition: any woman giving birth to a fetus or infant and undergoing a hysterectomy and/or IR procedure in the same clinical episode. Most peripartum hysterectomies are emergency, life-saving interventions following vaginal or caesarean delivery, most commonly performed to manage major obstetric haemorrhage but occasionally because of a severe pelvic infection. Hysterectomy in the first or second trimester, termed abortion hysterectomy, is occasionally required to prevent or manage major obstetric haemorrhage resulting from an abnormally invasive placenta (AIP) or an ectopic pregnancy in the cornua, the cervix or a caesarean scar. A peripartum hysterectomy during caesarean delivery (CD), termed ‘caesarean hysterectomy’, can be performed electively following antenatal diagnosis of a severe AIP or a gynaecological cancer. Failure of IR was defined as the need for urgent peripartum hysterectomy to control the haemorrhage.

Data collection forms sought information on maternal characteristics, medical, surgical and obstetric history, details of the index pregnancy, details of the delivery, the circumstances of the adverse event, its management, and the outcome for mother and neonate. The cause of the obstetric haemorrhage was deduced from the clinical signs reported by the clinician and classified according to the 4T’s mnemonic: tonus, tissue, trauma and thrombin. These labels were taken to stand for the following characteristics: tonus: uterine atony; tissue: AIP, placenta praevia, combination of AIP and placenta praevia, placental remnants or retained placenta without AIP; trauma: uterine rupture, genital tract lacerations, bleeding caused by unintended injuries during caesarean delivery or curettage (uncomplicated CDs were not included in this category); and thrombin: abnormal coagulation before onset of bleeding (DIC secondary to major haemorrhage was not included in this category). Cases where the cause of bleeding remained unclear or unknown were classified as ‘Other’.

The Belgian perinatal registries — Studiecentrum voor Perinatale Epidemiologie (SPE) in Flanders,17 18and Centre d’Epidémiologie Périnatale in Brussels19 and Wallonia20 — record data on a mandatory basis, covering nearly 100% of births in Belgium. A selected set of perinatal data are recorded by the obstetrician, midwife and neonatologist immediately after birth. Births are registered as such in cases of a live birth or in cases of a stillbirth at a birth weight of 500 g or more and/or after 22 completed weeks of gestation, although in Flanders (SPE) live births or stillbirths with a birth weight below 500 g, irrespective of gestational age, are not registered.

The aim was to record all cases of peripartum hysterectomy and arterial embolisation during a 2-year study period, meaning that no fixed sample size was set at start of the study. The prevalence of the obstetric events targeted was estimated using as denominator the total number of deliveries in Belgium in 2012 and 2013, corrected for the three hospitals that did not participate in B.OSS (n=248 681 women).

Comparison of the distributions of sociodemographic and obstetric characteristics of reported cases relative to the background population as obtained from the Belgian perinatal registry (2012–2013) was pursued through use of ORs. ORs and accompanying 95% CI were calculated in univariate analysis, using weighted cases where appropriate. Unadjusted and adjusted ORs were calculated in order to compare patient and organisational characteristics between interventions (hysterectomy vs IR) in univariate analysis and subsequently through the use of unconditional binary logistic regression models. Comparison between successful and failed IR cases was made only in crude analyses, as numbers of observations were too small to support logistic regression models. Non-parametric tests were used to compare distributions of outcomes between groups, specifically χ2 for categorical variables, and Kruskal-Wallis and Mann-Whitney U tests for continuous variables. p Values of <0.05 were considered statistically significant. Data were analysed using the statistical software package IBM SPSS V.22.0 statistics.21 The Strengthening the Reporting of Observational Studies in Epidemiology guidelines for reporting observational studies were followed.

Results

Data collection forms of 166 confirmed cases of peripartum hysterectomy and/or IR were retrieved and included in the analysis. Details of reported cases and completeness of data collection forms are presented in figure 1.

Overall, 84 women (50.6%) underwent a peripartum hysterectomy, corresponding to a prevalence of 3.3 per 10 000 deliveries (95% CI 2.7 to 4.1). Of these, 17 women (10.2%) underwent a hysterectomy following a previous IR procedure. Eighty-two women (49.4%) were managed with IR only, corresponding to a prevalence of 3.3 per 10 000 deliveries (95% CI 2.6 to 4.0). Three women (1.8%) needed IR because of persistent bleeding despite hysterectomy. In 160 women the intervention occurred within 24 hours of delivery. In eight women it occurred later (from 24 hours to 6 weeks postpartum). The discrepancy between the resultant sum of 168 and the 166 figure above is because two of the data collection forms were not returned, one of these being the only case in this data set of a hysterectomy performed because of severe peritonitis. All other women underwent hysterectomy and/or IR to treat or prevent major obstetric haemorrhage. The prevalence in Belgium of major obstetric haemorrhage requiring peripartum hysterectomy and/or IR is estimated at 6.6 (95% CI 5.7 to 7.7) per 10 000 deliveries.

Figure 1. Flow chart of case reporting and data collection.

We compared the women included in this registry with all the women who gave birth in Belgium during the study period. Patient characteristics of the background population were derived from the Belgian perinatal registries. In univariate analysis the women who underwent peripartum hysterectomy and/or IR for major obstetric haemorrhage were significantly more likely compared with the background population to be older (≥35 years), to be multiparous (parity ≥3), to have given birth at a preterm gestational age, to have had a twin pregnancy, to have had a CD in a previous pregnancy, to have had a CD in the index pregnancy and to have conceived with artificial reproductive technology (table 1). Ten of these 166 women were known to have a uterus myomatosus (6.0%) and 12 experienced haemorrhage in a previous pregnancy (7.2%).

Table 1. Patient characteristics of women with (imminent) obstetric haemorrhage who underwent peripartum hysterectomy and/or IR compared to all women who gave birth in Belgium during the study period. Risk factor Cases of H and/or IR Background population # Unadjusted OR % (n=166) (n=252,272) for H and/or IR n, % n, % (95% CI) Maternal age ≥ 35 years 72 (43.4) 43,256 (17.1) 3.7 (2.7-5.0) §

BMI at booking ≥ 30 kg/m2 12 (7.2) £ 29,453 (11.6) 0.7 (0.3-1.2) Singletons ≥22 weeks Singletons ≥22 weeks Gestational age (weeks) (n= 144) (n= 241,136) - Extreme preterm (<28w) 8 (5.4) 1,221 (0.5) - Very preterm (28-31w6) 6 (4.0) 1,474 (0.6) 7.4 (5.3-10.5) § - Late preterm (32-36w6) 37 (24.8) 13,798 (5.7) - Term (>37) 93 (62.4) 224,643 (93.1) Reference Parity ≥ 3 30 (18.0) 18,855 (7.4) 2.7 (1.8-4.1) §

Multiplets (twin) 13 (7.8) 4,690 (1.8) 4.4 (2.5-7.9) §

Previous CD & 61 (36.7) £ 27,007 (10.7) 4.8 (3.5-6.6) §

TOLAC (≥22 weeks) 14 / 57 (24.5) 12,754 / 27,007 (47.2) 0.3 (0.1-0.66) §

Delivery Mode (≥22 weeks) - CD $ 91 / 160 (56.8) £ 54,369 (21.5) 4.8 (3.5-6.5) § compared to VD - CD before onset of labour 66 / 91 (72.5) 28,586 / 54,369 (52.5) 2.3 (1.5-3.7) compared to CD after onset of labour - Assisted VD compared 10 / 69 (14.4) 23,100 /198,444 (11.6) 1.3 (0.6-2.5) to spontaneous VD ART (IVF/ICSI) 19 (11.4) £ 9,233 (3.7) 3.4 (2.1-5.4) §

Birthweight ≥ 4000 g 11 (6.6) 19,967 (7.8) 0.8 (0.4-1.5)

CD: Caesarean delivery, VD: Vaginal delivery, H: Hysterectomy, IR: Interventional Radiology, TOLAC: Trial of Labour After Caesarean Delivery, ART: Artificial Reproductive Technologies, IVF: In Vitro Fertilisation, ICSI: Intracytoplasmic Sperm Injection # Data from the background population were retrieved from SPE reports (Flanders) and CEpiP reports (Brussels and Wallonia) from 2012-2013. % unadjusted Odds ratios (with 95% Confidence Interval) for peripartum hysterectomy and/or IR calculated for the different risk factors: the odds (=probability of peripartum hysterectomy and/or IR occurring divided by the probability of H/IR not occurring) in the women having the risk factor, divided by the odds in the women not having the risk factor. £ Missing data in cases of hysterectomy and/or IR : BMI (n=27), previous CD (n=1), mode of delivery (n=1), ART (n=2), haemorrhage in previous pregnancy (n=5), uterus myomatosus (n=5) & Women with previous Caesarean Delivery: 9 with Placenta Praevia, 9 with AIP, 17 with Placenta Praevia and AIP $Women with Caesarean Delivery as delivery mode: 17 with Placenta Praevia, 8 with AIP, 23 with Placenta Praevia and AIP Reason for Caesarean Delivery: Bleeding (Placenta Praevia, AIP, abruptio placentae): n= 15; Placenta Praevia / AIP not bleeding: n= 29; Maternal shock / cardiac arrest: n=1; Other: n = 46 § p < 0.05

IR service availability, antenatal and postnatal transfers, and urgent and planned interventions The circumstances in which the interventions occurred in these 166 women are shown in figure 2. Three women were transferred antenatally to another hospital with emergency IR service availability, a total of 107 women gave birth in a maternity unit with this availability, while 20 gave birth in a unit where IR service is available but not 24/7 or where obstetricians can call on vascular surgeons for IR procedures, and 39 in a unit with no IR service. To refer to these different circumstances below, we use the terms 24/7-IR unit, not 24/7-IR unit and non-IR unit, respectively.

Figure 2. Circumstances in which peripartum hysterectomy and interventional radiology occurred in 166 women.

Seventeen women were transferred postnatally to a 24/7-IR unit and underwent an urgent IR procedure, including one woman for persistent bleeding despite hysterectomy and one woman for persistent bleeding despite planned intra-arterial balloon occlusion and arterial embolisation by the vascular surgeons in the referring hospital. Three of the postnatally transferred women consecutively underwent an urgent hysterectomy for persistent bleeding despite arterial embolisation. Of the three patients who were transferred antenatally, one underwent an urgent embolisation on arrival because of a bleeding placenta praevia, one was first planned for caesarean hysterectomy but then underwent this procedure urgently because of a bleeding AIP before the fixed date, and one underwent a planned CD combined with a planned IR procedure because of AIP. Seven more women were transferred postnatally to the intensive care unit (ICU) of a tertiary referral centre. Overall, intervention for imminent obstetric haemorrhage was planned in 15 women. The indication was AIP in 13 women and placenta praevia in 2 women.

Reported cause of the obstetric haemorrhage Eighteen women (11%) presented with antepartum haemorrhage due to abnormal placentation (placenta praevia, AIP or both) or abruptio placentae (n=2). Four women (2%) started bleeding during termination of pregnancy (<24 weeks) due to a cervical pregnancy (n=1) or an AIP (n=3). One hundred and thirty-eight women (83%) presented with early postpartum haemorrhage (within 24 hours) and six (4%) with late postpartum haemorrhage (from 24 hours to 6 weeks). The reported cause of the obstetric haemorrhage was uterine atony (n=91, 54.8%) or AIP (n=38, 22.8%) in the majority of women. In 54 women (32.5%) more than one cause was reported, with atony and abnormal placentation (AIP and/or placenta praevia) in 20 women (12%) as the most frequent combination. The cause of the (imminent) obstetric haemorrhage classified according to the 4T’s mnemonic is shown in figure 3.

Figure 3. Cause of obstetric haemorrhage according to the 4T’s mnemonic.

Hysterectomy was performed in all the cases of haemorrhage reported to be associated with uterine rupture (n=12) and in 73.7% of the cases associated with AIP (n=28/38). IR was able to stop the bleeding and avert the need for peripartum hysterectomy in the majority of cases of haemorrhage reported to be associated with genital tract lacerations (n=10/12, 83.3%), placental remnants or retention (n=17/26, 65.4%) and uterine atony (n=28/43, 65.1%). Other measures that were attempted to manage the obstetric haemorrhage before or while proceeding to the hysterectomy or IR are listed in table 2. The first-line measures were the administration of uterotonic agents and bimanual uterine compression, second-line measures being intra-uterine (balloon) tamponade, haemostatic brace suturing (B-lynch or other) and ligation of the uterine or internal iliac arteries, besides blood transfusion and the administration of fresh frozen plasma (FFP), platelets or clotting factors. An intra-uterine balloon was inserted in 35 women (21%), in 7 cases before transfer to another hospital with availability of an IR service. Brace suturing and arterial ligation were attempted in 6 and 4 women, respectively. In 19 women (11.4%) no other measures were undertaken besides hysterectomy (n=13), IR (n=2) or hysterectomy following IR (n=4). The reported cause of haemorrhage in these women was AIP (n=12), placenta praevia (n=2), uterine rupture (n=2), retained placenta (n=2) and in one case a late postpartum haemorrhage. The intervention was planned in five of these women. Eight of the 19 had minor haemorrhage and did not require transfusion, while three women undergoing urgent peripartum hysterectomy needed massive transfusion (defined as receiving eight or more units of red blood cells (RBCs)). Table 2. First and second line measures in women undergoing hysterectomy (H) and /or IR (IR) according to the circumstances of the obstetric haemorrhage. Total Planned Urgent H and/or IR H n=151 and/or IR IR service No IR Postnatal IR service not 24/7 service Transfer

24/7

n=166 n=15 n=98 n=12 n=25 n=16* n (%) n (%) n (%) n (%) n (%) n (%) Cause of obstetric haemorrhage Abnormal 2 (13) 56 (34) 15 (100) 29 (30) 5 (42) 5 (20) placentation Atony only 43 (26) - 29 (30) 5 (42) 7 (28) 2 (13) Uterine rupture 12 (7) - 6 (6) 1 (8) 5 (20) - Miscellaneous£ 55 (33) - 34 (34) 1 (8) 8 (32) 12 (75) First and second line measure Uterotonics 134 (81) 10 (67) 81 (83) 10 (83) 19 (76) 15 (94) Oxytocin 123 (74) 10 (67) 73 (75) 9 (75) 18 (72) 13 (81) PG 109 (66) 5 (33) 64 (65) 17 (68) 8 (67) 15 (94) Misoprostol 61 (37) 1 (7) 34 (35) 6 (50) 10 (40) 10 (63) Carboprost 63 (39) 5 (33) 42 (43) 3 (25) 9 (25) 9 (56) Oversewing# 30 (18) 1 (7) 19 (20) 1 (8) 7 (28) 2 (13) (Bimanual) 64 (39) 3 (20) 36 (37) 5 (42) 9 (36) 11 (69) compression Balloon 35 (21) 2 (13) 21 (22) 2 (17) 3 (12) 7 (44) tamponade B-lynch or 10 (6) - 5 (5) 1 (8) 4 (16) - ligation Transfusion 139 (84) 7 (47) 85 (87) 12 (100) 21 (84) 15 (88) Massive 51 (31) 2 (13) 24 (24) 7 (58) 10 (40) 8 (50) transfusion$ FFP 108 (65) 4 (27) 64 (65) 11 (92) 18 (72) 11 (69) Platelets 57 (34) 1 (7) 29 (30) 9 (75) 10 (40) 8 (50) Clotting factors% 23 (14) - 8 (8) 5 (42) 7 (28) 3 (19) * One woman was transferred postnatally for persistent bleeding due to AIP despite planned IR in the referring hospital; she was included in the ‘planned’ group. £ Miscellaneous group: Placental remnants or retention n=25, Surgical trauma n=16, Genital tract laceration n=11, Intra-abdominal arterial bleeding n=4, Coagulation disorders n= 3, Unknown n= 6. The total number exceeds n = 55 due to overlap in between cases. PG: Prostaglandin. This can be any or a combination of the following: misoprostol, sulprostone, dinoproston, carboprost. # Oversewing of the placental implantation site, bleeding points, tears or ruptures. $ Massive transfusion: transfusion of 8 or more units of red blood cells % Clotting: Fibrinogen, recombinant Factor VIIa, PPSB (Prothrombin-Proconvertin-Stuart Factor- Antihemophilic Factor B) or factors a combination

The management of the (imminent) obstetric haemorrhage in the women with AIP (n=38) is illustrated in figure 4 (for details see online supplementary figure 1). Seventeen women (45%) were susceptible to AIP due to previous CD and placenta praevia in the index pregnancy, another seven women (18%) had placenta praevia in the index pregnancy and nine women (24%) had previous CD. In 25 women (65.7%) there was antenatal suspicion of the AIP, of whom 16 had placenta praevia and previous CD, 5 had placenta praevia and 4 had previous CD only. All women with antenatal suspicion of AIP were planned to undergo an elective CD in a maternity unit with IR service availability (24/7 n=20, not 24/7 n=5). This occurred in 21 women, with 3 women having an emergency CD because of antepartum haemorrhage and 1 woman delivering vaginally following acute onset of labour. Further preventive measures were undertaken in 13 women (31.5%): antenatal transfer to a centre with IR service availability (n=2), placement of intra-arterial catheters precaesarean delivery (n=12), planned caesarean hysterectomy (n=3) or planned hysterectomy at a later stage (n=2).

In 13 women (34%) there was no antenatal suspicion of the AIP. In retrospect, though, all of them had risk factors: one woman had placenta praevia and previous CD, four had placenta praevia, six had previous CD, one had previous myomectomy and adhesiolysis for Asherman syndrome and another had a previous manual removal of the placenta. Nine women delivered in a maternity unit with IR service availability (24/7 n=7, not 24/7 n=2) and four delivered in a non-IR unit. The mode of delivery in those 13 women who had no antenatal suspicion of AIP was elective CD in two women, emergency CD in five women and vaginal delivery in six women. The placenta was left in situ in 11 women with AIP: four in an emergency setting where three women underwent immediate hysterectomy and one woman underwent urgent arterial embolisation followed by urgent hysterectomy because of bleeding after 3 days. Another seven women had their placenta left in situ electively combined with planned IR: two women consecutively underwent planned hysterectomy, two an urgent hysterectomy within 24 hours, two a planned hysterectomy after 1 month, and one elective evacuation of retained products of conception after 1 month. In conclusion, conservative management, defined as aiming to preserve the placenta and the uterus in a multidisciplinary approach, was attempted in six women and was successful in three women.

As anticipated, the women for whom measures were planned antenatally had significantly better outcomes than those where AIP was diagnosed peripartum: less estimated blood loss (median 625 mL (range 300–1250) vs median 4500 mL (range 500–6000), p=0.08), less need for transfusion (46.2% vs 92.3%, p=0.056), less need for massive transfusion (≥8 units) (8.3% vs 53.8%, p=0.045) and fewer hysterectomies (61.5% vs 76.9%, p=0.6). The maternal outcome was not significantly different between women with AIP undergoing IR as first intervention and those with hysterectomy as first intervention: estimated blood loss (median 4000 mL (range 500– 15000) vs median 1000 mL (range 300–8500), p=0.3), need for transfusion (76.5% vs 70%, p=0.9) and need for massive transfusion (≥8 units) (29.4% vs 30%, p=1.0). All women with planned interventions were managed in tertiary referral centres.

The decision of an obstetrician dealing with a major obstetric haemorrhage to attempt IR rather than turning to a hysterectomy immediately will be guided by many clinical and organisational factors. Some of the risk factors that may have contributed to this decision are presented in table 3. The women with double interventions (n=20) were excluded from this analysis to facilitate interpretation. High parity (≥3), previous CD, CD in the index pregnancy, AIP and lack of emergency IR service availability are significantly positively associated with peripartum hysterectomy in univariate analysis. In unconditional binary logistic regression, the odds of undergoing a hysterectomy for obstetric haemorrhage are 7.2 times higher in women with high parity and 8.2 times higher in women delivering in a non-IR unit, independent of the other risk factors in the model (Hosmer and Lemeshow test p=0.67). We repeated the analysis in a subgroup, namely the women who delivered in a 24/7- IR unit, with exclusion of the women with planned interventions, consisting of 31 women who underwent hysterectomy only and 60 women who underwent embolisation only. In unconditional binary logistic regression, the odds of undergoing a peripartum hysterectomy for obstetric haemorrhage are significantly higher in women with high parity (odds 4.5 (95% CI 1.2 to 16.5)) and nearly significantly higher in women with previous CD (odds 3.7 (95% CI 0.98 to 14.6)), independent of the other risk factors in the model (Hosmer and Lemeshow test p=0.69).

Table 3. Risk factors possibly contributing to the choice of intervention in women undergoing hysterectomy or interventional radiology for obstetric haemorrhage. Unadjusted and adjusted OR (95% CI) for hysterectomy (only) per risk factor. Risk factors Number of women Unadjusted Adjusted OR Hysterectomy Interventional OR (95%CI) (95%CI) only radiology for for n=64 only Hysterectomy Hysterectomy n, % n=82 only only n, % Demographic factors Maternal age ≥ 35 years 32 (50.0) 30 (36.5) 1.7 (0.8-3.3) BMI at booking ≥ 30 kg/m2 6 (9.3) 3 (3.6) 2.7 (0.6-11.3) Obstetric factors Preterm (<37w) 28 (43.8) 24 (29.3) 1.8 (0.9-3.7) Parity ≥ 3 18 (28.1) 7 (8.5) 2.7 (1.4-9.4) § 7.2 (2.1-24.6) § Previous caesarean delivery 31 (48.4) 18 (21.9) 3.3 (1.6-6.8) § 2.3 (0.9-6.3) Caesarean delivery ( ≥22w) * 38 / 59 (64.4) 37 / 82 (45.1) 2.2 (1.1-4.3) § 1.6 (0.6-4.0) Cause of bleeding AIP and AIP+ praevia 16 (25.0) 9 (10.9) 2.7 (1.1-6.6) § 3.0 (0.9-10.0) Atony 29 (45.3) 52 (63.4) 0.4 (0.2-0.9) § 0.9 (0.3-2.3) Organisational factors Weekend and/or non-office 24/60 (40.0) 34/80 (42.5) 0.9 (0.4-1.7) hours # IR NOT available $ 30 (46.8) 15 (18.2) 3.9 (1.8-8.2) § 8.2 (3.1-21.6) § * Exclusion of 5 cases < 22 weeks £ Missing data: BMI (n=27), previous caesarean delivery (n=1), Delivery Mode (n=1), placenta location (n=13), Date of delivery (n=2), Time of delivery (n=5), week/weekend (n=3) # based on date and time of delivery IR: interventional radiology $ Emergency IR service availability in the maternity unit where the woman gave birth § p<0.05

Seventeen women in this data set had a peripartum hysterectomy performed following a previous IR procedure. In five of these women the hysterectomy was performed in a programmed setting following planned IR because of AIP, two of them with a delay of 1 month. These five IR procedures were considered successful in preventing or stopping the bleeding. In the other 12 women, an urgent hysterectomy was performed because of persistent bleeding despite the IR procedure. We compared the characteristics of these 12 women for whom IR failed with 87 women (including the 5 who underwent an elective hysterectomy) for whom IR was successful. In univariate analysis, an increased risk of urgent hysterectomy despite IR was observed for women with CD in a previous pregnancy, for women with AIP and for women who had no uterotonic agents administered. These findings are presented in table 4.

Table 4. Risk factors possibly contributing to failure of Interventional Radiology (IR). Unadjusted odds ratio’s (95% CI) for failed IR per risk factor. Riskfactors Number of women Unadjusted OR Failed IR Successful IR (95% CI) n=12 n=87 for failed IR n, % n, %

Demographic factors Maternal age ≥ 35 years 5 (41.7) 33 (37.9) 1.1 (0.3-3.9) BMI at booking ≥ 30 kg/m2 2 (18.2) 3 (3.9) 5.6 (0.8-37.6) Obstetric factors

Twins 1 (8.3) 6 ( 6.9) 1.2 (0.1-11.1) Preterm gestational age < 37w 7 (58.3) 28 (32.2) 2.9 (0.8-10.1) Preterm gestational age < 34w 3 (25.0) 6 (6.8) 4.5 (0.9-21.1) Parity ≥3 3 (25.0) 8 (9.2) 3.2 (0.7-14.6) Previous caesarean delivery 7 (58.3) 21 (24.1) 4.4 (1.2-15.3) § Caesarean delivery (≥22 weeks) 8 (72.7) 42 (48.3) 2.1 (0.6-7.6) Induction of labour 3 (25.0) 25 (28.7) 0.8 (0.2-3.3) Augmentation of labour 2 (16.7) 26 (29.9) 0. (0.09-2.2) Cause of haemorrhage Atony 7 (63.6) 52 (62.7) 0.9 (0.2-3.2) AIP and AIP+ praevia 7 (58.3) 14 (16.1) 7.3 (2.0-26.3) § Surgical trauma 2 (16.7) 6 (6.9) 2.7 (0.4-15.2) Genital tract laceration 1 (8.3) 10 (11.5) 0.7 (0.08-6.0) 1st and 2nd line treatment Uterotonic agents 8 (66.7) 82 (94.3) 0.1 (0.02-0.5) § No use of uterotonic agents 4 (33.3) 5 (5.7) 8.2 (1.8-36.8) § Intra-uterine balloon 3 (25.0) 26 (29.9) 0.7 (0.1-3.1) Clotting factors IV 2 (16.7) 8 (9.2) 1.9 (0.3-10.6) FFP 7 (58.3) 54 (62.1) 0.8 (0.2-2.9) Organisational factors Planned intervention 2 (14) 10 (12) 1.5 (0.3-8.0) Weekend and/or non-office hours# 5 / 10 (50.0) 51 / 85 (60.0) 0.6 (0.1-2.4) IR NOT available $ 3 (25.0) 10 (11.4) 2.5 (0.6-11.1) Emergency IR NOT available $ 6 (50.0) 20 (22.9) 3.3 (0.9-11.5) Postnatal transfer 3 (25.0) 13 (14.9) 1.9 (0.4-7.9) IR: interventional radiology; AIP: abnormally invasive placenta; FFP: Fresh Frozen Plasma ; CD: Caesarean delivery # based on date and time of delivery $ IR service availability (24/7 and not 24/7) in the maternity unit where the woman gave birth: $ Emergency IR service availability (24/7) in the maternity unit where the woman gave birth § p < 0.05 Blood transfusion was given to 139 women (83.7%, missing data for 2 women), with a median of 6.0 units of RBCs (range 0–31), 4.0 units of FFP (range 0–31) and 0.5 units of platelets (range 0–29). Fifty-one women (30.7%) were transfused ≥8 units of RBC and 23 women were administered clotting factors (fibrinogen, recombinant factor VIIa, prothrombin-proconvertin-Stuart factor-antihaemophilic factor B or a combination). One hundred and fourteen women (68.6%, missing data in 4 women) were admitted to an ICU for a median of 2.0 days (range 0–18 days); the median hospital stay was 8.0 days (range 2–33 days). Five women needed relaparotomy and three women needed arterial embolisation following hysterectomy because of persistent bleeding. The cause of haemorrhage in these women was AIP (n=1), AIP and atony (n=2), atony (n=3) and in 1 case unclear (n=1). DIC secondary to major haemorrhage was reported in 32 women (19%). Other reported maternal complications associated with major haemorrhage were renal failure (n=2), pulmonary oedema (n=4), acute respiratory distress syndrome (n=2), myocardial ischaemia (n=1) and pleural and pericardial effusion (n=1). Complications associated with peripartum hysterectomy were ureteral injuries (n=2) and ileus (n=4). One woman suffered from pain in the loins and legs, which was the only reported complication attributed to the IR procedure. One mother died (case fatality rate 0.6% (95% CI 0.1 to 3.3)). She underwent an emergency CD because of cardiac arrest caused by massive pulmonary embolism. She survived the hysterectomy performed because of major bleeding due to uterine atony and was transferred to the ICU of a tertiary referral centre using extracorporeal life support. She underwent a relaparotomy because of major hepatic haemorrhage, but died the subsequent day during pulmonary thrombectomy. Considering the whole group, the outcome in terms of transfusion rate and ICU admission in women who underwent hysterectomy did not significantly differ from those who underwent IR, except for a significant difference in total duration of hospitalisation (median 9 days, range (2–30) vs median 7 days (range 2–33)). When comparing the women according to the IR service availability in the maternity unit where they gave birth, the rate of peripartum hysterectomy was significantly higher in the group of women who delivered in a non-IR unit (29 women (74%)) than in the group who delivered in a 24/7-IR unit (44 women, 41%), and so was the rate of massive transfusion (defined as ≥8 units of RBC): 17 women (45%) who delivered in a non-IR unit compared with 25 women (24%) who delivered in a 24/7-IR unit. When comparing the women who gave birth in a non-IR unit according to whether they were transferred postpartum, there was no significant difference in estimated blood loss (median 2850 mL (range 500–5000) vs median 3500 mL (range 3000–5000), p=0.4) and need for transfusion (median number of units 6.5 (range 0–19) vs 9.5 (range 0–31), p=0.2), but all women who were not transferred (n=25/25) underwent a hysterectomy compared with 28% (n=4/14) in the transferred group (p=0.0).

Discussion

The main findings of this study are, first, that the prevalence in Belgium of major obstetric haemorrhage requiring peripartum hysterectomy and/or IR is estimated at 6.6 (95% CI 5.7 to 7.7) per 10 000 deliveries; second, that AIP was not detected antenatally in 34% of the cases in this study, despite the presence of risk factors; third, that women with major obstetric haemorrhage who gave birth in a non-IR unit were more likely to undergo an urgent peripartum hysterectomy (74%) than those who delivered in a 24/7-IR unit (41%); and fourth, that the need for urgent peripartum hysterectomy was averted in 72% of the women who delivered in a non- IR unit but transferred postnatally, with no significant difference in estimated blood loss and transfusion rate compared with the women who were not transferred. The strengths of the study are its methodology of collecting cases prospectively on a monthly basis and its national design, with excellent participation and response rates (98.9%) of Belgian maternity units. Under-reporting of cases, due to the non- mandatory reporting of severe pregnancy complications, may have biased our results to some extent. The small number of cases, inevitable with rare but severe obstetric complications, warrants caution in making strong inferences. Another limitation is the lack of a perspective on the overall picture of postpartum haemorrhage in Belgium. We have no accurate data of the total number of women suffering severe postpartum haemorrhage, the near-misses and maternal deaths due to severe haemorrhage who did not undergo embolisation or hysterectomy. We have no denominator data of the women who were successfully managed with other second-line measures (intra- uterine (balloon) tamponade, haemostatic brace suturing and ligation of the uterine arteries) during the same study period in Belgium, so we cannot draw conclusions on the failure rate of these second-line measures. These are interesting topics that can be investigated by B.OSS in future studies. This nationwide population-based study describes how women with major obstetric haemorrhage were managed with either hysterectomy or IR according to circumstances. However, this study does not allow firm conclusions to be drawn regarding the motivation or the appropriateness of this management. One should be aware that obstetricians are forced to make rapid decisions when dealing with a major obstetric haemorrhage. Belgian obstetricians cannot rely on national guidelines for postpartum haemorrhage or AIP. Small volume maternity units even lack local hospital guidelines for these matters. Therefore, the management of major obstetric haemorrhage will differ greatly between hospitals and individual obstetricians. An emergency IR service is available in about 38% of Belgian maternity units, complemented with an informal network between hospitals as shown in the data set: 20 women (12%) were transferred to a centre with IR availability. We cannot rule out the possibility that this study in fact demonstrates an overuse of IR procedures and underuse of other second-line measures in the management of postpartum haemorrhage in Belgium. This reasonable hypothesis has been previously suggested by Kayem et al22based on the results of a population-based study of women with postpartum haemorrhage managed with invasive therapies in France. The use of peripartum hysterectomy was significantly higher in multiparous women (parity ≥3) and in women who delivered in a non-IR unit. This result may not surprise, and performing a hysterectomy sooner rather than later must be considered good practice,6 as losing time trying another second-line measure can cause further blood loss and further morbidity. The decision is more easily made in cases of women who have no desire for continued fertility. Likewise, the transfer of a haemodynamically unstable patient to an IR service is hardly ever considered a safe alternative if this IR service is located in a nearby hospital. However, the higher rate of massive blood transfusion in the group of women who delivered in a non-IR unit strengthens the hypothesis that the rate of hysterectomy is associated with the IR service availability and with the expert knowledge and experience at larger units in the management of massive obstetric haemorrhage. A case-by-case in-depth analysis would be necessary to reveal whether the hysterectomies and arterial embolisations recorded in this study were appropriate or preventable. A structured audit based on medical records of 50 peripartum hysterectomies was recently performed in Denmark. This population-based audit revealed that more than 50% of peripartum hysterectomies were avoidable by simple measures such as the use of intrauterine balloons.23 AIP accounts for 23% of the interventions and 86% of the planned interventions in this data set. Targeted papers report good sensitivity and positive predictive value (PPV) of greyscale ultrasound as the screening tool for AIP in women with a high index of suspicion, enhanced by the use of transvaginal ultrasound and colour Doppler.24 25 Nonetheless, the diagnosis of AIP was missed antenatally in 34% of the women in this study, although all of them had risk factors that should have increased awareness. Several population-based studies report rather low antenatal diagnosis rates of AIP: 50% of AIP cases were missed in a study of the UK Obstetric Surveillance System, even though 30% of these missed cases had a previous CD and placenta praevia.26 Similarly, in the Nordic Obstetric Surveillance Study, 70% of AIP cases were missed, despite the fact that 33% had placenta praevia and 39% had previous CD.27 In view of these findings, we believe that antenatal diagnosis of AIP may be improved by raising the awareness of clinicians of the risk factors for AIP. This involves actively asking for previous uterine surgery at the first antenatal visit, checking the position of the placenta and focused imaging in the pregnancies at risk.28 Importantly, there is little information about the sensitivity or PPV of ultrasound or MRI as the screening tool for AIP in women who have had no previous uterine surgery or in women in whom the placenta is not implanted anteriorly in the lower uterine segment,24 which would be an interesting topic for future research. Despite increased awareness a number of cases of abnormal placentation will still be missed. For this reason, all maternity units should ensure a state of constant preparedness to deal with an unanticipated AIP, which causes a major haemorrhage at any time. Although numbers are small, this study aligns with the findings in previous studies that maternal outcome is better when women with suspicion of AIP give birth in

specialised tertiary centres and when preventive measures are taken.26 29 30 A planned preterm caesarean hysterectomy with the placenta left in situ is the indubitable treatment of choice in women with AIP and is incorporated in recommendations of the Royal College of Obstetricians and Gynaecologists (RCOG),14 the American College of Obstetricians and Gynecologists (ACOG)31 and the Society for Maternal-Fetal Medicine (SMFM).32 Alternatively, a conservative approach is advocated for women who want to preserve their fertility and are haemodynamically stable, which means a multidisciplinary approach aiming to leave the placenta in situ and to preserve the uterus.33 34 This study demonstrates that conservative management of AIP is not well incorporated in obstetric care in Belgium. The placenta was left in situ in 11 women with AIP (29%) in this data set. A conservative approach was attempted in six women and was successful in three women. Two of these women underwent a planned hysterectomy delayed until 1 month after the CD. A small number of papers support this management (leaving the placenta in situ with a planned delayed hysterectomy) in women who have a placenta percreta with infiltration in the bladder and/or other surrounding tissues and therefore a high risk of haemorrhage or adjacent tissue injury,34–36 the rationale being that this delay will allow the decrease of the periuterine vascularisation and thus prevent haemorrhage. Studies with larger numbers are needed to strengthen evidence for the claim that this methodology is safer than caesarean hysterectomy in women with severe placenta percreta. The role of IR in the management of AIP is still a matter of debate. Most guidelines agree that IR can be life-saving in the treatment of major postpartum haemorrhage.14 32 They currently do not recommend the routine use of preoperative placement of intra-arterial catheters to enable balloon occlusion or arterial embolisation before caesarean hysterectomy or conservative treatment of AIP, based on lack of evidence of benefit.14 31 32 37 Meanwhile, the number of papers reporting benefit of prophylactic IR is growing.38 39Our population-based study reports diverse management of women with AIP, inherent to a multicentre study. Twelve women with AIP were managed with programmed IR, the need for hysterectomy being averted in six of them. We cannot draw firm conclusions on the success rate of IR in the elective setting because of small numbers and lack of controls. This study has demonstrated a high success rate (87.8%) of IR in controlling postpartum haemorrhage, consistent with the success rates ranging between 85% and 95% reported by others.9–12 40–42 Univariate analysis (taking into account the small numbers) showed IR was more likely to fail in women with a CD in a previous pregnancy (failure rate 25%), women with AIP (33%) and women who had no uterotonic agents administered (44%). Several large, hospital-based, retrospective studies have analysed the predictive factors associated with failure of IR in the management of postpartum haemorrhage.12 40–43 Most papers report a lower success rate of IR in women with AIP,40 41 which explains to a great extent the lower success rates in women with previous CD and in women with CD in the current pregnancy. A population-based study in the Netherlands10 reported a 25% failure rate for IR following CD in the current pregnancy, compared with just 16% in this study and barely 10% in a large study reported by Lee et al (n=176).44Other researchers have consistently reported a higher failure rate of IR in women with various clinical determinants related to major haemorrhage: haemodynamic shock,43 45 diffuse intravascular coagulopathy,12 45 low fibrinogen and prothrombin time,41 and the need for massive transfusion.12 40 41However, these factors could be either cause or consequence of the failure of IR, which was not clearly differentiated in some of the aforementioned studies. Interestingly, hospital-to-hospital transfer was not associated with a higher failure rate of IR in this study, consistent with what has been reported by others.12 40 41

Conclusion

The prevalence in Belgium of major obstetric haemorrhage requiring peripartum hysterectomy and/or IR is estimated at 6.6 (95% CI 5.7 to 7.7) per 10 000 deliveries: roughly 1 in 3030 women who give birth in Belgium undergoes a peripartum hysterectomy, while another 1 in 3030 women is successfully managed by IR, thereby preserving the uterus. Abnormal placentation and/or uterine atony are reported as the cause of haemorrhage and reason for intervention in the majority of these women (83.7%). Further improvement in the management of obstetric haemorrhage could possibly be achieved by increased awareness of risk factors for abnormal placentation on the part of clinicians and by antenatal transfer to tertiary centres in case of suspicion. A case-by-case in-depth analysis is necessary to reveal whether the hysterectomies and arterial embolisations performed in this study were appropriate or preventable.

Acknowledgments

This study would not have been possible without the voluntary participation of the following Belgian maternity units: GZA Sint-Jozef, Mortsel; GZA Sint-Augustinus, Wilrijk; GZA Sint-Vincentius, Antwerpen; ZNA, SintErasmus, Borgerhout; ZNA, Jan Palfijn, Merksem; ZNA, Middelheim, Antwerpen; ASZ, campus Aalst; ASZ, campus Geraardsbergen; AZ Lokeren; Imelda Ziekenhuis, Bonheiden; Sint-Jozef ziekenhuis, Bornem; AZ Klina, Brasschaat; AZ Sint-Jan, Brugge; AZ Sint-Lucas, Assebroek; Centre de Santé des Fagnes; Centre Hospitalier de l’Ardenne; Centre Hospitalier EpiCura, site Hornu; Centre Hospitalier de Mouscron; Centre Hospitalier du Bois de l’Abbaye et de Hesbay; Centre Hospitalier Peltzer, La Tourelle; CHC Saint-Vincent; CHC Saint-Joseph; CHC Sainte-Elisabeth; CHIREC, Clinique Sainte-Anne, Saint-Remi; CHIREC Braine- l’Alleud-Waterloo; CHR de la Citadelle; CHR de Namur; CHR du Val de Sambre; CHR Haute Senne-Le Tilleriau; CHR Mons Clinique Saint-Joseph; CHU Ambroise Paré; CHU Brugmann; CHU Charleroi, André Vésale; CHU Notre-Dame des Bruyères; CHU Saint- Pierre; CHU Tivoli; CHwapi-Site Notre-Dame; Clinique Edith Cavell; Clinique et Maternité Sainte Elisabeth; Clinique Notre dame de Grâce; Clinique Reine Astrid;

Clinique SaintePiere; Cliniques de l’Europe, Saint-Michel; Cliniques de l’Europe, Sainte-Elisabeth; Cliniques du Sud Luxembourg; Cliniques Universitaires Saint-Luc; AZ SintVincentius, Deinze; AZ Sint-Blasius, Dendermonde; AZ Monica, Deurne; AZ Diest; AZ Sint-Maarten, Duffel; AZ Alma, Eeklo; Dimpna Ziekenhuis, Geel; AZ Jan Palfijn, Gent; AZ Maria Middelares, Gent; AZ Sint-Lucas, Gent; GHDC - Notre Dame, Charleroi; AZ Maria ziekenhuis, Halle; Jessaziekenhuis, Hasselt; AZ Sint Elisabeth, Herentals; CAZ Midden-Limburg, Heusden-Zolder; Hôpital Civil Marie Curie; Hôpital Erasme, Bruxelles; Hôpitaux Iris Sud – Site Ixelles; Jan Yperman ziekenhuis, Ieper; IFAC Hôpital Princesse Paola; INDC Entité Jolimontoise; Sint-Jozef ziekenhuis, Izegem; Kliniek Sint- Jan; Klinik Saint-Josef; AZ Zeno, campus Knokke-Heist; AZ Groeninge, Kortrijk; Heilig Hart ziekenhuis, Leuven; UZ Leuven; Heilig Hart ziekenhuis, Lier; AZ Sint-Jozef, Malle; AZ Sint-Maarten, Mechelen; AZ Delta, campus Menen; Heilig Hart ziekenhuis, Mol; Onze-Lieve-Vrouw ziekenhuis, campus Aalst; Onze-LieveVrouw ziekenhuis, campus Asse; AZ Damiaan, campus Sint-Jozef, Oostende; AZ Sint-Jan, campus Oostende; AZ Oudenaarde; Mariaziekenhuis Noord-Limburg, Overpelt; AZ Heilige Familie, Reet AZ Delta, campus Brugsesteenweg, Roeselare; AZ Delta, campus Wilgenstraat, Roeselare; AZ Glorieux, Ronse; AZ Nikolaas, Sint-Niklaas; Sint-Trudo ziekenhuis, Sint-Truiden; Saint-Nikolaus Hospital; SintAndries ziekenhuis, Tielt; Heilig Hartziekenhuis, Tienen; AZ Vesalius, Tongeren; Sint-Rembertziekenhuis, Torhout; AZ Turnhout; UZ Antwerpen; UZ Brussel; AZ SintAugustinus, Veurne; AZ Jan Portaels, Vilvoorde; Onze- Lieve-Vrouw van Lourdes ziekenhuis, Waregem; Ziekenhuis Oost-Limburg, campus Sint-Jan, Genk; AZ SintElisabeth, Zottegem. We would like to thank Fatima Bercha, Marlies Deblaere and Ann Langedock for their contribution in collecting the data. References

1. Evaluating the quality of care for severe pregnancy complications: the WHO near-miss approach for maternal health. Geneva, Switzerland: WHO press 2011. Retrieved from http://apps.who.int/iris/bitstream/10665/44692/1/9789241502221_eng.pdf. Accessed 22 Dec. 16. 2. Stones W, Lim W, Al-Azzawi F, Kelly M. An investigation of maternal morbidity with identification of life-threatening 'near miss' episodes. Health trends. 1991;23(1):13-5. 3. Say L, Souza JP, Pattinson RC. Maternal near miss--towards a standard tool for monitoring quality of maternal health care. Best practice & research Clinical obstetrics & gynaecology. 2009;23(3):287-96. 4. Lewis, G (ed) 2007. The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving Mother's Lives: reviewing maternal deaths to make motherhood safer-2003-2005. The Seventh Report on Confidential Enquiries into Maternal Deaths in the United Kingdom. London: CEMACH. 5. Knight M, Tuffnell D, Kenyon S, , Kenyon S, Shakespear J, Gray R, Kurinczuk JJ (Eds.) on behalf of MBRRACE-UK. Saving Lives, Improving Mother's Care - Surveillance of maternal deaths in the UK 2011- 13 and lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2009-13. Oxford: National Perinatal Epidemiology Unit, University of Oxford 2015. 6. The Royal College of Obstetricians and Gynaecologists. Postpartum Haemorrhage, Prevention and Management. Green-top Guideline No. 52. Published: 11/05/2009. 7. Postpartum Hemorrhage. The American College of Obstetricians and Gynaecologistis. ACOG Practice Bulletin. Number 76, October 2006. Reaffirmed 2015. 8. Dahlke JD, Mendez-Figueroa H, Maggio L, Hauspurg AK, Sperling JD, Chauhan SP, et al. Prevention and management of postpartum hemorrhage: a comparison of 4 national guidelines. American journal of obstetrics and gynecology. 2015;213(1):76.e1-10. 9. Kayem G, Kurinczuk JJ, Alfirevic Z, Spark P, Brocklehurst P, Knight M. Specific second-line therapies for postpartum haemorrhage: a national cohort study. BJOG : an international journal of obstetrics and gynaecology. 2011;118(7):856-64. 10. Zwart JJ, Dijk PD, van Roosmalen J. Peripartum hysterectomy and arterial embolization for major obstetric hemorrhage: a 2-year nationwide cohort study in the Netherlands. American journal of obstetrics and gynecology. 2010;202(2):150.e1-7. 11. Inoue S, Masuyama H, Hiramatsu Y. Efficacy of transarterial embolisation in the management of post- partum haemorrhage and its impact on subsequent pregnancies. The Australian & New Zealand journal of obstetrics & gynaecology. 2014;54(6):541-5. 12. Lee HY, Shin JH, Kim J, Yoon HK, Ko GY, Won HS, et al. Primary postpartum hemorrhage: outcome of pelvic arterial embolization in 251 patients at a single institution. Radiology. 2012;264(3):903-9. 13. The Royal College of Obstetricians and Gynaecologists. The role of emergency and elective interventional radiology in postpartum haemorrhage. (Good Practice No. 6). June 2007. 14. The Royal College of Obstetricians and Gynaecologists. Placenta Praevia, Placenta Praevia Accreta and Vasa Praevia : Diagnosis and Management. Green-top Guideline No. 27. Published January 2011. 15. Investigation into 10 Maternal Deaths at, or following Delivery at, Northwick Park Hospital, North West London Hospitals NHS Trust, between April 2002 and April 2005. Commission for Healthcare Audit and Inspection. London, August 2006. 16. Vandenberghe G, De Blaere M, Van Leeuw V, Roelens K, Englert Y, Hanssens M, et al. Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2016;6(5):e010415. 17. H. Cammu EM, G. Martens, C. Van Mol, Y Jacquemyn. Perinatale activiteiten in Vlaanderen 2013. 2014. Retrieved from: https://www.zorg-en- gezondheid.be/sites/default/files/atoms/files/SPE_jaarrapport%202013.pdf. Accessed on 22 Dec. 16 18. H. Cammu EM, G. Martens, C. Van Mol, Y Jacquemyn. Perinatale Activiteiten in Vlaanderen 2012. 2013. Retrieved from: https://www.zorg-en- gezondheid.be/sites/default/files/atoms/files/SPE_jaarrapport%202012.pdf. Accessed on 22 Dec. 16.

19. Van Leeuw V LC, Englert Y. Perinatale gegevens in het Brussels Gewest - Jaren 2013. Centre d'Epidémiologie périnatale, 2014. Retrieved from: http://www.cepip.be/pdf/rapport_CEPIP_Bxl2013_NL.pdf. Accessed on 22 Dec. 16. 20. Leroy Ch VLV, Englert Y. Données périnatales en Wallonie - Année 2013. Centre d'Epidémiologie Périnatale, 2014. Retrieved from : http://www.cepip.be/pdf/rapport_CEPIP_wallonie2013_tma.pdf. Accessed on 22 Dec. 16. 21. Armonk NIC. IBM SPSS Statistics for Windows, Version 22.0. In: Corp I, editor. 2013. 22. Kayem G, Dupont C, Bouvier-Colle MH, Rudigoz RC, Deneux-Tharaux C. Invasive therapies for primary postpartum haemorrhage: a population-based study in France. BJOG : an international journal of obstetrics and gynaecology. 2016;123(4):598-605. 23. Colmorn LB, Krebs L, Langhoff-Roos J. Potentially Avoidable Peripartum Hysterectomies in Denmark: A Population Based Clinical Audit. PloS one. 2016;11(8):e0161302. 24. Comstock CH, Bronsteen RA. The antenatal diagnosis of placenta accreta. BJOG : an international journal of obstetrics and gynaecology. 2014;121(2):171-81; discussion 81-2. 25. D'Antonio F, Iacovella C, Bhide A. Prenatal identification of invasive placentation using ultrasound: systematic review and meta-analysis. Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology. 2013;42(5):509-17. 26. Fitzpatrick KE, Sellers S, Spark P, Kurinczuk JJ, Brocklehurst P, Knight M. The management and outcomes of placenta accreta, increta, and percreta in the UK: a population-based descriptive study. BJOG : an international journal of obstetrics and gynaecology. 2014;121(1):62-70; discussion -1. 27. Thurn L, Lindqvist PG, Jakobsson M, Colmorn LB, Klungsoyr K, Bjarnadottir RI, et al. Abnormally invasive placenta-prevalence, risk factors and antenatal suspicion: results from a large population-based pregnancy cohort study in the Nordic countries. BJOG : an international journal of obstetrics and gynaecology. 2016;123(8):1348-55. 28. Collins SL, Ashcroft A, Braun T, Calda P, Langhoff-Roos J, Morel O, et al. Proposal for standardized ultrasound descriptors of abnormally invasive placenta (AIP). Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology. 2016;47(3):271-5. 29. Chantraine F, Braun T, Gonser M, Henrich W, Tutschek B. Prenatal diagnosis of abnormally invasive placenta reduces maternal peripartum hemorrhage and morbidity. Acta obstetricia et gynecologica Scandinavica. 2013;92(4):439-44. 30. Eller AG, Bennett MA, Sharshiner M, Masheter C, Soisson AP, Dodson M, et al. Maternal morbidity in cases of placenta accreta managed by a multidisciplinary care team compared with standard obstetric care. Obstetrics and gynecology. 2011;117(2 Pt 1):331-7. 31. Placenta Accreta. The American College of Obstetricians and Gynaecologistis. Committee Opinion. Number 529. Published July 2012. Reaffirmed 2015. 32. Belfort MA. Placenta accreta. American journal of obstetrics and gynecology. 2010;203(5):430-9. 33. Kayem G, Davy C, Goffinet F, Thomas C, Clement D, Cabrol D. Conservative versus extirpative management in cases of placenta accreta. Obstetrics and gynecology. 2004;104(3):531-6. 34. Sentilhes L, Ambroselli C, Kayem G, Provansal M, Fernandez H, Perrotin F, et al. Maternal outcome after conservative treatment of placenta accreta. Obstetrics and gynecology. 2010;115(3):526-34. 35. Fox KA, Shamshirsaz AA, Carusi D, Secord AA, Lee P, Turan OM, et al. Conservative management of morbidly adherent placenta: expert review. American journal of obstetrics and gynecology. 2015;213(6):755-60. 36. Chantraine F, Nisolle M, Petit P, Schaaps JP, Foidart JM. Individual decisions in placenta increta and percreta: a case series. Journal of perinatal medicine. 2012;40(3):265-70. 37. Dilauro MD, Dason S, Athreya S. Prophylactic balloon occlusion of internal iliac arteries in women with placenta accreta: literature review and analysis. Clinical radiology. 2012;67(6):515-20. 38. Cali G, Forlani F, Giambanco L, Amico ML, Vallone M, Puccio G, et al. Prophylactic use of intravascular balloon catheters in women with placenta accreta, increta and percreta. European journal of obstetrics, gynecology, and reproductive biology. 2014;179:36-41. 39. Duan XH, Wang YL, Han XW, Chen ZM, Chu QJ, Wang L, et al. Caesarean section combined with temporary aortic balloon occlusion followed by uterine artery embolisation for the management of placenta accreta. Clinical radiology. 2015;70(9):932-7. 40. Sentilhes L, Gromez A, Clavier E, Resch B, Verspyck E, Marpeau L. Predictors of failed pelvic arterial embolization for severe postpartum hemorrhage. Obstetrics and gynecology. 2009;113(5):992-9. 41. Poujade O, Zappa M, Letendre I, Ceccaldi PF, Vilgrain V, Luton D. Predictive factors for failure of pelvic arterial embolization for postpartum hemorrhage. International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics. 2012;117(2):119-23. 42. Yamasaki Y, Morita H, Miyahara Y, Ebina Y, Okada T, Yamaguchi M, et al. The factors associated with the failure of transcatheter pelvic arterial embolization for intractable postpartum hemorrhage. Journal of perinatal medicine. 2014;42(3):359-62. 43. Touboul C, Badiou W, Saada J, Pelage JP, Payen D, Vicaut E, et al. Efficacy of selective arterial embolisation for the treatment of life-threatening post-partum haemorrhage in a large population. PloS one. 2008;3(11):e3819. 44. Lee HJ, Jeon GS, Kim MD, Kim SH, Lee JT, Choi MJ. Usefulness of pelvic artery embolization in cesarean section compared with vaginal delivery in 176 patients. Journal of vascular and interventional radiology : JVIR. 2013;24(1):103-9. 45. Kim YJ, Yoon CJ, Seong NJ, Kang SG, An SW, Kim YS, et al. Failed pelvic arterial embolization for postpartum hemorrhage: clinical outcomes and predictive factors. Journal of vascular and interventional radiology : JVIR. 2013;24(5):703-9.

Footnotes

Contributors Project conception: MH, YE. Study design: GV, VVL, JV, MH, YE. Data collection: GV, MG, IR, VVL, JV. Data analysis: GV, MG, IJ, ER. Data interpretation: GV, MG, IJ, KR, MH, ER, YE, HV. Manuscript first draft: GV. Contributed to the writing of the manuscript: MG, IJ, KR, MH, YE, HV. Manuscript revision: MG, IJ, VVL, KR, MH, ER, JV, YE, HV. All authors approved the final version, agreed to be accountable for all aspects of the work and met the ICMJE criteria for authorship. Funding The study was funded by the College for Mother and Newborn, a consultative body of the Belgian Public Health Service. GV was funded by the Flemish Research Foundation (FWO). Competing interests All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and hereby declare that they have no financial relationships with any organisations that might have an interest in the submitted work in the past 3 years, nor that they have partaken of any other relationships or activities that could appear to have influenced the submitted work. Patient consent Data were retrieved in retrospect from the case notes by means of data collection forms. No personally identifiable information was obtained. Participants were informed and enabled to opt-out. Ethics approval Medical Ethics Committee of the Ghent University Hospital. Provenance and peer review Not commissioned; externally peer reviewed. Data sharing statement There are no additional unplished data from the study available.

THE INOSS STUDY OF UTERINE RUPTURE: A DESCRIPTIVE MULTI-COUNTRY POPULATION BASED STUDY Griet Vandenberghe1, Kitty Bloemenkamp2, Silvia Berlage3, Lotte Colmorn4, Catherine Deneux-Tharaux5, Mika Gissler6,7, Marian Knight8, Jens Langhoff-Roos9, Pelle G Lindqvist10,11, Willi Oberaigner12,13, Jos Van Roosmalen14,15, Joost Zwart16, Kristien Roelens1. On behalf of INOSS (the International Network of Obstetric Survey Systems).

1. Department of Obstetrics & Gynaecology, Ghent University Hospital, Ghent, Belgium. 2. Department of Obstetrics & Gynaecology, Wilhelmina’s Children Hospital, University Medical Centre Utrecht, the Netherlands. 3. Center for Quality and Management in Health Care, Medical Association of Lower Saxony, Berliner Allee 20, 30175 Hannover, Germany. 4. Department of Obstetrics, Rigshospitalet, University Hospital of Copenhagen, Copenhagen, Denmark. 5. Inserm U1153, Obstetrical, Perinatal and Pediatric Epidemiology Research Team, Center for Epidemiology and Statistics Sorbonne Paris Cité, Paris Descartes University, Paris, France. 6. THL National Institute for Health and Welfare, Information Services Department, Mannerheimintie 166, 00270 Helsinki, Finland. 7. Department of Neurobiology, Care Sciences and Society, Division of Family Medicine, Karolinska Institute, 171 77 Stockholm, Sweden. 8. National Perinatal Epidemiology Unit, University of Oxford, Oxford, United Kingdom. 9. Department of Obstetrics, Rigshospitalet, University Hospital of Copenhagen, Copenhagen, Denmark. 10. Department of Clinical Science Intervention and Technology (CLINTEC), Karolinska Institute, Stockholm, Sweden. 11. Department of Obstetrics and Gynecology, Sodersjukhuset, Stockholm, Sweden. 12. Department of Clinical Epidemiology of the Tirol Kliniken Ltd., Innsbruck, Austria. 13. Institute of Public Health, Medical Decision Making and Health Technology Assessment, Department of Public Health, Health Services Research and Health Technology Assessment, UMIT - University for Health Sciences, Medical Informatics and Technology, Hall i.T., Austria. 14. Athena Institute, VU University Amsterdam, the Netherlands. 15. Department of Obstetrics, Leiden University Medical Centre, Leiden, The Netherlands. 16. Department of Obstetrics and Gynaecology, Deventer Hospital, Deventer, the Netherlands.

Abstract Objective International comparison of complete uterine rupture. Design Descriptive multi-country population-based study. Setting International. Population International Network of Obstetric Survey Systems (INOSS) Methods We merged individual data, collected prospectively in nine population-based studies, of women with complete uterine rupture, defined as complete disruption of the uterine muscle and the uterine serosa, regardless of symptoms and rupture of fetal membranes. Main Outcome Measures Prevalence of complete uterine rupture, regional variation and correlation with rates of caesarean section (CS) and trial of labour after CS (TOLAC). Severe maternal and perinatal morbidity and mortality. Results We identified 864 complete uterine ruptures in 2,625,017 deliveries. Overall prevalence was 3.3 (95% CI 3.1-3.5) per 10,000 deliveries, 22 (95% CI 21-24) in women with and 0.6 (95% CI 0.5-0.7) in women without previous CS. Prevalence in women with previous CS was negatively correlated with previous CS rate (rho=-0.917) and positively correlated with TOLAC rate of the background population (rho=0.600). Uterine rupture resulted in peripartum hysterectomy in 87 of 864 women (10%, 95% CI 8-12%) and in a perinatal death in 116 of 874 infants (13.3%, 95% CI 11.2-15.7) whose mother had uterine rupture. Overall rate of neonatal asphyxia was 28% in neonates who survived. Conclusions Higher prevalence of complete uterine ruptures per TOLAC was observed in countries with low previous CS and high TOLAC rates. Rates of hysterectomy and perinatal death are about 10% following complete uterine rupture, but in women undergoing TOLAC the rates are extremely low (only 2.2 and 3.2 per 10,000 TOLACs respectively). Funding Research Foundation Flanders (FWO). Keywords Uterine rupture, caesarean section, trial of labour after caesarean section, population- based, severe maternal morbidity. Tweetable abstract Prevalence of complete uterine rupture is higher in countries with low previous CS and high TOLAC rates.

Introduction

Complete uterine rupture is the disruption of the full-thickness of the uterine wall, including myometrium and uterine serosa, during pregnancy or delivery.1, 2 This obstetric complication is rare with prevalences between 2.4 and 3.0 per 10,000 deliveries3, 4, 5, but associated with severe outcomes for both mother and child. An emergency intervention is needed in order to prevent perinatal asphyxia and death, while the mother may suffer from life threatening haemorrhage and end up with peripartum hysterectomy. An increasing prevalence of complete uterine rupture can be largely explained by an increase in trial of labour after caesarean section (TOLAC) (Supporting Figure S1).6-9 In general, TOLAC has been appreciated as a way to reduce increasing CS rates, since Edward Cragin’s statement “once a cesarean always a cesarean” one hundred years ago (1916).10 Practice guidelines on trial of labour after caesarean section published by leading professional associations also recommend TOLAC as a clinically safe choice for the majority of women with one previous CS, to limit further increase of CS rates and maternal morbidity associated with multiple CSs.11-15 However, TOLAC has become the main risk factor for uterine rupture especially in high-income countries and fear of medical litigation is an important reason for obstetricians to be reluctant to offer TOLAC to their patients.16 Population-based rates of TOLAC differ considerably between countries.16-18 The International Network of Obstetric Survey Systems (INOSS) is a multinational collaboration of organisations conducting prospective population-based studies of serious disorders in pregnancy and childbirth.19 Nine INOSS members have conducted population-based studies on uterine rupture, the first started in 2004 and the latest completed in 2014, using comparable definitions and data collection forms, enabling international comparison of prevalence, risk factors, management and outcome. We aimed to assess the overall prevalence of complete uterine rupture in these nine countries, to compare prevalences between the different countries and to assess the correlation between prevalence of complete uterine rupture and the countries’ caesarean section and trial of labour after caesarean section rates. We hypothesise that rates of uterine rupture per country will be proportional to rates of previous CS per country and rates of trial of labour after caesarean section per country. Further, we aimed to describe overall maternal and perinatal outcome associated with complete uterine rupture.

Methods

This study is a descriptive multi-country population-based study. We merged individual data of women with complete uterine rupture collected prospectively in nine population-based studies within INOSS: AuOSS (Austrian Obstetric Surveillance System) in Austria, Belgian Obstetric Surveillance System (B.OSS) in Belgium, EPIMOMS (EPIdemiologie de la MOrbidité Maternelle Sévère) in France, Nordic Obstetric Surveillance System (NOSS) with data from Denmark, Finland and Sweden, GerOSS (German Obstetric Surveillance System) in Lower Saxony, Germany; LEMMoN (Landelijke studie naar Etnische determinanten van Maternale Morbiditeit , now called NethOSS, Netherlands Obstetric Surveillance System) in the Netherlands and UKOSS (United Kingdom Obstetric Surveillance System) in the United Kingdom. All population-based studies were nationwide, except for GerOSS in Germany that encompassed the region of Lower Saxony only, and EPIMOMS in France that encompassed six regions (Alsace, Auvergne, Basse-Normandie, Île-de-France, Lorraine and Rhône-Alpes) covering one-fifth of national births. Women were actively involved designing, managing and interpreting data for individual component studies.

For the purpose of this study we included only cases that correspond to an objective anatomical definition of complete uterine rupture, namely complete disruption of the uterine muscle and the uterine serosa, irrespective of the condition of the fetal membranes. The definitions of uterine rupture used by the different participating countries are listed in Supporting Table S1. There were two countries where a compromise was needed to enable inclusion of their registered cases. The Netherlands used a clinical definition rather than an anatomical definition, namely cases of uterine rupture where clinical symptoms led to an emergency CS or laparotomy; we agreed to include all registered cases, aware that some may have been incomplete ruptures. France registered cases of complete uterine rupture only if associated with Severe Acute Maternal Morbidity (SAMM). This comprises only cases of complete uterine rupture with severe obstetric haemorrhage and cases which underwent peripartum hysterectomy or laparotomy for complete uterine rupture after vaginal birth.20 Uterine ruptures occurring in labour after a previous caesarean section were used as a proxy for uterine ruptures in women undergoing TOLAC. Short term maternal outcomes were peripartum hysterectomy and maternal mortality. Severe outcome of neonates was defined as neonatal asphyxia and/or perinatal mortality. Neonatal asphyxia was defined as umbilical artery pH ≤7.0 and/or

Apgar score <7 at 5 minutes and/or need for resuscitation including endotracheal ventilation and/or neurologic manifestations (seizures).21 Stillbirth was defined as intrauterine fetal death after 24 completed weeks. Perinatal mortality was defined as stillbirths and early neonatal deaths (up to 7 days of life). Stillbirths that occurred before uterine rupture were not included in perinatal mortality rates (PMR).

Data-collection in each population-based study was accomplished according to prospective case-collection schemes, similar to UKOSS methodology.22 Call for cases took place on a (bi-, tri- or mostly) monthly basis by sending a report card by email or fax to a nominated clinician in each maternity unit in the country or region under surveillance, asking them to report all cases of uterine rupture that occurred in the past month and to complete the report card also when there was no case to report. Data-collection differed in France, Finland and Denmark: in France clinicians were contacted by monthly calls or emails, in Denmark and Finland clinicians reported their cases spontaneously, besides registers (Medical Birth Register, Hospital Discharge Register and/or delivery logbooks) were used to identify missing cases which were added to the study after they were found to be true cases. For every reported case clinicians, research midwives or scientific collaborators completed data collection forms (paper or electronic). These sought confirmation of the appropriate case definition and additional information on risk factors, management and outcomes. All countries used paper-based or web-based data-collection forms (www.geross.de, www.b-oss.be, www.noss.nu). Each population-based study group processed and cleaned their data. National data were published in peer-reviewed journals and/or organisational reports before onset of this international study.17, 18, 23-25 Anonymous data were transferred to the principal investigator of this study, according to applicable data transfer policies, which differed by country.

To facilitate matching of nine different databases, 37 variables were selected based on 1) their presence in data collection forms of all nine countries and 2) their significance for analysis. Set of 37 variables and codebook are listed in Supporting Table S2. Variables comprised details about previous obstetric history, previous medical history, current pregnancy and labour, uterine rupture and its management, maternal and perinatal outcome. Proportion of women who had previously delivered by CS (%), CS rates, trial of labour after caesarean section (TOLAC) rates and vaginal birth after caesarean section (VBAC) rates in each country were derived from the most recent available perinatal data or the perinatal data closest to the study period or from previous population based studies.17, 18, 24-32 At the time of analysis, a core

outcome set (COS) for severe maternal morbidity in general or uterine rupture specifically, had not yet been developed.33

Data were analysed using statistical software package IBM SPSS Statistics for Windows, Version 22.0 .34 For most research objectives, variables were expressed as relative frequencies and analysis confined to descriptive statistics. Prevalences were calculated with 95% confidence intervals using the Wilson score method.35 Lower limits of gestational age appeared to differ within and between countries in uterine rupture studies as well as in civil birth registers (Supporting Table S3). In an attempt to adjust for this discrepancy, prevalences were re-calculated using as numerator cases of complete uterine rupture ≥ 24 weeks of gestational age and as denominator the estimated number of deliveries in the background population ≥ 24 weeks of gestational age, if information was available. These adjusted prevalences differed little (maximum 0.67 per 10,000 deliveries) from the overall prevalences (Supporting Table S4). Estimated numbers of deliveries with and without previous CS (Table 1) were calculated based on estimates of the proportion of women previously delivered by CS (%) in each country. Correlations between prevalence of complete uterine rupture and previous CS, CS, TOLAC and VBAC rates per country were calculated using nonparametric Spearman Rank Order correlations (rho). P-values of <0.05 were considered statistically significant. Estimated numbers per 10,000 deliveries per country (Figure 1) were calculated based on the numbers in Table 1 and Supporting Table S5. We calculated estimated numbers of women with previous CS per 10,000 deliveries, and, similarly, estimated numbers of women with previous CS who underwent TOLAC, who succeeded in VBAC, who underwent emergency CS following TOLAC and who underwent elective repeat CS (ERCS) per 10,000 deliveries (Supporting Figure S1). Perinatal mortality and neonatal asphyxia rates were calculated using as denominator all neonates of women with complete uterine rupture, with exclusion of immature neonates (<24 weeks, n=24), stillbirths prior to uterine rupture (n=8), neonates with major congenital anomalies (n=7) and neonates with unknown outcome (missing gestational age (n=4), data missing on stillbirth versus live birth (n=84), liveborn neonates with data missing to assess asphyxia (n=64), live born neonates suffering from asphyxia but missing data on neonatal mortality (n=5)).

Results

This multi-country study comprised 864 cases of complete uterine rupture in 2,625,017 deliveries from nine countries with an overall prevalence of 3.3 (95% CI 3.1-3.5) per 10,000 deliveries (Supporting Table S6). Finland reported the highest prevalence of 7.8 (95% CI 6.5-9.4) per 10,000 deliveries and Austria the lowest prevalence of 1.6 (95% CI 1.1-2.2) per 10,000 deliveries. Estimated prevalence of complete uterine rupture in women with previous CS was 22 (95% CI 21-24) per 10,000 deliveries as compared to 0.6 (95% CI 0.5-0.7) per 10,000 deliveries in women without previous CS. Estimated prevalences per country are shown in Supporting Table S7.

Associations between estimated prevalence of complete uterine rupture and rates of CS, previous CS, trial of labour after caesarean section and vaginal birth after caesarean section in countries are displayed in Table 1. We noted a strong negative correlation between previous CS rates and prevalence of uterine rupture (rho=-0.854 for total prevalence and rho=-0.917 for prevalence in women with previous CS). Likewise, we identified a strong positive correlation between TOLAC rates and prevalence of uterine rupture (rho=0.611 for total prevalence and rho=0.600 for prevalence in women with previous CS) (Supporting Figures S2-4). Additionally, VBAC rates were positively correlated with the proportion of TOLAC in the countries (rho=0.740). Based on the numbers in Table 1 and Supporting Table S5 we estimated the number of women with previous CS and, similarly, the number of women with previous CS who underwent TOLAC, who achieved VBAC, who underwent emergency CS following TOLAC and who underwent elective repeat caesarean section (ERCS) per 10,000 deliveries (Figure 1) (Supporting Figure S1). Countries with lower overall CS rates, thus more restrictive in performing CSs, and consequently having lower numbers of women with previous CS, appear more tolerant in their selection of candidates for TOLAC, and more successful in achieving VBAC, at the expense of a slightly higher number of uterine rupture per TOLAC. In contrast, countries with higher overall CS rates, thus more permissive in performing CSs, and consequently having a higher number of women with previous CS, are more selective in allowing TOLAC, and less successful in achieving VBAC, in favour of a slightly lower number of uterine rupture per TOLAC.

Management and maternal outcome of women with complete uterine rupture are reported in Supportive Table S8. The proportionally high number of women who had laparotomy following vaginal delivery, who underwent hysterectomy or received more than 4 units of Packed Cells (PC) in the French study is related to the different inclusion criteria, where Severe Acute Maternal Morbidity (SAMM) was a prerequisite for inclusion.20 All countries verified maternal deaths and only two were reported, both in the study from the United Kingdom resulting in an overall case fatality rate of 0.23% (95% CI 0.06-0.84%). Forty of 627 women with uterine rupture associated with TOLAC underwent hysterectomy (6.4%, 95% CI 4.7-8.6), resulting in a rupture-related hysterectomy rate of 2.2 (95% CI 1.6-3.0) per 10,000 women undergoing TOLAC, taking into account the overall uterine rupture rate of 35 (95% CI 32-37) per 10,000 TOLACs.

We analysed maternal and perinatal outcome resulting from complete uterine rupture in 864 women. The study involved 8 twins and 1 triplet pregnancy, resulting in 874 fetuses or newborns. Characteristics of women with complete uterine rupture and their fetuses/neonates are shown in Supportive Table S10. The majority of uterine ruptures occurred beyond 37 weeks (722 women with 727 neonates, 83.2%), of whom 161 neonates (18%) were beyond 41 weeks and 56 (6%) beyond 42 weeks. Perinatal outcome in all cases of complete uterine rupture is shown in Figure 2a. Perinatal outcome of a subgroup of women who underwent trial of labour after caesarean section, is shown in Figure 2b. The rupture-related perinatal mortality rate in this multi-country study was 116 (95% CI 95-141) per 1,000 neonates; neonatal asphyxia was observed in 280 (95% CI 248- 315) per 1,000 neonates who survived (supporting Table S11), resulting in 39.6% of neonates with severe outcome. In women who underwent trial of labour after caesarean section, perinatal mortality rate was 90 (95% CI 68-117) per 1,000 neonates and the rate of neonatal asphyxia was 263 (95% CI 226-304) per 1,000 surviving neonates, resulting in 35.3% of neonates with severe outcome. Perinatal mortality rate ranges from 0.8 to 13.2 per 10,000 TOLACs between countries (Table 2).

Discussion

There is a wide variation in prevalence of complete uterine rupture between countries (1.6 to 7.8 per 10,000 deliveries), which also exists in the rate of uterine rupture per TOLAC (16 to 80 per 10,000 deliveries), proportional to TOLAC uptake rates per country and inversely proportional to CS rates per country. Sixty percent of neonates of women with complete uterine rupture overall and 65% of neonates with TOLAC did not have severe outcome.

Strength of this study is its international design: in the INOSS collaboration we managed to collect a large number of cases of validated complete uterine ruptures from nine countries or regions by accumulation of small numbers registered according to analogous methods. Moreover, the multi-national design with participation of nine European countries facilitated comparison between high-income countries with varying CS and TOLAC rates. Ascertainment-bias caused by the inclusion of two studies that used different definitions of uterine rupture, however, is an important limitation. Systematic over-reporting of cases in the Netherlands and restriction of inclusions to cases associated with SAMM in France will certainly affect the comparison between cases, and conclusions should therefore be interpreted with caution. Another limitation is lack of information on intended mode of delivery. Thus, women who presented in spontaneous labour despite having planned elective repeat caesarean section (ERCS) (Supporting Figure S1) were classified by actual mode of delivery, and the number of women undergoing TOLAC may be overestimated. Also, TOLAC rates of the background populations were derived from the countries’ perinatal data using the same approximation in Austria, Belgium, France, Germany, Denmark, Finland and Sweden. To overcome these limitations of varying definitions of completeness of a uterine rupture, INOSS has developed international consensus definitions for severe maternal morbidities including complete uterine rupture using a Delphi procedure, and developed a set of core outcome measures for future studies.2

Prevalence of uterine rupture This is the largest prospective population-based report in high-income countries with clinical verification of uterine rupture i.e. not based on ICD-codes. Prevalences of uterine rupture in the literature differ greatly, according to whether studies are register-based, population- or facility-based, occurred in high- or low- and middle-

income countries, depending on the definition used (e.g. complete, incomplete, symptomatic rupture, dehiscence), on gestational age cut-off (e.g. including first and second trimester versus only term pregnancies) and on data sources. Register-based surveys that rely only on ICD-codes are known to over-estimate (false positives), whilst they miss a substantial number of true uterine ruptures that are under- reported (false negatives).36, 37 Other population-based reports on overall rates of uterine rupture are scarce. The World Health Organisation’s (WHO) systematic review in 2005 reported a median prevalence of 2.5 (range 1.6-7.0) per 10,000 deliveries based on five studies in high-income countries from 1995 to 1998.3 In a large Norwegian study register-based study validated by review of medical records the overall prevalence over 40 years was 1.3 (95% CI 1.2-1.5), increasing from 1.0 (95% CI 0.6-1.3) to 3.0 (95% CI 2.4-3.6) per 10,000 deliveries from the first to the last decade (2000-2008).4 A large population-based study in the United States reported a prevalence of 2.4 (95% CI 2.3-2.5) per 10,000 deliveries based on vital statistics data from 2011 to 2012.5 We further explored the literature for population-based reports of uterine rupture in women with previous CS in high-income countries. Compared to three relatively large sources (Al-Zirqi et al.,4 Guise et al. 38, Landon et al.7) reporting prevalences between 30 (95% CI 24-36) and 36 (95% CI 30-43) per 10,000 deliveries,4, 7, 9 the prevalence in our study is lower, with a notable wide range between countries. Utilisation patterns of trial of labour after caesarean section The range in prevalence in women with previous CS appears to be, amongst others, a consequence of the range in TOLAC rates between countries (from 47% to 72%). Practice guidelines on TOLAC published by four large professional associations11-14 differ only modestly on a few matters, e.g. availability of appropriate personnel to perform emergency caesarean section. Guidelines recommend that the final decision to undergo TOLAC should be made through shared decision-making by the woman in consultation with her obstetrician.11, 12 It stands to reason, supported by growing evidence, that obstetricians can easily influence women’s decisions.16, 39 Obstetricians, in turn, will be influenced by factors related to the health care system, institutional and organisational factors, and litigation.16 Health care providers can contribute to lower TOLAC rates by providing higher reimbursement for CSs compared to vaginal deliveries. Yee et al. identified a more than 2-fold increase in the likelihood of attempting TOLAC, with a similar increase in VBAC, in a night float call system (schedule in which the provider has clinical responsibilities for labouring patients only for a day or night shift) compared with a traditional call system.39 Obstetricians working in institutions with obstetric, paediatric, anaesthetic and operating room staff immediately available might be more likely to offer TOLAC. According to the NIH, ACOG recommendations on staff availability caused a number of American hospitals to stop offering TOLAC and therefore restricted women’s access to TOLAC. This international comparison study suggests a need to critically

appraise TOLAC policies in each country: first, the selection of appropriate candidates and second, management of labour in women who decided to undergo TOLAC. Maternal outcome The rate of hysterectomy related to uterine rupture in our study is 10% (n=87). A Norwegian population-based study over 40 years reported a rupture-related hysterectomy rate of 11.4% in the last decade (2000-2008) with a major decrease from 35.5% in the first decade (1967-1977).4 Overall hysterectomy rate in women with previous CS undergoing TOLAC was 0.17% in a systematic review by Guise et al.38 and 0.2% in a prospective cohort study by Landon et al.7 Hence, we can state that the rupture-related hysterectomy rate in women undergoing TOLAC is extremely low (2.2 per 10,000 women in this study). This study included two maternal deaths, resulting in a case fatality rate of 0.23% (95% CI 0.06-0.84%). In comparison, the World Health Organization’s (WHO) reported a case fatality rate between 1% and 13% in low- and middle-income countries.3 The overall rate of maternal death in women with previous CS undergoing TOLAC in the systematic review of Guise et al. was 0.004%,16, 38 while Landon et al. reported a rate of 0.02%.7 Perinatal outcome The rupture-related rate of perinatal death in women undergoing TOLAC is extremely low: 3.2 per 10,000 TOLACs in our study and ranging from 1.1 to 4.7 per 10,000 TOLACs in the literature.7, 40, 41 In a systematic review based on five cohort studies Guise et al. found a pooled rupture-related rate of perinatal deaths of 13 (range 6 to 29) per 10,000 TOLACs.38 The lowest Perinatal Mortality Rate (PMR) in France (0.8 per 10,000 TOLACs) is probably related to the different definition of uterine rupture limited to cases associated with SAMM, the highest rate in Sweden (13.2 per 10,000 TOLACs) probably due to the large number of missing data of neonatal outcome in the Swedish study. PMR in the other countries ranges between 1.2 and 4.1 per 10,000 TOLACs. The overall rate of neonatal asphyxia associated with complete uterine rupture in our study was 28% in the neonates who survived and 26% in the subgroup of babies born after uterine ruptures associated with TOLAC, resulting in a rate of 9.1 (95% CI 8.4 – 9.7) per 10,000 TOLACs. However, because of missing data, use of a proxy for asphyxia, and lack of information on long-term effects in this study, it is inappropriate to make firm conclusions on perinatal morbidity related to TOLAC. Rupture-related perinatal asphyxia reported in the literature ranges from 4.6 per 10,000 TOLACs (based on HIE)7, 42 to 15 and 40 per 10,000 TOLACS (based on umbilical artery pH<7.0).40, 43 Future research Obviously, outcome of complete uterine rupture is strongly related to the interval between the time of diagnosis and time of delivery of the neonate. Many authors refer to two observational studies of Leung et al. (n=116 uterine ruptures) and

Holmgren et al. (n=36 uterine ruptures) who both reported a diagnosis-to-delivery interval (DDI) of less than 18 minutes associated with umbilical cord pH levels above 7.0.44, 45 Holmgren et al. further reported that all neonates delivered within 30 minutes had uncomplicated long-term outcome. In this respect, the American College of Obstetricians and Gynecologists (ACOG) practice bulletin’s recommendation that TOLAC should be undertaken in facilities with staff immediately available to perform emergency CS, seems highly appropriate.11 Further studies are needed to increase evidence of the impact of organisational factors on outcome of complete uterine rupture related to TOLAC. These studies should include well-defined outcome parameters (stillbirth, early and late neonatal mortality, umbilical artery pH, long term morbidity), comparable groups (adjusted for gestational age, previous CS and TOLAC) and take into account in-house availability of staff to perform immediate CS, number of annual births per hospital and diagnosis-to-decision-to-delivery interval. Benchmarking population-based TOLAC rates opposed to uterine rupture rates reveals that countries with high TOLAC rates pay a higher price in uterine ruptures per TOLAC, whilst countries with low TOLAC rates pay a higher price of repeat CSs per VBAC. To evaluate whether one is preferable above the other, associated morbidity and mortality must be taken into consideration. Long-term maternal morbidities associated with CS and multiple scars to the uterus, such as abnormally invasive placenta (AIP) and peripartum hysterectomy, are equally relevant in determining optimal TOLAC rates. The NOSS study group, among others, demonstrated an increasing prevalence of AIP with increasing number of previous CSs.46 Considering TOLAC inevitably involves a trade-off between the rate of AIP (1/363 in women with two previous CS in the NOSS) and the rate of uterine rupture (1/294 in women undertaking TOLAC in our study). Once again, the results of this study emphasize that population-based rates of unnecessary primary caesarean sections be kept as low as possible.47

Conclusion

We observed a higher prevalence of complete uterine rupture per TOLAC in countries with relatively low previous CS and high TOLAC rates. Rupture-related rates of hysterectomy and perinatal mortality in women undergoing trial of labour after caesarean section are extremely low (2.2 and 3.2 per 10,000 TOLACs respectively). Future studies are needed to improve prediction of complete uterine rupture and management of TOLAC to further decrease the 10% rate of hysterectomy and perinatal death associated with complete uterine rupture.

Acknowledgements

Austria: We would like to acknowledge all obstetricians in Austria who reported the cases to AuOSS. Part of the study was sponsored by the Anniversary Fund of the Österreichische Nationalbank – ÖNB. Belgium: We would like to acknowledge all the B.OSS reporting clinicians who contributed data to the study; Marlies De Blaere and Virginie Van Leeuw who contributed in the acquisition of Belgian data; Yvon Englert, Myriam Hanssens, Hans Verstraelen who contributed in the design of the study and interpretation of data. B.OSS is funded by the College for Mother and Newborn, a consultative body of the Belgian Public Health Service. GV was funded by the Flemish Research Foundation (FWO) (2015-2016). Denmark: We would like to acknowledge all the NOSS reporting clinicians who contributed data to the study and Lone Krebs from the Danish NOSS organisation. Finland: Maija Jakobsson, Anna-Maija Tapper, Outi Palomäki, Kati Ojala, Nanneli Pallasmaa, and Maija-Riitta Ordén from the Finnish NOSS group France: We would like to acknowledge all the EPIMOMS obstetricians, midwives, and anesthetists who contributed to case identification and documentation in their hospital, and research assistants who collected the data, as well as members of the Epimoms Study Group. Germany: Sponsored by Qualitätsinitiative – Niedersächsischer Verein zur Förderung der Qualität im Gesundheitswesen e.V. until 2016. Now GerOSS is financed by the Center for Quality and Management in Health Care, Medical Association of Lower Saxony. The Netherlands: NethOSS board: Jos van Roosmalen, Thomas van den Akker, Kitty Bloemenkamp, Timme Schaap and Joost Zwart. We would like to acknowledge all clinicians reporting data to the LEMMoN study between 2004-2006 Sweden: We would like to acknowledge all the NOSS reporting clinicians who contributed data to the study and and Karin Gottvall and Karin Källen at the Medical Birth Registry United Kingdom: We would like to acknowledge all the UKOSS reporting clinicians who contributed data to the study.

Disclosure of interest

None of the authors have any conflict of interest to declare.

Contribution to authorship

GV, KB, SB, LC, CD, MG, MK, JLR, PL, WO, JV, JZ and KR contributed in the conception of the study and provided data. GV and KR designed the study. GV coordinated data collection, processing and analysis. GV, KB, SB, LC, CD, MG, MK, JLR, PL, WO, JV, JZ and KR contributed in the interpretation of the data. GV drafted the article. GV, KB, SB, LCCD, MG, MK, JVL, PL, WO, JV, JZ and KR critically revised the article and approved the final version. GV, KB, SB, LC, CD, MG, MK, JLR, PL, WO, JV, JZ and KR agree to be accountable for all aspects of the work. Authors are listed in alphabetical order between the first and last author.

Details of ethical approval

Each population-based study was approved individually by their Ethics Committees (Supporting Table S12). The international study was approved by the Medical Ethics Committee of the Ghent University Hospital in Belgium (23 March 2016; EC UZG 2016/0453).

Funding

GV was funded by the Research Foundation Flanders (FWO) (2015-2016).

References

1. Landon MB. Predicting uterine rupture in women undergoing trial of labor after prior cesarean delivery. Seminars in perinatology. 2010 Aug;34(4):267-71. 2. Schaap T, Bloemenkamp K, Deneux-Tharaux C, Knight M, Langhoff-Roos J, Sullivan E, et al. Defining definitions: a Delphi study to develop a core outcome set for conditions of severe maternal morbidity. BJOG : an international journal of obstetrics and gynaecology. 2017 Jul 29. 3. Hofmeyr GJ, Say L, Gulmezoglu AM. WHO systematic review of maternal mortality and morbidity: the prevalence of uterine rupture. BJOG : an international journal of obstetrics and gynaecology. 2005 Sep;112(9):1221-8. 4. Al-Zirqi I, Stray-Pedersen B, Forsen L, Daltveit AK, Vangen S. Uterine rupture: trends over 40 years. BJOG : an international journal of obstetrics and gynaecology. 2016 Apr;123(5):780-7. 5. Vilchez G, Nazeer S, Kumar K, Warren M, Dai J, Sokol RJ. Contemporary epidemiology and novel predictors of uterine rupture: a nationwide population-based study. Archives of gynecology and obstetrics. 2017 Sep 01. 6. Sachs BP, Kobelin C, Castro MA, Frigoletto F. The risks of lowering the cesarean-delivery rate. The New England journal of medicine. 1999 Jan 07;340(1):54-7. 7. Landon MB, Hauth JC, Leveno KJ, Spong CY, Leindecker S, Varner MW, et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. The New England journal of medicine. 2004 Dec 16;351(25):2581-9. 8. Al-Zirqi I, Stray-Pedersen B, Forsen L, Vangen S. Uterine rupture after previous caesarean section. BJOG : an international journal of obstetrics and gynaecology. 2010 Jun;117(7):809-20. 9. Guise JM, Denman MA, Emeis C, Marshall N, Walker M, Fu R, et al. Vaginal birth after cesarean: new insights on maternal and neonatal outcomes. Obstetrics and gynecology. 2010 Jun;115(6):1267-78. 10. Cragin EB. Conservatism in obstetrics. NY Med J. 1916;104:1-3. 11. ACOG Practice bulletin no. 115: Vaginal birth after previous cesarean delivery. Obstetrics and gynecology. 2010;116(2 Pt 1):450-63. 12. RCOG. Green-top Guideline No. 45. Birth After Previous Caesarean Birth. 2015 [Available from: https://www.rcog.org.uk/globalassets/documents/guidelines/gtg_45.pdf. 13. SOGC clinical practice guidelines. Guidelines for vaginal birth after previous caesarean birth. Number 155 (Replaces guideline Number 147), February 2005. International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics. 2005;89(3):319-31. 14. Sentilhes L, Vayssiere C, Beucher G, Deneux-Tharaux C, Deruelle P, Diemunsch P, et al. Delivery for women with a previous cesarean: guidelines for clinical practice from the French College of Gynecologists and Obstetricians (CNGOF). European journal of obstetrics, gynecology, and reproductive biology. 2013 Sep;170(1):25-32. 15. Reif P, Brezinka C, Fischer T, Husslein P, Lang U, Ramoni A, et al. Labour and Childbirth After Previous Caesarean Section: Recommendations of the Austrian Society of Obstetrics and Gynaecology (OEGGG). Geburtshilfe und Frauenheilkunde. 2016 Dec;76(12):1279-86. 16. National Institutes of Health Consensus Development Conference Statement vaginal birth after cesarean: new insights March 8-10, 2010. Seminars in perinatology. 2010 Oct;34(5):351-65. 17. Colmorn LB, Petersen KB, Jakobsson M, Lindqvist PG, Klungsoyr K, Kallen K, et al. The Nordic Obstetric Surveillance Study: a study of complete uterine rupture, abnormally invasive placenta, peripartum hysterectomy, and severe blood loss at delivery. Acta obstetricia et gynecologica Scandinavica. 2015 Jul;94(7):734-44. 18. Fitzpatrick KE, Kurinczuk JJ, Alfirevic Z, Spark P, Brocklehurst P, Knight M. Uterine rupture by intended mode of delivery in the UK: a national case-control study. PLoS medicine. 2012;9(3):e1001184. 19. Knight M. The International Network of Obstetric Survey Systems (INOSS): benefits of multi-country studies of severe and uncommon maternal morbidities. Acta obstetricia et gynecologica Scandinavica. 2014 Feb;93(2):127-31.

20. Deneux-Tharaux CB-C, Marie-Helene; for the EPIMOMS study group. Severe acute maternal morbidity in France: the epimoms population-based study. American journal of obstetrics and gynecology. 2017;216(1, Supplement):S345-S6. 21. Executive summary: Neonatal encephalopathy and neurologic outcome, second edition. Report of the American College of Obstetricians and Gynecologists' Task Force on Neonatal Encephalopathy. Obstetrics and gynecology. 2014 Apr;123(4):896-901. 22. Knight M, Kurinczuk JJ, Tuffnell D, Brocklehurst P. The UK Obstetric Surveillance System for rare disorders of pregnancy. BJOG : an international journal of obstetrics and gynaecology. 2005 Mar;112(3):263-5. 23. Zwart JJ, Richters JM, Ory F, de Vries JI, Bloemenkamp KW, van Roosmalen J. Uterine rupture in The Netherlands: a nationwide population-based cohort study. BJOG : an international journal of obstetrics and gynaecology. 2009 Jul;116(8):1069-78; discussion 78-80. 24. Vandenberghe G, De Blaere M, Van Leeuw V, Roelens K, Englert Y, Hanssens M, et al. Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2016 May 17;6(5):e010415. 25. Colmorn LB, Langhoff-Roos J, Jakobsson M, Tapper AM, Gissler M, Lindqvist PG, et al. National Rates of Uterine Rupture are not Associated with Rates of Previous Caesarean Delivery: Results from the Nordic Obstetric Surveillance Study. Paediatric and perinatal epidemiology. 2017 May;31(3):176-82. 26. Zentrum für Qualität und Management im Gesundheitswesen. Geburtshilfe Jahresstatistik 2010-2013: Niedersächsische Perinatalerhebung. Available online at: https://www.aekn.de/fileadmin/media/Downloadcenter/ZQ/Projekt_Perinatalerhebung/Statistiken/pe gesamtstatistik2010.pdf 27. Knight HE, Gurol-Urganci I, van der Meulen JH, Mahmood TA, Richmond DH, Dougall A, et al. Vaginal birth after caesarean section: a cohort study investigating factors associated with its uptake and success. BJOG : an international journal of obstetrics and gynaecology. 2014 Jan;121(2):183-92. 28. Kwee A, Bots ML, Visser GH, Bruinse HW. Uterine rupture and its complications in the Netherlands: a prospective study. European journal of obstetrics, gynecology, and reproductive biology. 2006 Sep- Oct;128(1-2):257-61. 29. Epidemiologic Research Unit on Perinatal Health and Women and Children's health I-U, Paris. French National Perinatal Survey 2010. 2011. 30. Institut für klinische Epidemiologie. Geburtenregister Österreich. Bericht Geburtsjahr 2011-2013. 31. Zhao Y, Zhang J, Hukkelhoven C, Offerhaus P, Zwart J, Jonge A, et al. Modest Rise in Caesarean Section from 2000-2010: The Dutch Experience. PloS one. 2016;11(5):e0155565. 32. Blondel B, Lelong N, Kermarrec M, Goffinet F. Trends in perinatal health in France from 1995 to 2010. Results from the French National Perinatal Surveys. Journal de gynecologie, obstetrique et biologie de la reproduction. 2012 Jun;41(4):e1-e15. 33. Khan K. The CROWN Initiative: journal editors invite researchers to develop core outcomes in women's health. BJOG : an international journal of obstetrics and gynaecology. 2016 Sep;123 Suppl 3:103-4. 34. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp. 35. Brown LD, Cai T, Das Gupta A. Interval Estimation for a binomial proportion. Statistical Science. 2001;16(2):101-33. 36. Thisted DL, Mortensen LH, Hvidman L, Rasmussen SC, Larsen T, Krebs L. Use of ICD-10 codes to monitor uterine rupture: validation of a national birth registry. European journal of obstetrics, gynecology, and reproductive biology. 2014 Feb;173:23-8. 37. Al-Zirqi I, Stray-Pedersen B, Forsen L, Daltveit AK, Vangen S. Validation study of uterine rupture registration in the Medical Birth Registry of Norway. Acta obstetricia et gynecologica Scandinavica. 2013 Sep;92(9):1086-93. 38. Guise JM, McDonagh MS, Osterweil P, Nygren P, Chan BK, Helfand M. Systematic review of the incidence and consequences of uterine rupture in women with previous caesarean section. BMJ (Clinical research ed). 2004 Jul 03;329(7456):19-25. 39. Yee LM, Liu LY, Grobman WA. Obstetrician call schedule and obstetric outcomes among women eligible for a trial of labor after cesarean. American journal of obstetrics and gynecology. 2017 Jan;216(1):75.e1-.e6.

40. Chauhan SP, Martin JN, Jr., Henrichs CE, Morrison JC, Magann EF. Maternal and perinatal complications with uterine rupture in 142,075 patients who attempted vaginal birth after cesarean delivery: A review of the literature. American journal of obstetrics and gynecology. 2003 Aug;189(2):408-17. 41. Smith GC, Pell JP, Pasupathy D, Dobbie R. Factors predisposing to perinatal death related to uterine rupture during attempted vaginal birth after caesarean section: retrospective cohort study. BMJ (Clinical research ed). 2004 Aug 14;329(7462):375. 42. Spong CY, Landon MB, Gilbert S, Rouse DJ, Leveno KJ, Varner MW, et al. Risk of uterine rupture and adverse perinatal outcome at term after cesarean delivery. Obstetrics and gynecology. 2007 Oct;110(4):801-7. 43. Bujold E, Gauthier RJ. Neonatal morbidity associated with uterine rupture: what are the risk factors? American journal of obstetrics and gynecology. 2002 Feb;186(2):311-4. 44. Leung AS, Leung EK, Paul RH. Uterine rupture after previous cesarean delivery: maternal and fetal consequences. American journal of obstetrics and gynecology. 1993 Oct;169(4):945-50. 45. Holmgren C, Scott JR, Porter TF, Esplin MS, Bardsley T. Uterine rupture with attempted vaginal birth after cesarean delivery: decision-to-delivery time and neonatal outcome. Obstetrics and gynecology. 2012 Apr;119(4):725-31. 46. Thurn L, Lindqvist PG, Jakobsson M, Colmorn LB, Klungsoyr K, Bjarnadottir RI, et al. Abnormally invasive placenta-prevalence, risk factors and antenatal suspicion: results from a large population-based pregnancy cohort study in the Nordic countries. BJOG : an international journal of obstetrics and gynaecology. 2016 Jul;123(8):1348-55. 47. Obstetric care consensus no. 1: safe prevention of the primary cesarean delivery. Obstetrics and gynecology. 2014 Mar;123(3):693-711.

THE BELGIAN OBSTETRIC SURVEILLANCE SYSTEM TO MONITOR SEVERE MATERNAL MORBIDITY Griet Vandenberghe1, Kristien Roelens1, Virginie Van Leeuw2,3, Yvon Englert4,5, Myriam Hanssens6, Hans Verstraelen1.

1. Department of Obstetrics & Gynaecology, Ghent University Hospital, 9000 Ghent, Belgium. 2. Perinatal Epidemiology Center (Centre d’Épidémiologie Périnatale, CEpiP), 1070 Brussels, Belgium 3. School of Public Health, Université Libre de Bruxelles (ULB), 1050 Brussels, Belgium. 4. Perinatal Epidemiology Center (Centre d’Épidémiologie Périnatale, CEpiP), 1070 Brussels, Belgium 5. Faculty of Medicine, Research Laboratory on Human Reproduction, Université Libre de Bruxelles (ULB), 1050 Brussels, Belgium. 6. Department of Obstetrics & Gynaecology, Leuven University Hospital, 300 Leuven, Belgium. Key Words: Severe maternal morbidity, maternal near miss, quality of care, obstetric health care, obstetric surveillance system, population-based.

ABSTRACT

Background In 2011 the Belgian Obstetric Surveillance System (B.OSS) was set up to monitor severe maternal morbidity in Belgium. Aim The aim of B.OSS is to get an accurate picture of the obstetric complications under investigation and secondly, to improve the quality and safety of obstetric care in Belgium by practical recommendations based on the results. Methodology Data are obtained through prospective active collection of cases by a monthly call according to the principle of nothing-to-report, along with data collection forms that confirm the diagnosis and gather detailed information. Data-collection occurs web- based since August 2013 through www.b-oss.be. Results B.OSS achieves excellent participation rates and response rates. The results of the first registration round are gradually brought out by means of scientific publications and presentations, biennial reports, newsletters and the website. The international comparison of results within the International Network of Obstetric Survey Systems (INOSS) gives important added value. No alternative mandatory data sources are appropriate to check for underreporting. Conclusions B.OSS is successful in monitoring severe maternal morbidity thanks to the willingness of the Belgian OB-GYNs. The results of the first studies suggest the need to develop nationally adopted guidelines. Furthermore, the results invite to critically evaluate the current organisation of obstetric health care in Belgium. B.OSS aims to monitor the impact on patient safety in future surveys when guidelines and recommendations are put into practice.

Background: Why did we develop the Belgian Obstetric Surveillance System?

The key to improve is to measure. Measurements can reveal weaknesses and weaknesses are opportunities to improve. Measurements allow to discover trends and to benchmark results within a group. Measurements support policy-makers in their decisions to change. Subsequent measurements then monitor the results when interventions were implemented. This wisdom is generally true and widely used: in pursuit of a long tradition in the business world, surveys and registers have become common practice in health care. In measuring the quality of obstetric health care, since two decades the focus has shifted from maternal mortality as major indicator to severe maternal morbidity (SMM) and maternal near-miss (MNM) (Stones et al., 1991; Say et al., 2009): the last- but-one level in a continuum from uncomplicated pregnancy to maternal death (Figure 1). The study of women who experienced and survived a severe obstetric complication has advantages compared to the study of maternal deaths: an adequate number of cases for analysis is achieved more rapidly, the women can be interrogated and clinicians may provide information more easily, the comparison with cases of maternal death can reveal risk factors associated with mortality and recommendations will be more generalisable (Say et al., 2009; Knight et al., 2014). Severe obstetric complications are studied in a topic-based approach by the Obstetric Surveillance Systems: nationwide surveys that identify and study near-miss events and rare diseases of pregnancy. The United Kingdom Obstetric Surveillance System (UKOSS) as pioneer completed around three dozen studies since its initiation in 2006, which resulted in improved patient safety (Knight et al., 2005; Knight and Lindquist, 2013). The methodology of UKOSS served as a template to several high-income countries united in the International Network of Obstetric Survey Systems (INOSS), having the mission to co-operate, share information and enable cross-national comparisons and collaborative studies of very rare conditions (Knight, 2014). Health care in Belgium is considered of high quality and easily accessible (Björnberg, 2017). Obstetric care is mainly specialist-led and hospital-based. Belgium has a high number of small volume maternity units (57% have less than 1000 annual births), besides 17 tertiary referral centres providing Neonatal and Maternal Intensive Care (NIC and MIC). The number of deliveries in Belgium is around 124,000 per year (Agentschap Zorg en Gezondheid, 2017; CEpiP, 2017). There is a relatively high rate of induction of labour (27.1% in 2015)(Euro-Peristat, 2013) and a steadily increasing caesarean section rate (20.7% in 2015)(Devlieger et al., 2015; Leroy et al., 2015; Van Leeuw et al., 2015), still below the European median rate (25.2% in 2010)(Euro- Peristat, 2013). Maternal and perinatal data registration in Belgium is organised in two parallel institutes for Perinatal Registry: Studiecentrum voor Perinatale Epidemiologie (SPE) in Flanders and Centre d’Epidémiologie Périnatale (CEpiP) in Brussels and Wallonia

(Agentschap Zorg en Gezondheid, 2017; CEpiP, 2017). Both register on a mandatory basis, covering 100% of births in Belgian maternity units and home births. A selected set of maternal and perinatal data are recorded by the OB-GYN, midwife and neonatologist immediately after birth. At this point, Belgium has no structured national system to survey the cases of maternal deaths, unlike most European countries. In 2011 the Belgian Obstetric Surveillance System (B.OSS) was set up to register and analyse severe maternal morbidity. The aim of B.OSS is to get an accurate picture of the obstetric complications under investigation and secondly, to improve the quality and safety of obstetric care in Belgium by practical recommendations based on the findings of B.OSS. The aim of this manuscript is to describe the development, methodology, first achievements and future plans of B.OSS.

Methodology: How does the Belgian Obstetric Surveillance System work?

The Belgian Obstetric Surveillance System, briefly B.OSS, is initiated with support of the College of Physicians for Mother and Newborn, a consultative body of the Federal Public Service of Health. B.OSS is endorsed by the two professional associations for OB-GYNs (Vlaamse Vereniging voor Obstetrie en Gynecologie (VVOG) and Groupement des Gynécologues Obstétriciens de Langue Française de Belgique (GGOLFB)), and by the perinatal registries SPE and CEPiP. The College operates as the steering committee of B.OSS. In 2017, a formal scientific board is constituted with representatives of SPE, CEpiP, the Belgian Health Care Knowledge Centre (KCE), the Scientific Institute of Public Health (WIV-ISP) and the College. Daily reporting and data collection tasks are carried out by two cooperating teams: one in Flanders, another in Brussels and Wallonia.

B.OSS has adopted the methodology for case reporting developed by the UK Obstetric Surveillance System (UKOSS)(Knight et al., 2005). An appointed contact person (OB- GYN or senior midwife) in each participating maternity unit is invited by monthly mailing to report a selected number of rare obstetric complications that occurred in the preceding month or alternatively to state that there was ‘nothing to report’, as is mostly the case. In the event of a case being reported, the contact person is asked to complete an extensive data collection form. In case of incomplete reporting, the contact person is encouraged repeatedly by email and phone to provide the missing data.

Initially, the data of the reported cases were obtained through use of a standardised form, filled out electronically or on hard copy. Web-based data-collection was gradually introduced following the launch of the B.OSS website (www.b-oss.be) in August 2013, facilitating monthly reporting and completion of data collection forms online. Monthly emails calling to report for the previous month are generated automatically, with reminders for missing reporting forms and incomplete data collection forms. Restricted access to the website is provided to the appointed contact person, one per maternity unit, via a personal login. They have access to the reporting forms and data collection forms of their maternity unit. Data protection is secured by the use of anonymous hash codes, replacing person-identifiable information such as the woman’s name, date of birth or hospital number.

The data collection forms seek information on maternal characteristics, medical, surgical and obstetric history, details of the index pregnancy, details of the delivery, the circumstances of the adverse event, its management and the outcome for mother and neonate. The data are retrieved in retrospect from the woman’s case notes.

The collected data are coded and exported as comma-separated values (csv) files, then cleaned and processed into operational SPSS data files. Data-analysis is performed using IBM SPSS statistics (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp), EpiTools epidemiological calculators (Sergeant, ESG, 2017. Epitools epidemiological calculators. Ausvet Pty Ltd. Available at: http://epitools.ausvet.com.au) and MedCalc for Windows, version 15.0 Software (MedCalc Software, Ostend, Belgium). The prevalence of the obstetric events targeted is estimated using as denominator the total number of deliveries in Belgium during the survey period, corrected for the maternity units that did not participate. Reference data are obtained from the perinatal registries (SPE and CEpiP) when available.

The B.OSS methodology was approved by the Medical Ethics Committee of Ghent University Hospital (EC UZG 2012/734; B670201215359) and by the Medical Ethics Committee of the Erasme University Hospital, Brussels (EC ULB 2012/111; B406201213660) at the beginning. The Medical Ethics Committee of the Ghent University Hospital became Central Ethics Committee in 2015 (EC UZG 2015/1470;

B670201526875). The women eligible for inclusion are informed by their OB-GYN and offered an information letter enabling them to opt-out. Confidentiality is guaranteed for mother, provider and hospital. Person-identifiable information is eliminated from data-analysis.

Results: What is the output of the Belgian Obstetric Surveillance System?

At start, 110 of 113 Belgian maternity units (97.3 %) formally agreed to participate in B.OSS; the three remaining maternity units have been recruited since. The number of maternity units dropped through the years as a result of merging and closure, with 106 of 107 (99%) participating in January 2017. B.OSS achieved an excellent response rate (the number of monthly reporting forms returned) at start in 2012-2013 (98.9%), which slightly dropped in the following years (Figure 2).

B.OSS completed a two-year study of uterine rupture (2012-2013) and peripartum hysterectomy and/or arterial embolisation of the uterine arteries (2012-2013) and a three-year study of eclampsia (2012-2014). Currently B.OSS is conducting a four-year study of antenatal pulmonary embolism (APE) (2015-2018) and is participating in the INOSS international studies of spontaneous haemoperitoneum in pregnancy (SHiP) (2015-2017) and anaphylaxis in pregnancy (2016-2018).

The results of the first registration round are brought out gradually. The prevalence, risk factors, management and outcome of the obstetric complications under investigation are communicated towards obstetric and other health care providers by means of scientific publications and presentations.(Vandenberghe et al., 2016; Vandenberghe et al., 2017) A newsletter is sent out every six months, providing intermediate results and practical information, besides asking for feedback and ideas for improvement. A biennial report summarising the achievements of B.OSS in the previous two years, is distributed to all OB-GYNs and maternity units. The website (http://www.b-oss.be) provides information on the background and methodology of B.OSS. Study protocols, data collection forms, the patient information letter, scientific publications and presentations of B.OSS-results are freely accessible. The contact persons have restricted access to the newsletters, biennial reports and a continuously updated response rate.

Despite monthly emails and frequent reminders, underreporting of cases cannot be excluded. Underreporting is attributed to the non-mandatory participation of busy clinicians and to the case’s nature: severe maternal complications and near miss cases, maybe evoked by some degree of concern on the quality of care. We consulted two national data sources, collecting medical data on a mandatory basis, that could possibly serve as a control for underreporting for the data recorded by B.OSS. Firstly, the Belgian hospital discharge register (Minimale Ziekenhuis Gegevens (MZG), Résumé Hospitalier Minimum (RHM)), which used the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) (CDC, 2013) codes up to January 2015. From this data source we derived the total number of peripartum hysterectomy, eclampsia and uterine rupture, indicated by clinicians from January until December 2012. The number of arterial embolisations derived from the hospital discharge register was not accurate, due to the lack of a specific ICD-9-cm code indicating the embolisation of the uterine arteries. As shown in Table I, all numbers from the hospital discharge register exceeded the numbers derived by B.OSS during the same period, even twice (eclampsia) or three times (uterine rupture). This is in line with the findings of true validation studies, such as a study from Denmark (Thisted et al., 2014) in which only 60.4% of the women registered with an ICD-10 code for uterine rupture during labour actually had a uterine rupture. Likewise, a validity study from the United States demonstrated a high positive predictive value (PPV) for ICD-9-cm codes indicating procedures, such as hysterectomy (PPV 100%, 95% CI 86-100%) and a low PPV for ICD-9-cm codes indicating conditions that require the interpretation of clinical information, such as eclampsia (PPV 55%, 33-76%) (Sigakis et al., 2016). We trust that the methodology of the obstetric surveillance system, through monthly reporting and detailed data collection forms, should be more precise than the hospital discharge register. The second data source theoretically allowing control for underreporting is the Inter Health Insurance Funds Agency (het InterMutualistisch Agentschap (IMA), l’Agence InterMutualiste (AIM))(IMA-AIM, 2017). The IMA-AIM records administrative and billing data of all Belgian inhabitants affiliated to the seven Belgian Health Insurance funds. The database of IMA-AIM includes information of every medical act that was carried out and reimbursed in Belgium based on a specific nomenclature code, such as the timing and location of the medical act and (coded) patient data. From this data source we derived the total number of hysterectomies and arterial embolisations carried out from January 2012 until December 2013 within 4 weeks following a vaginal birth or caesarean section. In Table II is shown that the number of peripartum hysterectomies recorded by IMA-AIM was almost equal, while the number of arterial embolisations largely exceeded the number reported by B.OSS. Possible explanations for this discrepancy are the accuracy of the nomenclature code, which is not specific for embolisation of the uterine arteries, and the hypothesis that in some cases embolisation catheters were placed prophylactically before planned caesarean

sections with expected high risk of obstetric haemorrhage while the actual arterial embolisation did not need to be carried out. This data-source could not provide data of uterine ruptures or eclampsia, because those complications do not involve a specific medical act with traceable nomenclature code. Hence, the IMA-AIM database may only be of use to validate surveys of procedures indicative of severe maternal morbidity.

The use of similar methods and data collection forms based on the UKOSS, enables the comparison of the Belgian data with the results of studies performed by obstetric surveillance systems in other countries. The international comparison of prevalences, risk factors, management and outcome adds important value on top of the conclusions that can be made based on national data. Within the INOSS network, we conducted an international comparison study of uterine rupture between nine countries, resulting in the largest prospective population-based report of uterine rupture in high-income countries (unpublished data, paper submitted October 2017). This study reports a prevalence of uterine rupture in women with previous caesarean section of 22 (95% CI 21-24) per 10,000 deliveries, ranging from 8 (95% CI 6-11) to 68 (95% CI 56-83) per 10,000 deliveries. This wide range in prevalence between countries appears to be, amongst others, a consequence of the range in trial of labour after caesarean section (TOLAC) rates between countries (from 47% to 72%). Benchmarking population-based TOLAC rates opposed to uterine rupture rates reveals that countries with high TOLAC rates pay a slightly higher price in uterine ruptures per TOLAC. This international comparison study suggests a need to critically appraise TOLAC policies in each country: first, the selection of appropriate candidates and second, the management of labour in women who decided to undergo TOLAC. With regard to Belgium, the prevalence of uterine rupture in women with previous caesarean section (21 (95% CI 16-27) per 10,000 deliveries) is in the lower-middle of prevalences, while the TOLAC rate (percentage of women with previous caesarean section who underwent TOLAC) is the second lowest within this INOSS-group (47.2%). Furthermore, Belgium has the second lowest peripartum hysterectomy rate (3.1 (95% CI 2.6-4.0) per 10,000 deliveries) in an international comparison study between ten INOSS countries (range 3.0 to 7.8 per 10,000 deliveries) (unpublished data). And the prevalence of eclampsia applying the UKOSS criteria (Schaap et al., 2014) is 1.6 (95% CI 1.2-2.1) per 10,000 deliveries in Belgium (unpublished data), which is very low compared to the prevalence in the UK (2.7 (95% CI 2.4-3.2) per 10,000 deliveries) (Knight, 2007) and the Netherlands (5.4 (95% CI 4.6-6.2) per 10,000 deliveries) (Schaap et al., 2014).

The B.OSS findings give rise to practical recommendations: which changes should be made in clinical practice to improve the quality and safety of obstetric care? Based on the study of uterine rupture,(Vandenberghe et al., 2016) we believe that obstetric care in Belgium would benefit from a nationally adopted guideline on trial of labour after caesarean section (TOLAC). This guideline should result from the comparison and fusion of the existing international guidelines (SOGC, 2005; Sentilhes et al., 2013; RCOG, 2015; ACOG, 2017) and have the consent of the professional associations for OB-GYNs, the professional associations for midwifes and the College of Physicians for Mother and Newborn. The guideline should be developed to promote safe TOLAC in appropriate candidates on the one hand, in order to enhance the low current TOLAC rate in Belgium. On the other hand, the guideline should prevent unsafe procedures, which have been observed among the uterine rupture cases in the B.OSS study. Furthermore, the guideline should include recommendations on the mode of delivery in women with other uterine surgical procedures (such as myomectomy, hysteroscopic resection of a uterine septum or salpingectomy), in whom a management similar to women with previous caesarean section is justified. A follow-up registration of uterine rupture by B.OSS should be planned over time, to monitor the effect of and compliance with the national guideline. Based on the study of peripartum hysterectomy and arterial embolisation, (Vandenberghe et al., 2017) we believe that the management of obstetric haemorrhage in Belgium could be further improved if all women with abnormal placentation were managed in a tertiary referral centre. Therefore, we need to enhance the awareness of risk factors for abnormal invasive placenta (AIP) on the part of clinicians in order to improve the identification of women at risk and increase the antenatal transfer to tertiary referral centres in case of suspicion. This can be achieved by, besides the development of a new national guideline, the propagation of the B.OSS publication accompanied by clear instructions and ultrasound criteria for AIP, through the websites of B.OSS and the professional associations for OB-GYNs and through presentations on regional and national conferences. More influential would be a restructuring of (obstetric) health care in Belgium by the government, making referral to tertiary referral centres obviously and compulsory in case of certain conditions, such as AIP. A follow-up registration by B.OSS should include all women with massive obstetric haemorrhage, using the INOSS international consensus definition,(Schaap et al., 2017) and all women successfully managed with other second-line measures (such as intra-uterine balloon tamponade), to get the overall picture of postpartum haemorrhage in Belgium. Furthermore, a case-by-case in-depth analysis would be interesting to reveal whether the hysterectomies and arterial embolisations performed in this study were appropriate or preventable. Based on the current findings, we hypothesise that hysterectomies may be performed more easily

in smaller maternity units, while there may be an overuse of arterial embolisations when this service is easily accessible, without first resorting to second-line measures. Based on the study of eclampsia, we believe that there is scope for improvement by increasing the awareness of health care providers, mainly general practitioners and emergency room doctors besides OB-GYNs and midwifes, of the use of magnesium sulphate as the first choice treatment in case of an eclamptic fit. In current B.OSS study magnesium sulphate was administered as first treatment in only 54% of the women. This calls for another national guideline.

Discussion: How can we further improve the Belgian Obstetric Surveillance System?

Although relying on the willingness of busy clinicians, B.OSS achieves an excellent participation rate and response rate. The success of B.OSS demonstrates the hunger for information on severe maternal morbidity on a national basis. Rates gradually dropped in the latest years however, blamed partly on the extremely rare conditions currently under investigation. To further motivate clinicians and boost the response rate, we suggest several action points. Firstly, starting new studies that investigate more common conditions that are relevant for every obstetrician. Secondly, proclaiming the results of the first B.OSS studies in presentations and publications and proclaiming the impact on quality of care, when recommendations have been implemented over time. Thirdly, rewarding the maternity units that register in a pay- for-quality project or making the B.OSS registration mandatory by the government. Thanks to its methodology, that is the prospective collection of cases through an active search on a monthly basis according to the principle of nothing-to-report, the number of missed cases (false negatives) can be reduced to a great extent. Besides, the number of false positives is limited to a minimum thanks to the detailed data collection forms that confirm the diagnosis. However, there is currently no appropriate way to check the data of B.OSS for completeness and correctness. The validity of B.OSS data could be further improved by involving more than one clinician per maternity unit, such as a senior midwife, an obstetric anaesthesist or an intensive care specialist. This would follow the lead of the UKOSS, where four clinicians are nominated in each hospital (Knight et al., 2005). The launch of the B.OSS website (http://www.b-oss.be) substantially reduced the workload of the research team and contact persons by enabling web-based data collection. For example, the monthly reporting is now easily done in three clicks by an email link. However, the commitment to the B.OSS project does involve additional work to contact persons on top of their clinical practice. Moreover, clinicians are regularly approached to participate in all kind of surveys. To further improve the workload, attention must be paid to keep data collection forms of future surveys

more concise and clear. Furthermore, future data collections will adhere to the INOSS Core Dataset (ICoD), a standardized minimum dataset developed by the International Network of Obstetric Survey System, to enhance the comparison of data between different countries. Over time, it would be most expedient if B.OSS could join the federal project Healthdata.be (WIV-ISP), a platform developed to reduce the administrative burden of clinicians. The Healthdata-platform aims to standardise and harmonise the registration and storage of health data, while absolute confidentiality is guaranteed. It enables to gather health data in a standardised way from different health care institutions and health care providers using many different medical record systems, by implementing a specifically developed software program. Since start in 2014, they succeeded to successfully integrate 42 registries into the platform. We await further development and results of Healthdata, to assess whether the B.OSS data collection could be (partially) covered by this platform.

The B.OSS surveys give an overall picture of the severe obstetric complications under surveillance, by an epidemiological description of prevalence, risk factors, management and outcome for mother and child. They do not allow to make firm conclusions on the quality of care provided to individual women included in the surveys. This requires a case-by-case in-depth analysis based on the full medical records, thereby assessing the management against relevant guidelines in order to make a judgement whether there is evidence of quality of care concerns. A structured review of near-miss cases by a multidisciplinary team of appointed experts is done on an annual basis in the UK since 2014, in a programme additional to the Confidential Enquiries into Maternal Deaths (CEMD) and the UKOSS, the so-called Confidential Enquiries into Maternal Morbidities (CEMM) (Knight et al., 2014). To set up audit studies in addition to B.OSS surveys, we first need to develop the practice of performing confidential enquiries in Belgium, and preferably have drafted nationally adopted guidelines against which to assess the quality of care.

Belgium can no longer lag behind its neighbouring high-income countries and other even low-and middle-income countries which have a well-established Confidential Enquiries into Maternal Deaths (CEMD) since many years. Now that B.OSS has a firm position, supported by the great majority of Belgian OB-GYNs, it is the right time to add maternal mortality to the project. There is clearly a need for a structured maternal mortality register in Belgium, since the B.OSS research team was asked by several clinicians where to report a case of maternal mortality. The added-value of a CEMD is obviously demonstrated by countries as UK and the Netherlands, where still half of cases is shown to be preventable and due to issues related to quality of care.

The College of Physicians for Mother and Newborn is currently taking the lead and appointed a working group to organise a structured Belgian CEMD.

Conclusion

The Belgian Obstetric Surveillance System has become a successful institute to monitor severe maternal morbidity in Belgium. The excellent participation rate and response rate demonstrates the willingness of Belgian OB-GYNs and their hunger for information on obstetric complications on a national basis. The results of the first studies suggest the need to develop nationally adopted guidelines. Furthermore, the results invite to critically evaluate the current organisation of obstetric health care in Belgium. B.OSS aims to monitor the impact on patient safety in future surveys when guidelines and recommendations are put into practice.

Acknowledgements

We would like to thank all the Belgian maternity units and in particular, the B.OSS contact persons in each unit. The achievements of B.OSS are made possible thanks to their voluntary participation. We would like to acknowledge Fatima Bercha, Marlies De Blaere, Letica Dias Nunes, Marine Guisset, Annejo Huybrechts, Ann Langedock, Charlotte Leroy and Caroline Roelens for their contributions to B.OSS.

Funding

B.OSS is funded by the College of Physicians for Mother and Newborn, a consultative body of the Belgian Public Health Service. GV was funded by the Flemish Research Foundation (FWO) (2015-2016).

Disclosure of interest

None of the authors have any conflict of interest to declare.

References

1. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 184: Vaginal Birth After Cesarean Delivery. Obstet Gynecol. 2017;130:e217-e233. 2. Agentschap Zorg en Gezondheid. Belangrijkste trends in geboorte en bevalling. Retrieved October, 2017, from https://www.zorg-en- gezondheid.be/sites/default/files/atoms/files/Evaluatierapport%20SPE%202015.pdf. 3. Björnberg A (2017). Euro Health Consumer Index 2016, Health Consumer Powerhouse Ltd., 2017. 4. Centers for Disease Control and Prevention. International Classification of Diseases, Ninth Revision, Clinical Modificiation (ICD-9-CM). 18 June 2013. Retrieved 15 October 2017, from https://www.cdc.gov/nchs/icd/icd9cm.htm. 5. Centre d'Épidémiologie Périnatale. CEpiP - Périnatalité. Retrieved October 2017, from http://www.cepip.be/. 6. Devlieger R, Martens E, Martens G et al. (2015). Perinatale activiteiten in Vlaanderen 2015. 7. Euro-Peristat project with SCPE and Eurocat. European Health Report. The health and care of pregnant women and babies in Europe in 2010. Euro-Peristat. 2013. Retrieved October 2017, from http://www.europeristat.com/images/doc/EPHR2010_w_disclaimer.pdf. 8. Het InterMutualistisch Agentschap - L'Agence InterMutualiste. IMA-AIM. 2017. Retrieved October 2017, from http://www.aim-ima.be/. 9. Knight M. Eclampsia in the United Kingdom 2005. Bjog. 2007;114:1072-1078. 10. Knight M. The International Network of Obstetric Survey Systems (INOSS): benefits of multi-country studies of severe and uncommon maternal morbidities. Acta Obstet Gynecol Scand. 2014;93:127-131. 11. Knight M, Kurinczuk JJ, Tuffnell D et al. The UK Obstetric Surveillance System for rare disorders of pregnancy. Bjog. 2005;112:263-265. 12. Knight M, Lewis G, Acosta CD et al. Maternal near-miss case reviews: the UK approach. Bjog. 2014;121 Suppl 4:112-116. 13. Knight M and Lindquist A. The UK Obstetric Surveillance System: impact on patient safety. Best Pract Res Clin Obstet Gynaecol. 2013;27:621-630. 14. Leroy C, Van Leeuw V, Englert Y et al. (2015). Santé périnatale en Wallonie. 15. Royal College of Obstetricians and Gynaecologists. RCOG. Green-top Guideline No. 45. Birth After Previous Caesarean Birth. 2015. Retrieved October 2017, from https://www.rcog.org.uk/globalassets/documents/guidelines/gtg_45.pdf. 16. Say L, Souza JP and Pattinson RC. Maternal near miss - towards a standard tool for monitoring quality of maternal health care. Best Pract Res Clin Obstet Gynaecol. 2009;23:287-296. 17. Schaap T, Bloemenkamp K, Deneux-Tharaux C et al. Defining definitions: a Delphi study to develop a core outcome set for conditions of severe maternal morbidity. Bjog. 2017; 18. Schaap TP, Knight M, Zwart JJ et al. Eclampsia, a comparison within the International Network of Obstetric Survey Systems. Bjog. 2014;121:1521-1528. 19. Sentilhes L, Vayssiere C, Beucher G et al. Delivery for women with a previous cesarean: guidelines for clinical practice from the French College of Gynecologists and Obstetricians (CNGOF). Eur J Obstet Gynecol Reprod Biol. 2013;170:25-32. 20. Sigakis MJ, Leffert LR, Mirzakhani H et al. The Validity of Discharge Billing Codes Reflecting Severe Maternal Morbidity. Anesth Analg. 2016;123:731-738. 21. Society of Obstetricians and Gynaecologists of Canada. SOGC. Guidelines for vaginal birth after previous caesarean birth. Number 155. Int J Gynaecol Obstet. 2005;89:319-331. 22. Stones W, Lim W, Al-Azzawi F et al. An investigation of maternal morbidity with identification of life- threatening 'near miss' episodes. Health Trends. 1991;23:13-15. 23. Thisted DL, Mortensen LH, Hvidman L et al. Use of ICD-10 codes to monitor uterine rupture: validation of a national birth registry. Eur J Obstet Gynecol Reprod Biol. 2014;173:23-28. 24. Van Leeuw V, Leroy C, Englert Y et al. (2015). Santé périnatale en Région bruxelloise. 25. Vandenberghe G, De Blaere M, Van Leeuw V et al. Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance System. BMJ Open. 2016;6:e010415.

26. Vandenberghe G, Guisset M, Janssens I et al. A nationwide population-based cohort study of peripartum hysterectomy and arterial embolisation in Belgium: results from the Belgian Obstetric Surveillance System. BMJ Open. 2017;7:e016208. 27. Healthdata.be. A short introduction. WIV-ISP. Retrieved October 2017, from https://healthdata.wiv- isp.be/en/about-healthdatabe.

Legend of Tables

Table I. Number of hospital stays for pregnancy, delivery or abortion, with report of severe obstetric complication in 2012 by the Belgian Hospital Discharge Register; compared to the number registered by the Belgian Obstetric Surveillance System (B.OSS). Table II. Number of peripartum hysterectomies and peripartum arterial embolisations in 2012-2013 recorded by IMA-AIM; compared to the number registered by the Belgian Obstetric Surveillance System (B.OSS).

Legend of Figures

Figure 1. The continuum from uncomplicated pregnancies to maternal mortality. Figure 2. Yearly response rate of the Belgian Obstetric Surveillance System in percentage.

Table I. Number of hospital stays for pregnancy, delivery or abortion, with report of severe obstetric complication in 2012 by the Belgian Hospital Discharge Register; compared to the number registered by the Belgian Obstetric Surveillance System (B.OSS). Peripartum Peripartum Eclampsia$ Uterine Severe obstetric hysterectomy embolisation§ rupture# complication * n n n n Belgian Hospital Discharge Register 57 24 54 116 (MZG, RHM)

B.OSS 44 53 24 41

* ICD-9-CM codes: 68.31,68.39, 68.41, 68.49, 68.51,68.59, 68.61, 68.69, 68.71, 68.79, 68.9 § ICD-9-CM codes: 38.86, 39.79. The codes 68.24 and 68.25: uterine artery embolization (UAE) with coils and without coils, were not yet available. $ ICD-9-CM codes : 642.6x # ICD-9-CM codes : 665.0, 665.1

Table II. Number of peripartum hysterectomies and peripartum arterial embolisations in 2012-2013 recorded by IMA-AIM; compared to the number registered by the Belgian Obstetric Surveillance System (B.OSS). Peripartum Peripartum Severe obstetric hysterectomy embolisation complication n n

IMA – AIM* 77 139

B.OSS 79 96

*IMA-AIM records billing data of Belgian inhabitants affiliated to the seven Belgian Health Insurance funds, using nomenclature codes.

Figure 1. The continuum from uncomplicated pregnancy to maternal mortality. Adapted from The Pyramid of Disease. PowerPoint Presentation. Introduction to UKOSS. Retrieved from https://www.npeu.ox.ac.uk/ukoss. and The spectrum of morbidity: from noncomplicated pregnancies to maternal death. Say et al., (2009).

100

90 80 70 60 50 40 30 20

10 Percentage of reporting forms returned forms reporting of Percentage 0 2012 2013 2014 2015 2016

Figure 2. Yearly response rate of the Belgian Obstetric Surveillance System in percentage

.

DISCUSSION

Main findings: infographic

- During the study period (2012-2013) 252,272 women gave birth in Belgium. 10.7% had a previous Caesarean Section (CS) 47.2% of women with previous CS underwent TOLAC* - Uterine rupture (clinical definition) occurred in 1 in 2,777 women (3.6 per 10,000 deliveries (95% CI 2.9-4.4)) 1 in 370 women with previous CS (27 per 10,000 deliveries (95% CI 21-33)) 1 in 14,286 women without previous CS (0.7 per 10,000 deliveries (95% CI 0.4-1.2)) - Complete uterine rupture (anatomical definition) occurred in 1 in 3,448 women (2.9 per 10,000 deliveries (95% CI 2.3-3.7)) 1 in 474 women (2.1 per 10,000 deliveries (95% CI 16.3-27.3)) 1 in 14,286 women without previous CS (0.7 per 10,000 deliveries (95% CI 0.4-1.2)) - Of the women with uterine rupture 9% underwent hysterectomy 22% was admitted to an intensive care unit (ICU) 40% had transfusion - 16% had >4units of PC and FFP - Of neonates born to mothers with complete uterine rupture 13% died 38% suffered from asphyxia - Major obstetric haemorrhage requiring peripartum hysterectomy and/or interventional radiology (IR) occurred in 1 in 1,515 women (6.6 per 10,000 deliveries (95% CI 5.7-7.7)) 1 in 3,030 women underwent hysterectomy (3.3 per 10,000 deliveries (95% CI 2.7-4.1)) 1 in 3,030 women underwent IR (3.3 per 10,000 deliveries (95% 2.6-4.0))

*: Number of TOLAC derived from the perinatal registries (SPE and CEpiP), based on the number of women with previous CS undergoing primary and secondary caesarean. This will be an over-estimation, because women who presented in spontaneous labour despite having planned elective repeat caesarean section (ERCS) cannot be recognised as such and have been included as TOLAC.

- Participation rate: (the number of maternity units who formally agree to participate in B.OSS, in relation to the total number of maternity units) At start, 110 of 113 Belgian maternity units (97.3%) formally agreed to participate in B.OSS; the three remaining maternity units have been recruited since. The number of maternity units dropped through the years as a result of merging and

closure, with 106 of 107 (99%) participating maternity units in January 2017 (Figure 6 and 7).

- Response rate: (the number of monthly reporting forms returned in relation to the total number of reporting forms sent) B.OSS achieved an excellent response rate at start in 2012-2013 (98.9%), which slightly dropped in the following years (Figure 8).

- Data collection forms completed: We retrieved data collection forms for 92% and 89% of the notified cases in the study of uterine rupture and the study of peripartum hysterectomy and arterial embolisation respectively.

- Accuracy of completion: No further cases needed to be excluded because of missing data. The majority of data collection forms returned were complete. Missing and inaccurate data were retrieved or modified by means of repeated emails and phone calls to contact persons. In comparison, the number of missing data in the study of eclampsia was higher, attributable to a longer and more extensive data collection form (unpublished results).

- Performance of B.OSS within the INOSS: The abovementioned numbers are in the same range as in comparable studies of the United Kingdom1 (100% participation rate, 90% data collection forms returned) and the Netherlands2 (100% participation rate, 97% response rate, 100% data collection forms returned). B.OSS provided data for two comparative INOSS studies (the study of uterine rupture and the study of peripartum hysterectomy) and is currently collecting data for two collaborative INOSS studies of extremely rare conditions (spontaneous hemoperitoneum in pregnancy and anaphylaxis in pregnancy). Further, B.OSS conducted at short notice the one- week Global Maternal Sepsis study organised by the WHO.

Figure 6. Belgian maternity units participating in B.OSS at initiation in 2012.

Figure 7. Belgian maternity units participating in B.OSS in 2017.

100

90

80

70

60

50

40

30

20

Percentage of reporting forms returned forms reporting of Percentage 10

0 2012 2013 2014 2015 2016

Figure 8. Yearly response rate of the Belgian Obstetric Surveillance System in percentage.

Strengths and weaknesses.

i. Although relying on the willingness of busy clinicians, B.OSS achieves an excellent participation rate and response rate. The success of B.OSS demonstrates the hunger for information on severe maternal morbidity on a national basis. Rates gradually dropped in the latest years however, blamed partly on the extremely rare conditions currently under investigation. Mostly lack of time is the justification given for a delay in response. Some ob/gyn specialists mentioned the lack of an apparent organisation behind B.OSS as a barrier to participate. On a single occasion, the lack of a financial reimbursement was named.

ii. Thanks to its methodology, that is the prospective collection of cases through an active search on a monthly basis according to the principle of nothing-to-report, the number of missed cases (false negatives) can be reduced to a great extent. Besides, the number of false positives is limited to a minimum thanks to the detailed data collection forms that confirm the diagnosis. However, there is currently no appropriate way to check the data of B.OSS for completeness and correctness. We checked two national data sources collecting medical data on a mandatory basis, that theoretically allow control for underreporting: the Belgian hospital discharge register (MZG, RHM) and the Inter Health Insurance Funds Agency (IMA-AIM). We concluded that both were not accurate to serve for validation of B.OSS data, except for surveys of procedures indicative of severe maternal morbidity (such as hysterectomy), where the MA-AIM database could be consulted for validation.3 Cases could be not reported when the B.OSS contact person is not aware of them, e.g. when the woman is admitted to ICU or to another department, when the maternity unit consists of more than one campus, or of more than one association (group of ob/gyn specialists). Furthermore, we cannot rule out that cases are hold back intentionally, when there were obvious quality of care concerns involved. iii. The launch of the B.OSS website (http://www.b-oss.be) substantially reduced the workload of the research team and contact persons by enabling web-based data collection. For example, the monthly reporting is now easily done in three clicks by an email link. However, the commitment to the B.OSS project does involve additional work to contact persons on top of their clinical practice. Moreover, clinicians are regularly approached to participate in all kind of surveys. Additionally, many surveys, registers and trials overlap by requesting similar information to a great or lesser extent. iv. B.OSS can rely on the perinatal registries (SPE and CEpiP) to derive data from the background population, to use as denominator to calculate prevalences and to compare socio-demographic and obstetric characteristics as possible risk factors.

The perinatal data from three different regions (Flanders, Brussels and Wallonia) is provided in three different databases, which must be aggregated to obtain the national data. Moreover, there is overlap of hospitals between regions, which must be taken into account. Additionally, there is a slight discrepancy in the inclusion criteria between SPE and CEpiP: in Brussels and Wallonia (CEpiP) births are registered as such, in case of a live birth or in case of a stillbirth at a birth weight of 500 grams or more and/or after 22 completed weeks of gestation, whereas in Flanders (SPE) live births or stillbirths with a birth weight below 500 grams, irrespective of gestational age, are not registered. All this may give rise to small inaccuracies. B.OSS gathers information of cases only, not controls, thereby restricting the output to descriptive epidemiological studies, whereas case- control studies would generate additional information (e.g. the inclusion of control women with previous caesarean section who did not suffer from uterine rupture, would enable to use logistic regression and adjust for possible confounding factors in the evaluation of possible risk factors for uterine rupture).

Recommendations to improve the performance of B.OSS. i. To further motivate clinicians and boost the response rate, we suggest several action points. Firstly, starting new studies that investigate ‘more common’ conditions that are relevant for every obstetrician, combined with the study of extremely rare conditions eventually. Secondly, proclaiming the results of the first B.OSS studies in presentations and publications and proclaiming the impact on quality of care, when recommendations have been implemented over time. Thirdly, formalising the organisation of B.OSS and announcing this official structure to the Belgian maternity units. To this end, a formal scientific board has been constituted in 2017 with representatives of SPE, CEpiP, the Belgian Health Care Knowledge Centre (KCE), the Scientific Institute of Public Health (WIV-ISP) and the College of Physicians for Mother and Newborn. The College, operating as the steering committee since the start of B.OSS, officially becomes the Governing Board. Fourthly, rewarding the maternity units that register to B.OSS in a “pay for quality (P4Q)” or “pay for performance (P4P)” program. This involves, as the name implies, providing a financial reward to health care providers or institutions, when they can demonstrate quality of care, assessed based on a list of indicators, typically structure, process and outcome indicators. Participation in B.OSS was proposed as a possible non-mandatory indicator of the Belgian P4Q project, but has not been not selected, ironically, because B.OSS is performing too well. One step further would be elevating B.OSS to become a mandatary registration; which is not even the case for UKOSS in the UK however. ii. The validity of B.OSS data could be further improved by involving more than one clinician per maternity unit. This would follow the lead of the UKOSS, where four

clinicians are nominated in each hospital.4 Eligible are a senior midwife (which would overcome the problem of multiple associations of ob/gyn specialists in one maternity unit), an obstetric anaesthetist (in larger centres) or general anaesthetist (aware of pregnant women presenting at the emergency department, intensive care unit (ICU) or undergoing surgical interventions), an ICU specialist (who would pick up women at ICU transferred from other hospitals). Eventually a neonatologist, who could identify mothers with severe maternal morbidity related to neonates admitted to the NICU. And, where applicable, a medical specialist in line with the complication surveyed (e.g. a pneumologist when surveying pulmonary embolism). Multiple reporting clinicians per hospital would require an upgrade of the website to enable web-based data collection, which is currently rather basic to keep the expenses low. iii. To further improve the workload of reporting clinicians, attention must be paid to keep data collection forms of future surveys more concise and clear. Worth mentioning in this context is that future data collections will adhere to the INOSS Core Dataset (ICoD), a standardized minimum dataset developed by the International Network of Obstetric Survey System (INOSS), to enhance the comparison of data between different countries.5 Over time, it would be most expedient if B.OSS could join the federal project Healthdata.be 6, a platform developed to reduce the administrative burden of clinicians. The Healthdata- platform aims to standardise and harmonise the registration and storage of health data, while absolute confidentiality is guaranteed. The ideology of the platform is ‘the only-once principle’: information - health data in this case - should be provided only once and can then be shared and reused. The platform enables to gather health data in a standardized way from different health care institutions and health care providers using many different medical record systems, by implementing a specifically developed software program. Since start in 2014, they succeeded to successfully integrate 42 registries into the platform. We await further development and results of Healthdata, to assess whether the B.OSS data collection could be (partially) covered by this platform. There are some reservations about the feasibility of this however, firstly, the automated identification of cases, which most likely would be based on codes (ICD codes, nomenclature codes, billing codes); it is known however that administrative data are not accurate to survey severe maternal morbidity, mainly due to overestimation (false positives).3, 7, 8Further there are doubts on how such platform could retrieve the sequence and time line of events, information that is written in the case notes by the people involved. Nevertheless, there would already be a reduction of workload, if only part of the required data was gathered by this platform, e.g. sociodemographic data, medical history. Moreover, the Health data platform is perfectly in line with the new European Data Privacy Law and is financed by the government.

iv. The daily work by the research team, which mainly involves the pursuit of missing data by emailing and calling, is carried out by two cooperating teams: one team is taking care of the maternity units in Brussels and Wallonia, another team of the maternity unit in Flanders. The work in Brussels and Wallonia is managed by the personnel of the perinatal registry CEpiP. In Flanders B.OSS is currently depending on ob/gyn trainees and medical students who contribute to B.OSS studies as part of a thesis. It would be most logical, if the B.OSS activities in Flanders became part of the Flemish perinatal registry SPE. In this way, B.OSS could unite both perinatal registries into one conjoint national perinatal registry, where discrepancies and joint projects can be discussed. Furthermore, this would enable to integrate new variables in the perinatal registries that are interesting for the analyses of B.OSS studies, such as ‘the intended mode of delivery’ to determine the true rate of trial of labour after caesarean (TOLAC) and vaginal birth after caesarean (VBAC).

Recommendations for further development and future studies of B.OSS.

The B.OSS surveys give an overall picture of the severe obstetric complications under surveillance, by an epidemiological description of prevalence, risk factors, management and outcome for mother and child. They do not allow to make firm conclusions on the quality of care provided to individual women included in the surveys. A case-by-case in-depth analysis is required based on the full medical records, thereby assessing the management against relevant guidelines, to judge whether there is evidence of quality of care concerns. In the UK the MBRRACE-UK is even going beyond, by organising structured reviews of near-miss cases by a multidisciplinary team of appointed experts. This is done on an annual basis since 2014 in a programme additional to the Confidential Enquiries into Maternal Deaths (CEMD) and the UKOSS, the so-called Confidential Enquiries into Maternal Morbidities (CEMM).9 An interesting audit study would be the in-depth analysis of the hysterectomies and the arterial embolisations described in the B.OSS study10, to reveal whether they were appropriate or preventable. In the discussion we hypothesised that hysterectomies may be performed more easily in smaller maternity units, while on the other hand there may be an overuse of arterial embolisations when this service is easily accessible. These are merely assumptions derived from the study findings, that cannot be confirmed unless an additional audit study is undertaken. In doing so, it is important to realise that the timely decision to perform a peripartum hysterectomy can save a woman’s life.11 The decision is usuallly taken under stressful conditions and various factors are taken into consideration when deciding. The assessment of the full

medical notes of every case would enable to check all circumstances – e.g. the woman’s desire for continued fertility, her haemodynamic status, the presence of a senior ob/gyn specialist and an anaesthetist, the immediate availability of an operating room and staff, the availability of or distance to the nearest interventional radiology service, among others – that could have played a role in the decision to perform a hysterectomy. Further, it would enable to point out the sequence and timing of events and interventions from delivery to hysterectomy, insofar as detailed narratives are written in the case notes. A population based audit in Denmark could serve as a model: they performed a structured audit based on medical records of 50 peripartum hysterectomies. By the use of a simple audit evaluation form they classified the hysterectomies as 1) avoidable, 2) potentially avoidable and 3) unavoidable. More than 50% of peripartum hysterectomies were avoidable by simple measures such as the use of intrauterine balloons.12 The cases of arterial embolisations in the B.OSS study merit a similar in-depth analysis. See further, topic ‘Future studies, ii’. To set up audit studies in addition to B.OSS surveys in Belgium, we first need to develop the practice of performing audits and confidential enquiries, and preferably have drafted nationally adopted guidelines against which to assess the quality of care. See further, topic ‘Guidelines’.

Belgium can no longer lag behind its neighbouring countries (the UK, the Netherlands and France)13-15 which have a well-established Confidential Enquiries into Maternal Deaths (CEMD) since many years. Even certain middle- and low-income countries (Malaysia, South Africa, India, Ghana, Indonesia … )16 succeed in conducting a Maternal Death Surveillance and Response system (MDSR17) to some extent, mostly an adapted version of the CEMD. As in all high-income countries, Belgium can provide estimates of the maternal mortality ratio (MMR) and a basic classification into direct, indirect and late maternal deaths based on the vital statistics.18 (See above, ‘Introduction’, topic ‘Obstetric registers and surveys in Belgium’) The knowledge that this number is most likely an under-estimation of the true MMR19 is not the main reason why establishing a Belgian maternal mortality surveillance system would be worthwhile. ‘Beyond the numbers: reviewing maternal deaths and complications to make pregnancy safer’, the title of a document authored by the World Health Organisation (WHO), clearly expresses why: because knowing the level of maternal mortality is not enough.20 In high-income countries every maternal death must be considered a ‘sentinel event’21, 22, which is defined by the Joint Commission as a patient safety event that resulted in an unanticipated death, not primarily related to the natural course of the patient’s illness or underlying condition.22, 23 Named “sentinel”, the lookout or guard, because it signals the need for immediate investigation or response:23 sentinel events raise questions about safety and quality of

care and require careful investigation to identify the causal and contributory factors.19, 22, 23 Obviously not all maternal deaths are true sentinel events, in the sense that some will be related to the natural course of the underlying condition. Reviewing all cases of maternal death remains useful however, because the aim is not to blame or punish but to improve patient safety and quality of care by learning from adverse events.21 Now that B.OSS has a firm position, supported by the great majority of Belgian ob/gyn specialists, it is the right time to add maternal mortality to the project. There is clearly a need for a structured maternal mortality register in Belgium, since the B.OSS research team was asked by several clinicians where to report a case of maternal mortality. Even a family member of a mother who died, residing in a low-income country with an established (adapted version) of CEMD, questioned the lack of a Belgian CEMD. A simple poll indicated that a great majority of Belgian ob/gyn specialists (95% (72/76 maternity units) recognise the need for a Belgian maternal mortality registration. Only 88% (52/59 maternity units) however find it useful to participate in a Belgian CEMD with a structured review of all cases of maternal deaths. The College of Physicians for Mother and Newborn is currently taking the lead and appointed a working group to develop a structured Belgian CEMD. The main challenge will be to reassure the ob/gyn specialists that confidentiality will be guaranteed, because cases could be easily recognisable in our very small country and fear of litigation is the most important objection to participate.

The UKOSS was inspired by the British Paediatric Surveillance Unit (BPSU), a similar system that surveys uncommon disorders of childhood since 1986.24 The reverse could happen in Belgium, where B.OSS may inspire to develop a system to survey rare neonatal complications, that are relevant for ob/gyn specialists: e.g. subgaleal heamorrhage and skull fractures, Erb’s palsy, water birth complications etc. However, neonatologists are already involved in the perinatal registries (SPE and CEpiP) and are conducting the NIC-audit since 2004, surveying all neonates admitted to the 19 Neonatal Intensive Care (NIC) units based on 40 quality indicators.25 The NIC-audit is an initiative of the Newborn’s Section of the College of Physicians for Mother and Newborn and aims to improve the quality of care, similarly as the B.OSS initiative. Regardless, the B.OSS surveillance would benefit from the involvement of neonatologists, by identifying mothers with severe maternal morbidity related to neonates admitted to the NICU (See above), and to assist in reporting the outcome of neonates born to women with severe maternal morbidity, especially at a later stage.

i. Firstly, being able to discuss the topic ‘future studies’ is an achievement of the B.OSS project to be proud of. In the selection of interesting and suitable topics to survey by B.OSS, we can rely upon different sources: 1) the eight conditions for which INOSS has developed a consensus definition5, 2) conditions that are currently surveyed by or have been surveyed by other INOSS member countries, to enable comparative studies, 3) multi-country INOSS studies of extremely rare conditions, such as the study of anaphylaxis in pregnancy, 4) other topics suggested by Belgian ob/gyn specialists themselves. The decision will be made at the annual meeting of the scientific board by agreement between the parties represented: SPE, CEpiP, KCE, WIV-ISP and the College of Mother and Newborn. A broad and interesting study would be to register all women who suffer severe primary postpartum haemorrhage in Belgium, according to the recently agreed INOSS definition5 (postpartum blood loss exceeding 2000 cc and/or the need of transfusion of at least four units of red blood cells, within 24 hours after the end of the pregnancy, in a pregnancy of at least 20 weeks gestational age), further extended with the registration of all women who do not meet this definition but underwent peripartum hysterectomy and/or arterial embolisation or were successfully managed with other second-line measures (intra-uterine balloon, haemostatic brace suturing or arterial ligation). This would provide an overall picture of postpartum haemorrhage in Belgium and would provide denominator data enabling to calculate the failure rate of the second-line measures used. ii. A French population-based study by Kayem et al. indicated a high rate of invasive therapies used for primary postpartum haemorrhage (PPH) in France, nearly one procedure in 550 deliveries, including conservative surgery (26%), arterial embolisation (62%) and hysterectomy (12%).26 As one of the possible explanations for this high rate they suggested an overuse of arterial embolisation for moderately severe PPH, because ob/gyn specialists tend to anticipate its use while radiologists usually don’t re-evaluate the need for the procedure. 26 It is debatable whether an ‘overuse’ of this potentially life-saving procedure should be considered unfavourable, in the knowledge that related complications are rare27 (procedure related complications in 3-9%, fever and pelvic pain in 3-9%, major vascular and neurologic complications in 1-2%) and future fertility is not affected28, 29 apart from a possible increased risk of repeat postpartum haemorrhage related to abnormally invasive placenta.29 It is however a time- consuming procedure, mostly in acute thus unscheduled settings, that can only be carried out by a small number of radiologists having the skills. A simple inquiry of B.OSS contact persons demonstrated the availability of an interventional radiology (IR) service in at least 38% of Belgian maternity units, not including those that can call upon vascular surgeons for similar procedures.10 Triggered by the hypothesis of Kayem et al., we would propose an audit study analysis of the

cases of peripartum arterial embolisation in Belgium, paying particular attention to the use of first and second-line measures and the sequence and timing of the interventions before the decision to resort to interventional radiology. This study combined with a thorough charting of the availability and network of interventional radiology services, would reveal to what extent IR is used efficiently in the management of PPH and what the optimal geographic distribution of IR services would be. iii. The unique INOSS collaboration allows for the study of extremely rare conditions by enabling the inclusion of an adequate number of cases in a relatively short time frame.30 The international study of uterine rupture included 120 cases in women without previous caesarean section, resulting in a prevalence of 0.6 (95% CO 0.5-0.7) per 10,000 deliveries. We plan an in-depth analysis of these very rare cases of uterine rupture in the unscarred uterus, known to have a higher risk for severe outcome for both mother and child. The aim is to describe the presentation (symptoms and signs, mode of delivery, location), the potential risk factors (maternal characteristics and history, obstetric factors), the management and outcome for mother and neonate. We plan a similar analysis of the cases of uterine ruptures that occurred preterm or in the non-labouring uterus. iv. B.OSS aims to monitor the impact on patient safety in future surveys, when guidelines and recommendations are put into practice (e.g. a reduction in number of cases of uterine rupture, peripartum hysterectomy, eclampsia in a follow-up study after 10 years).

Recommendations to improve obstetric care in Belgium.

Registration and gaining knowledge is one thing, using this knowledge to ‘improve quality and safety of care’ is quite another. B.OSS has proved to be successful in providing an accurate picture of the obstetric complications under surveillance (primary objective, see above ‘Objectives’). However, identifying so-called ‘weaknesses in the health care chain’ based on the first two completed surveys, that can be translated into practical recommendations (our secondary objective), is not that clear-cut. Even though, obstetric care would certainly benefit from the implementation of the generally applicable recommendations described below.

Why would Belgium and its ob/gyn specialists need national guidelines for obstetric care? There is no need to reinvent the wheel. Belgian ob/gyn specialists can rely on different highly respected international guidelines, which they can consult easily on the net. Prominent guidelines are those by the RCOG and the ACOG, besides the

guidelines of the Nederlandse Vereniging voor Obstetrie & Gynaecologie (NVOG) often referred to by the Flemish ob/gyn specialists. There are however several reasons why Belgium would benefit from nationally adapted guidelines: 1) To enhance the awareness of certain good practices and risks, considered as commonly known by the majority of clinicians, yet not always applied in practice (e.g. the use of Magnesium Sulphate as first choice treatment for control of seizures with eclampsia, e.g. the increased risk of uterine rupture when using prostaglandins for Trial of Labour after Caesarean, e.g. to check the placenta localisation at the routine 20 and 30 weeks screening ultrasound, especially in women with previous CS. 2) To enable the conduction of audit studies and confidential enquiries, we need nationally adapted guidelines supported by the appropriate authorities, against which management and quality of care can be assessed. Here too, the international guidelines could be of use, but the minor discrepancies in between those guidelines (e.g. the on-site availability of ob/gyn specialists in case of TOLAC) and the discrepancies with the Belgian situation (e.g. the fee-for-service payment model in Belgium) could lead to discussion. 3) To serve as a basis to develop hospital-based clinical guidelines. The requirements that hospitals need to fulfil to obtain accreditation or quality labels are constantly rising. Encouraged by these quality institutions numerous ob/gyn departments develop their own hospital-based clinical guidelines. A national guideline would facilitate this process and may harmonise the approaches that currently differ between ob/gyn specialists, associations, hospitals and regions. We make a start by developing a nationally adopted guideline on trial of labour after caesarean section (TOLAC). This guideline will result from the comparison and fusion of the existing international guidelines 31-34 and have the consent of the professional associations for ob/gyn specialists, the professional associations for midwifes and the College of Physicians for Mother and Newborn. The guideline is developed to promote safe TOLAC in appropriate candidates on the one hand, in order to enhance the low current TOLAC rate in Belgium. On the other hand, the guideline should prevent unsafe procedures, which have been observed among the uterine rupture cases in the B.OSS study. Furthermore, the guideline will include recommendations on the mode of delivery in women with other uterine surgical procedures (such as myomectomie, hysteroscopic resection of a uterine septum or salpingectomy), in whom a management similar to women with previous caesarean section is justified. It is hoped that many other guidelines will follow. By implementing guidelines we automatically create opportunities for future and further surveys, to monitor the adherence to the guideline and the impact on the quality and safety of the obstetric health care.

Besides the national audit studies, B.OSS surveys should encourage the initiation of audits in ob/gyn departments in hospitals. The subject of these local audits could be: 1) the cases of severe maternal morbidities surveyed by B.OSS or currently under surveillance by B.OSS, 2) the cases that meet at least one of the two screening criteria for SMM, recommended by the American College of Obstetricans and Gynecologists (ACOG) and the Society for Maternal Fetal Medicine (SMFM): a) transfusion of four or more units of blood and b) admission of a pregnant or postpartum woman to an ICU.35, 3) the cases of SMM agreed upon by the local ob/gyn specialists. The aim of reviewing SMM cases at a hospital level is to learn from own adverse events by performing a detailed multi-disciplinary analysis in search for weaknesses and to improve by undertaking corrective actions, to prevent similar events in the future. Local audits are easier to set up compared to national initiatives and would help in creating a culture of learning instead of blame, which would pave the way to introduce a system of national audits.

The findings of B.OSS studies invite to critically evaluate the current organisation of obstetric health care in Belgium. More specifically: 1. Belgium has a rather low Trial of Labour after Caesaeran section (TOLAC) rate, being the second lowest within the INOSS-group (47.2%). This number derived from the perinatal registries (SPE and CEpiP) will be an over-estimation even, because women who presented in spontaneous labour despite having planned elective repeat caesarean section (ERCS) cannot be recognised as such and have been included as TOLAC. This finding brings us to three questions. Firstly, why is our TOLAC rate lower than others? Secondly, should we make efforts to increase our TOLAC rate, would it improve the obstetric health care? Thirdly, are there ways to increase our TOLAC rate? Possible clarifications why Belgian ob/gyn specialists less commonly undertake TOLAC are i) elective repeat caesarean sections (ERCS) can be scheduled, are less time- consuming and do not involve a risk of uterine rupture, it should be no wonder that ob/gyn specialist stick to the old statement “once a cesarean always a cesarean”; ii) the higher payments for caesarean sections compared to vaginal deliveries,36 because women stay longer in the hospital and can be charged supplemental fees in private rooms; iii) the majority of maternity units cannot guarantee the presence of an ob/gyn specialist 24 hours a day, but have a scheduled on-call system where the on- call doctor should be able to come in within 15 minutes. Besides, there is still a number of Belgian ob/gyn specialists who are available day and night for their (private) patients. An interesting study by Yee et al. showed that the likelihood of

attempting TOLAC increased 2-fold, with a similar increase in VBAC, in a night float call system (a schedule in which the doctor on-call has clinical responsibilities for labouring patients only for a day or night shift) compared with a traditional call system.37 Moreover, international guidelines, and particularly the ACOG Practice Bulletin, recommend that TOLAC should be attempted in facilities that can provide emergency caesarean sections, which involves the immediate availability of obstetric, paediatric, anaesthesiology and operating room staff.34 Would the Belgian obstetric health care be improved by increasing the TOLAC rate? It would help to halt the increasing CS rate, which is currently around one in five (20.7% in 2015)38-40 and substantially higher than the Netherlands, Sweden and Finland. Halting the CS rate, rather than reducing, is certainly what Belgium intends, triggered by recent reports from the World Health Organisation (WHO).41 An intended result besides reducing short-term maternal morbidity associated with CS, is to prevent an increase in prevalence of abnormal invasive placenta (AIP), which is associated with an increasing number of previous caesarean sections (1/363 in women with two previous CSs)42. The mean number of children per family in Belgium is 2.1 (Source: Famifed 2014). Further, we must take into account that a higher TOLAC rate will probably be associated with a slightly higher uterine rupture rate, as demonstrated by the INOSS study (unpublished results). Moreover, there is more to gain from the reduction of primary caesarean sections in nulliparous women with single cephalic pregnancies ≥ 37 weeks (Robson group 1 and 2).43, 44 How could we increase the TOLAC rate in Belgium? Besides raising awareness by a national guideline, an effective way would be by reformations in the financing of deliveries and caesarean sections. This can be very time-consuming seen the request to introduce an ‘urgency fee’, which means a higher remuneration for night time deliveries, was ongoing since 20 years. More invasive, and probably illusory, would be to obligate by law the 24 hours on-site availability of the staff and resources necessary to perform an immediate caesarean section in every maternity unit. This would require major organisational reforms, such as the fusion of the numerous small maternity units to become a limited number of large maternity units, equipped with all facilities. Another ambitious plan would be to centralise the care for pregnant women with previous CS (estimated 10,7% of our pregnant women, thus around 13,200 annual deliveries) in tertiary referral centres that provide 24 hours on-site continuity.

2. The study of peripartum hysterectomy and arterial embolisation can give some insight into the transfer policy and use of tertiary referral centers, the so- called Maternal Intensive Care (MIC)-centers. Of 166 women who required invasive therapy for postpartum haemorrhage (PPH), 68% (n=113) were managed in a MIC- center, of whom 3 following antenatal transfer and 12 following postnatal transfer. Five women were transferred postnatally to a non-MIC center with interventional radiology (IR) service availability. The peripartum hysterectomy rate was 76% of women who delivered in a maternity unit without emergency IR service. Considering

those women in whom a postpartum haemorrhage was highly predictable, 87% (34 / 38) of women with abnormal invasive placenta (AIP) and 67% (12 / 18) of women with placenta praevia only (non-AIP) delivered in a MIC center. These findings give rise to the following questions: 1) Should the availability of emergency Interventional Radiology services in Belgium be optimised? The time is now, since major reforms to the Belgian’s health care organisation are currently ongoing. Would it not be better to have all MIC-centers equipped with an on-site IR service? Should hospitals with small maternity units (less than 1000, even 500 annual deliveries) have on-site IR service in the first place? A thorough charting of the availability and network of interventional radiology services is needed (See above, Future studies), to reveal whether the geographic distribution of IR services could be improved. The geographic distribution of MIC-centers should be questioned in the same way.45 2) Should the referral of women with suspicion of abnormal invasive placenta to MIC- centers, with on-site availability of interventional radiology service and staff and resources necessary to perform an immediate caesarean section, become a formal MIC admission criterium? Similarly, should the referral of women with placenta praevia to MIC-centers become a formal MIC admission criterium? The effectiveness and efficiency of the MIC function in Belgium has been assessed in a large research project.45 The researchers concluded that the MIC-services were not used appropriately, with only 40% of eligible women actually delivering in a MIC-service, while the so-called MIC-beds were variably but always under-occupied. According to the researchers, this is due to the absence of formal admission criteria or referral pathways, besides the unequal geographical distribution of MIC-centers concentrated in the large cities. Their recommendation was therefore to formalise the admission criteria and referral pathways and to re-examine the distribution of MIC-beds.45

Table 2. Women who required invasive therapy for postpartum haemorrhage according to type of hospital (MIC-centre / non-MIC -centre; IR availability; annual number of deliveries)

MIC IR Annual Number Number of women delivered in this number of of type of hospital who underwent deliveries hospitals invasive therapy for PPH (Hysterectomy (H), Embolisation (E)) (Total n=166) yes yes ≥2000 9 90 22 H (1 antenatal transfer) 1000-1999 5 45 E (2 antenatal transfer) 8 H post E 1 E post H Yes ≥2000 1 4H not 3E 24/7 2H post E No ≥2000 1 3H 1000-1999 1 2 E (2 postnatal transfer) No Yes ≥2000 2 31 11 H 1000-1999 11 18 E 500-999 10 2 H post E <500 2 Yes 1000-1999 1 11 3 H not 6 E (3 postnatal transfer) 500-999 4 24/7 2 H post E <500 1

No ≥2000 2 34 22 H 1000-1999 16 8 E (8 postnatal transfer) 500-999 28 3 H post E (3 postnatal <500 8 transfer) 1 E post H (1 postnatal transfer) Not 500-999 4 / / known <500 1

Figure 10. Geographic distribution of maternity units and Maternal Intensive Care (MIC)-centres with Interventional Radiology service in Belgium. Source: Inquiry via B.OSS contact persons.

Women suffering postpartum haemorrhage benefit greatly when they are in hands of a multidisciplinary team, that is trained in situation awareness, decision making, communication, teamwork and leadership, the so-called ‘non-technical skills’, besides the medical technical skills. Obstetric and multidisciplinary team training in a simulation setting for acute obstetric emergencies can significantly improve the performance of a team46 and thus the patient safety by preventing ‘human errors’.47 Obstetric emergency training in medical simulation centers, or even better when organised in the own environment of the maternity unit, is gradually making entrance in Belgium. To enhance this positive upcoming trend, the organisation of and/or participation in multidisciplinary team training in simulation setting should become another quality indicator in the pay for quality program (see higher, ‘Recommendations to improve the performance of B.OSS’). Since two years we organise the “Obstetrische skills en drills” at the maternity unit of the Ghent university hospital. The obstetric emergencies we simulated were: postpartum haemorrhage, antepartum haemorrhage, eclamptic fit, umbilical cord prolapse, shoulderdystocia and unexpected breech delivery. Besides ob/gyn specialists and

midwives, the anaesthesiologists and neonatologists were involved. It takes time and effort to organise, which is definitely worthwile but may be only manageable at a university referral center. Similar team trainings are offered by several simulation centers in their center or on-site.

References

1. Fitzpatrick KE, Kurinczuk JJ, Alfirevic Z, Spark P, Brocklehurst P, Knight M. Uterine rupture by intended mode of delivery in the UK: a national case-control study. PLoS medicine. 2012;9(3):e1001184. 2. Zwart JJ, Richters JM, Ory F, de Vries JI, Bloemenkamp KW, van Roosmalen J. Uterine rupture in The Netherlands: a nationwide population-based cohort study. BJOG : an international journal of obstetrics and gynaecology. 2009 Jul;116(8):1069-78; discussion 78-80. 3. Vandenberghe G, Roelens K, Van Leeuw V, Englert Y, Hanssens M, Verstraelen H. The Belgian Obstetric Surveillance System to monitor Severe Maternal Morbidity. (2017) Unpublished results, paper submitted to Facts, views and visions in Obgyn. 4. Knight M, Kurinczuk JJ, Tuffnell D, Brocklehurst P. The UK Obstetric Surveillance System for rare disorders of pregnancy. BJOG : an international journal of obstetrics and gynaecology. 2005 Mar;112(3):263-5. 5. Schaap T, Bloemenkamp K, Deneux-Tharaux C, Knight M, Langhoff-Roos J, Sullivan E, et al. Defining definitions: a Delphi study to develop a core outcome set for conditions of severe maternal morbidity. BJOG : an international journal of obstetrics and gynaecology. 2017 Jul 29. 6. WIV-ISP. Healthdata.be. A short introduction. Retrieved October 2017 from: https://healthdata.wiv- isp.be/en/about-healthdatabe 7. Thisted DL, Mortensen LH, Hvidman L, Rasmussen SC, Larsen T, Krebs L. Use of ICD-10 codes to monitor uterine rupture: validation of a national birth registry. European journal of obstetrics, gynecology, and reproductive biology. 2014 Feb;173:23-8. 8. Sigakis MJ, Leffert LR, Mirzakhani H, Sharawi N, Rajala B, Callaghan WM, et al. The Validity of Discharge Billing Codes Reflecting Severe Maternal Morbidity. Anesthesia and analgesia. 2016 Sep;123(3):731-8. 9. Knight M, Lewis G, Acosta CD, Kurinczuk JJ. Maternal near-miss case reviews: the UK approach. BJOG : an international journal of obstetrics and gynaecology. 2014 Sep;121 Suppl 4:112-6. 10. Vandenberghe G, Guisset M, Janssens I, Leeuw VV, Roelens K, Hanssens M, et al. A nationwide population-based cohort study of peripartum hysterectomy and arterial embolisation in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2017 Nov 08;7(11):e016208. 11. Royal College of Obstetricians and Gynaecologists. Postpartum Haemorrhage, Prevention and Management (Green-top Guideline No. 52). 2009. 12. Colmorn LB, Krebs L, Langhoff-Roos J. Potentially Avoidable Peripartum Hysterectomies in Denmark: A Population Based Clinical Audit. PloS one. 2016;11(8):e0161302. 13. Lewis G. Saving Mothers' Lives: the continuing benefits for maternal health from the United Kingdom (UK) Confidential Enquires into Maternal Deaths. Seminars in perinatology. 2012 Feb;36(1):19-26. 14. Schutte JM, Steegers EA, Schuitemaker NW, Santema JG, de Boer K, Pel M, et al. Rise in maternal mortality in the Netherlands. BJOG : an international journal of obstetrics and gynaecology. 2010 Mar;117(4):399-406. 15. Saucedo M, Deneux-Tharaux C, Bouvier-Colle MH. Ten years of confidential inquiries into maternal deaths in France, 1998-2007. Obstetrics and gynecology. 2013 Oct;122(4):752-60. 16. Hussein J. Improving the use of confidential enquiries into maternal deaths in developing countries. Bulletin of the World Health Organization. 2007 Jan;85(1):68-9. 17. Smith H, Ameh C, Roos N, Mathai M, Broek NVD. Implementing maternal death surveillance and response: a review of lessons from country case studies. BMC pregnancy and childbirth. 2017 Jul 17;17(1):233.

18. Algemene Directie Statistiek - Statistics Belgium. Metagegevens. De statistiek van de moedersterfte. Retrieved November 2017 from: http://statbel.fgov.be/nl/binaries/meta_matmort_nl_tcm325- 281108.pdf 19. Bouvier-Colle MH, Mohangoo AD, Gissler M, Novak-Antolic Z, Vutuc C, Szamotulska K, et al. What about the mothers? An analysis of maternal mortality and morbidity in perinatal health surveillance systems in Europe. BJOG : an international journal of obstetrics and gynaecology. 2012 Jun;119(7):880- 9; discussion 90. 20. Lewis G. Beyond the numbers: reviewing maternal deaths and complications to make pregnancy safer. British medical bulletin. 2003;67:27-37. 21. The American College of Obstetricians and Gynecologists. Severe Maternal Morbidity: Clarification of the New Joint Commission Sentinel Event Policy. 2015. Retrieved November 2017 from: https://www.acog.org/About-ACOG/News-Room/Statements/2015/Severe-Maternal-Morbidity- Clarification-of-the-New-Joint-Commission-Sentinel-Event-Policy 22. The Joint Commission. Sentinel Events (SE). Comprehensive Accreditation Manual for Hospitals; 2016. Retrieved December 2017 from: https://www.jointcommission.org/assets/1/6/CAMH_24_SE_all_CURRENT.pdf 23. The Joint Commission. Sentinel Event Policy and Procedures. 2017 June 29. Retrieved December 2017 from: https://www.jointcommission.org/sentinel_event_policy_and_procedures/ 24. Verity C, Preece M. Surveillance for rare disorders by the BPSU. The British Paediatric Surveillance Unit. Archives of disease in childhood. 2002 Oct;87(4):269-71. 25. Vanhaesebrouck P, Charon A, Devlieger H, Haumont D, Langehendries J, Michel M, et al. NICaudit Synoptic Report 2000-2007: Federal Public Service Health, Food chain safety and Environment. Newborn's section of the College of Physicians for the Mother and Newborn.; 2008. 26. Kayem G, Dupont C, Bouvier-Colle MH, Rudigoz RC, Deneux-Tharaux C. Invasive therapies for primary postpartum haemorrhage: a population-based study in France. BJOG : an international journal of obstetrics and gynaecology. 2016 Mar;123(4):598-605. 27. Ruiz Labarta FJ, Pintado Recarte MP, Alvarez Luque A, Joigneau Prieto L, Perez Martin L, Gonzalez Leyte M, et al. Outcomes of pelvic arterial embolization in the management of postpartum haemorrhage: a case series study and systematic review. European journal of obstetrics, gynecology, and reproductive biology. 2016 Nov;206:12-21. 28. Gizzo S, Saccardi C, Patrelli TS, Di Gangi S, Breda E, Fagherazzi S, et al. Fertility rate and subsequent pregnancy outcomes after conservative surgical techniques in postpartum hemorrhage: 15 years of literature. Fertility and sterility. 2013 2013/06/01/;99(7):2097-107. 29. Poggi SH, Yaeger A, Wahdan Y, Ghidini A. Outcome of pregnancies after pelvic artery embolization for postpartum hemorrhage: retrospective cohort study. American journal of obstetrics and gynecology. 2015 Oct;213(4):576.e1-5. 30. Knight M. The International Network of Obstetric Survey Systems (INOSS): benefits of multi-country studies of severe and uncommon maternal morbidities. Acta obstetricia et gynecologica Scandinavica. 2014 Feb;93(2):127-31. 31. The Royal College of Obstetricians and Gynaecologists. Green-top Guideline No. 45. Birth After Previous Caesarean Birth. 2015. Retrieved December 2017 from: https://www.rcog.org.uk/globalassets/documents/guidelines/gtg_45.pdf. 32. The Society of Obstetricians and Gynecologists of Canada. Guidelines for vaginal birth after previous caesarean birth. Number 155. International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics. 2005 Jun;89(3):319-31. 33. Sentilhes L, Vayssiere C, Beucher G, Deneux-Tharaux C, Deruelle P, Diemunsch P, et al. Delivery for women with a previous cesarean: guidelines for clinical practice from the French College of Gynecologists and Obstetricians (CNGOF). European journal of obstetrics, gynecology, and reproductive biology. 2013 Sep;170(1):25-32. 34. The American College of Obstetricians and Gynecologists. Practice Bulletin No. 184: Vaginal Birth After Cesarean Delivery. Obstetrics and gynecology. 2017 Nov;130(5):e217-e33. 35. Kilpatrick SK, Ecker JL. Severe maternal morbidity: screening and review. American journal of obstetrics and gynecology. 2016 Sep;215(3):B17-22.

36. IMA-AIM, Regueras N, Guillaume J, De Zinno T, Ceupens A. Bevallen in België. Analyse van de gegevens 2008-2012. Prenatale zorgen in België in 2010 Vergelijking met de resultaten 2005, IMA, juni 2013. 37. Yee LM, Liu LY, Grobman WA. Obstetrician call schedule and obstetric outcomes among women eligible for a trial of labor after cesarean. American journal of obstetrics and gynecology. 2017 Jan;216(1):75.e1- .e6. 38. Leroy C, Van Leeuw V, Englert Y, Zhang W. Santé périnatale en Wallonie; 2015. 39. Van Leeuw V, Leroy C, Englert Y, Zhang W. Santé périnatale en Région bruxelloise, 2015. 40. Devlieger R, Martens E, Martens G, Van Mol C, Cammu H. Perinatale activiteiten in Vlaanderen 2015; 2015. 41. Ye J, Zhang J, Mikolajczyk R, Torloni MR, Gulmezoglu AM, Betran AP. Association between rates of caesarean section and maternal and neonatal mortality in the 21st century: a worldwide population- based ecological study with longitudinal data. BJOG : an international journal of obstetrics and gynaecology. 2016 Apr;123(5):745-53. 42. Thurn L, Lindqvist PG, Jakobsson M, Colmorn LB, Klungsoyr K, Bjarnadottir RI, et al. Abnormally invasive placenta-prevalence, risk factors and antenatal suspicion: results from a large population-based pregnancy cohort study in the Nordic countries. BJOG : an international journal of obstetrics and gynaecology. 2016 Jul;123(8):1348-55. 43. Robson M, Murphy M, Byrne F. Quality assurance: The 10-Group Classification System (Robson classification), induction of labor, and cesarean delivery. International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics. 2015 Oct;131 Suppl 1:S23-7. 44. Le Ray C, Blondel B, Prunet C, Khireddine I, Deneux-Tharaux C, Goffinet F. Stabilising the caesarean rate: which target population? BJOG: An International Journal of Obstetrics & Gynaecology. 2015;122(5):690-9. 45. Van Parijs A-S, Lucet C, Remacle A, Di Zinno T, Verstraelen H, Mambourg F, et al. Maternal Intensive Care in Belgium. Health Services Research (HSR). Brussel: Federaal Kenniscentrum voor de Gezondheidszorg (KCE); 2007. Report No.: (D/2008/10.273/77) 46. Fransen AF, van de Ven J, Merien AE, de Wit-Zuurendonk LD, Houterman S, Mol BW, et al. Effect of obstetric team training on team performance and medical technical skills: a randomised controlled trial. BJOG : an international journal of obstetrics and gynaecology. 2012 Oct;119(11):1387-93. 47. Merien AE, van de Ven J, Mol BW, Houterman S, Oei SG. Multidisciplinary team training in a simulation setting for acute obstetric emergencies: a systematic review. Obstetrics and gynecology. 2010 May;115(5):1021-31.

SUMMARY

The Belgian Obstetric Surveillance System, briefly B.OSS, was set up in 2011 with support of the College of Physicians for Mother and Newborn, a consultative body of the Federal Public Service of Health. B.OSS is a nationwide system that identifies and analyses rare severe obstetric complications and conditions in Belgium. The objective of B.OSS is measuring the so-called severe maternal morbidity (SMM) and maternal near-miss (MNM), the last-but-one level in a continuum from uncomplicated pregnancy to maternal death, to improve the quality and safety of obstetric health care in Belgium by practical recommendations based on the findings of the surveys. “The key to improve is to measure”. B.OSS has adopted the methodology developed by the UK Obstetric Surveillance System (UKOSS, https://www.npeu.ox.ac.uk/ukoss). Cases are identified actively and prospectively, by monthly mailing to all Belgian maternity units and according to the principle of ‘nothing-to-report’. Further, an extensive data collection form is completed in the event of a case being reported. Thanks to its methodology the number of false negatives and false positives is limited to a minimum; we trust that B.OSS is more precise than mandatory administrative databases (such as the hospital discharge register and the database of IMA-AIM). The launch of the B.OSS website (http://www.b-oss.be) in august 2013 substantially reduced the workload of the research team and contact persons by enabling web-based data collection. The methodology of UKOSS served as a template to several high-income countries, which are united within the so-called International Network of Obstetric Survey Systems (INOSS, https://www.npeu.ox.ac.uk/inoss)INOSS) The INOSS has the mission to co- operate, share information and enable cross-national comparisons and collaborative studies of very rare conditions. Since its start, B.OSS achieved an excellent participation rate of the Belgian maternity units (97.3%) and monthly response rate (98.9%), although relying on the willingness of busy clinicians. The success of B.OSS demonstrates the hunger for information on severe maternal morbidity on a national basis. The results of the first registration round (2012-2013) are brought out gradually, via scientific publications and presentations, news-letters, biennial reports and the website. Based on the findings of the B.OSS study of uterine rupture1 and the international comparative study of uterine rupture within INOSS2, we believe that obstetric care in Belgium would benefit from a nationally adopted guideline on trial of labour after caesarean section (TOLAC). The guideline should be developed to promote safe TOLAC in appropriate candidates on the one hand, in order to enhance the low current TOLAC rate in Belgium. On the other hand, the guideline should prevent unsafe procedures, which have been observed among the uterine rupture cases in the B.OSS study. Furthermore, the guideline should include recommendations on the mode of delivery in women with other uterine surgical procedures (such as myomectomy,

hysteroscopic resection of a uterine septum or salpingectomy), in whom a management similar to women with previous caesarean section is justified. The B.OSS study of peripartum hysterectomy and arterial embolisation3 reflects the use of these invasive therapies in the management of massive obstetric haemorrhage in Belgium. Ideally this study should be extended by the registration of all women who suffered from severe primary postpartum haemorrhage in Belgium, including all women who were successfully managed with other second-line measures (intra- uterine balloon, haemostatic brace suturing or arterial ligation). Further, this study requires an additional case-by-case in-depth analysis based on the full medical records to reveal whether the hysterectomies and arterial embolisations performed were appropriate or might have been preventable. We hypothesise that hysterectomies may be performed more easily in smaller maternity units, while on the other hand there may be an overuse of arterial embolisations when this service is easily accessible, without first resorting to first and second-line therapies. We expect the results of the study of eclampsia in Belgium very shortly. Currently B.OSS is registrating the cases of antenatal pulmonary embolism (APE), spontaneous hemoperitoneum in pregnancy (SHIP) and anaphylaxis in pregnancy. In conclusion the Belgian Obstetric Surveillance System has become a successful institute to monitor severe maternal morbidity in Belgium and a respectable partner within the INOSS group. The excellent participation rate and response rate demonstrate the willingness of Belgian ob/gyn specialists and their hunger for information on obstetric complications on a national basis.4 The results of the first studies invite to critically evaluate the current organisation of obstetric health care in Belgium and gave rise to the development of a national guideline on TOLAC.

Publications:

1. Vandenberghe G, De Blaere M, Van Leeuw V, Roelens K, Englert Y, Hanssens M, et al. Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2016 May 17;6(5):e010415. 2. Vandenberghe G, Bloemenkamp K, Berlage S, Colmorn L, Deneux-Tharaux C, Gissler M, et al. on behalf of INOSS. Results of the INOSS study of uterine rupture: a descriptive multi-country population based study. 2017. (Accepted for publication by BJOG, February 2018). 3. Vandenberghe G, Guisset M, Janssens I, Leeuw VV, Roelens K, Hanssens M, et al. A nationwide population-based cohort study of peripartum hysterectomy and arterial embolisation in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2017;7(11):e016208. 4. Vandenberghe G, Roelens K, Van Leeuw V, Englert Y, Hanssens M, Verstraelen H.The Belgian Obstetric Surveillance System to monitor Severe Maternal Morbidity. (Accepted for publication by Facts, Views and Visions in ObGyn. February 2018).

SAMENVATTING

Het Belgian Obstetric Surveillance System (B.OSS) werd in 2011 opgericht met de steun van het College voor Moeder en Pasgeborene van de FOD Volksgezondheid. B.OSS is een registratie-en evaluatiesysteem voor zeldzame ernstige obstetrische complicaties en aandoeningen in België. Het doel van B.OSS is door het meten van de ernstige maternale morbiditeit (severe maternal morbidity, SMM) en maternale near- miss (MNM), het voorlaatste stadium in een continuüm van probleemloze zwangerschap tot maternale sterfte, de kwaliteit en veiligheid van de obstetrische zorg te evalueren en waar mogelijk te verbeteren. “The key to improve is to measure”. B.OSS volgt hierin het voorbeeld van het UK Obstetric Surveillance System (UKOSS, https://www.npeu.ox.ac.uk/ukoss). Het verzamelen van gegevens gebeurt prospectief en actief, door een maandelijkse oproep aan alle Belgische materniteiten en volgens het principe van ‘nothing-to-report’. Voorts wordt bij een aangemelde casus een gedetailleerde vragenlijst ingevuld. Deze methodologie garandeert een minimum aan vals positieve en vals negatieve inclusies, in tegenstelling tot administratieve databases (zoals MKG en IMA). Via de website (www.b-oss.be) werkt B.OSS web- gebaseerd sinds augustus 2013, met een verbetering van de werklast voor deelnemers en team en van de anonimiteitsgarantie. De methodologie van UKOSS werd overgenomen door diverse landen, die zich verenigd hebben onder de noemer International Network of Obstetric Survey System (INOSS, https://www.npeu.ox.ac.uk/inoss). De INOSS groep maakt vergelijkende studies tussen landen mogelijk, alsook studies van uitzonderlijk zeldzame complicaties. B.OSS scoort vanaf zijn start excellent wat betreft deelname van ziekenhuizen (97.3%) en maandelijkse respons (98.9%). Dat Belgische gynaecologen bereid zijn mee te werken, ondanks hun drukke klinische praktijk, bewijst dat er nood is aan informatie op Belgisch niveau. De resultaten van de eerste registratieperiode (2012-2013) worden momenteel bekendgemaakt, via wetenschappelijke publicaties en presentaties, nieuwsbrieven, 2-jaarrapporten en de website. De studie naar uterusruptuur1 en de internationale vergelijking binnen de INOSS 2 leert ons dat de Belgische obstetrische zorg baat zou hebben bij een nationaal opgestelde richtlijn voor TOLAC (Trial Of Labour After Caesarean section), zodat vrouwen met een sectiolitteken goed geïnformeerd en zorgvuldig geselecteerd zouden worden. Het doel van de richtlijn is enerzijds onveilige procedures vermijden, welke geobserveerd werden in de B.OSS-studie, maar anderzijds ook streven naar een hoger aantal TOLAC’s en VBAC’s (Vaginal Birth After Caesarean). De richtlijn moet ook melding maken van het risico op uterusruptuur na andere uteriene chirurgie (myomectomie, hysteroscopische resectie van een uterien septum en salpingectomie). Bij zwangere vrouwen met dergelijke chirurgie in de voorgeschiedenis is een counseling en partusbeleid gerechtvaardigd gelijkaardig aan vrouwen met voorafgaande keizersnede.

De B.OSS studie naar peripartum hysterectomie en arteriële embolisatie3 heeft een beeld van hoe deze invasieve therapie wordt ingezet in de aanpak van massieve postpartumbloedingen in België. Idealiter moet deze studie worden opgevolgd door een studie naar àlle gevallen van ernstige postpartumbloeding in België, inclusief alle gevallen die succesvol zijn behandeld met tweedelijns interventies (intra-uteriene ballon en conservatieve chirurgische interventies). Verder vraagt deze studie naar een aanvullende audit-studie aan hand van de medische dossiers, om na te gaan of de uitgevoerde hysterectomieën en embolisaties mogelijk vermijdbaar waren. De hypothese is namelijk dat een peripartum hysterectomie sneller wordt uitgevoerd in de kleinere centra, terwijl embolisaties mogelijk te snel worden gedaan indien een dienst interventionale radiologie vlot beschikbaar is, zonder eerst eerste- en tweedelijnsinterventies toe te passen. We verwachten binnenkort de resultaten van de studie naar eclampsie in België. B.OSS registreert momenteel de gevallen van antenatale longembolen (APE), spontaan hemoperitoneum in de zwangerschap (SHiP) en anafylaxie in de zwangerschap. We concluderen dat B.OSS uitgebouwd is tot een efficiënt systeem voor het monitoren van ernstige maternale morbiditeit en tot een gerespecteerde partner binnen de INOSS. B.OSS heeft zijn succes te danken aan de vrijwillige medewerking van nagenoeg alle Belgische materniteiten. 4 De resultaten van de eerste registratieperiode nodigen uit om kritisch na te denken over de huidige organisatie van de obstetrische zorg in België en gaven aanleiding tot het opstellen van een nationale richtlijn voor TOLAC.

Publicaties: 1. Vandenberghe G, De Blaere M, Van Leeuw V, Roelens K, Englert Y, Hanssens M, et al. Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2016 May 17;6(5):e010415. 2. Vandenberghe G, Bloemenkamp K, Berlage S, Colmorn L, Deneux-Tharaux C, Gissler M, et al. on behalf of INOSS. Results of the INOSS study of uterine rupture: a descriptive multi-country population based study. 2017. (Accepted for publication by BJOG, February 2018). 3. Vandenberghe G, Guisset M, Janssens I, Leeuw VV, Roelens K, Hanssens M, et al. A nationwide population-based cohort study of peripartum hysterectomy and arterial embolisation in Belgium: results from the Belgian Obstetric Surveillance System. BMJ open. 2017;7(11):e016208. 4. Vandenberghe G, Roelens K, Van Leeuw V, Englert Y, Hanssens M, Verstraelen H.The Belgian Obstetric Surveillance System to monitor Severe Maternal Morbidity. (Accepted for publication by Facts, Views and Visions in ObGyn. February 2018).

CURRICULUM VITAE

Name Griet Vandenberghe

Date of birth 21 January 1978

Place of birth Roeselare, Belgium

Private address Auguste Van Ooststraat 11, 9050 Gentbrugge.

Professional Address Ghent University Hospital Department of Obstetrics and Gynaecology Corneel Heymanslaan 10 9000 Gent Belgium +32 9 33 27 849 E-Mail [email protected]

Marital status Partner of Wim Van Soom Mother of Frans (°2013), Riks (°2014) and Renée (°2017)

1996 - 1999 Katholieke Universiteit Leuven campus Kortrijk Belgium Kandidaat Arts (Bachelor) - 28/06/1999 Magna cum laude

1999 - 2003 Katholieke Universiteit Leuven Belgium Arts (Degree of Doctor of Medicine) – 26/06/2003 Magna cum laude

08/2003 - 07/2005 AZ Sint Lucas, Brugge Belgium Dr N Schockaert

08/2005 - 07/2006 UZ Gasthuisberg, Leuven Belgium Prof Dr I Vergote

08/2006 - 07/2007 Europaziekenhuizen - site Sint-Elisabeth, Brussel Belgium Dr S Vanderlinden

08/2007 - 07/2008 Watford General Hospital, Watford Herts United Mr AD Sanusi Kingdom

01/08/2008 Specialist in Obstetrics and Gynaecology Belgium

09/2008 - 08/2010 VUMC Amsterdam The Prof JMG van Vugt Netherlands

06/2010 - 08/2010 Lyell Mc Ewin Hosital, Adelaide Australia Women’s and Children’s Hospital, Adelaide Prof Dr GA Dekker

07/12/2010 Subspecialist in Perinatology The Recognition by the NVOG Netherlands

09/2010 - 08/2012 Resident Ghent University Hospital, Ghent, Belgium Department of Obstetrics & Gynaecology Maternal Intensive Care (MIC)

09/2012 - present Deputy head of clinic Ghent University Hospital, Ghent, Belgium Department of Obstetrics & Gynaecology Maternal Intensive Care (MIC)

2016 - 2019 Member of Vlaamse Vereniging voor Workgroup Obstetrics Obstetrie en Gynecologie (VVOG)

2016 - present Member of Instituut voor Training en Workgroup ITCIT Klinische Innovatieve Technologie Gent.

2011 - present Invited expert College of Physicians for Mother and Newborn, section Mother.

2011 - present Main Investigator Belgian Obstetric Surveillance System

04/2013 - present PhD Student Ghent University, Belgium

01/2015 -12/2016 Clinical PhD fellowship Research Foundation (part time) Flanders (FWO)

Academic year Begeleiding van de zwangere met diabetes. 2012 -2013 Posthogeschoolvorming Referentievroedvrouw Diabetes. - present VIVES Kortrijk

AY 2012 -2013 Capita Selecta Master in de Specialistische Geneeskunde. - present De zwangere vrouw in het ziekenhuismilieu. Ghent University

AY 2012 -2013 Borstvoeding en infecties, schildklierpathologie en - present contraceptie. Lesdag lactatiekunde. Ghent University Hospital

AY 2012 -2013 Fantoom onderwijs Schouderdystocie, Baringsmechanisme - present en vacuumextractie, Stuitbevalling. Postgraduaatlessen. Ghent University Hospital.

AY 2012 -2013 Mentorschap van groep van 8 studenten Geneeskunde - present vanaf eerste bachelor.Ghent University.

15/11/2013 Perinatale CMV-infectie. Bijscholingsavond Huisartsenkring Sint-Amandsberg.

04/2014 Verloskundige aspecten van pre-eclampsie. Avondsymposium Anesthesie. Ghent University Hospital.

02/2015 Wie wordt doorverwezen na een preconceptieconsultatie. Studiedag “Preconceptiezorg breed bekeken” VIVES Kortrijk

04/2016 Schildklierlijden in de zwangerschap. Vorming vroedvrouwen. Ghent University Hospital.

03/2015 Organisation of ‘Obstetric Skills and Drills’ 02/2016 Simulation Training in Obstetric Emergencies Maternity ward, Ghent University Hospital

AY 2015-2016 De normale zwangerschap. De normale bevalling. -present Pathologie tijdens arbeid, partus en postpartum. 1ste master REVAKI, Ghent University

11/2016 Begeleiding van de zwangere. Maternale pathologie. Avondcolloquia voor de practicus. Basisinformatie voor praktische counselling in de huisartspraktijk. Alumni Geneeskunde. Ghent University.

AY 2016-2017 Ernstige maternale morbiditeit en moedersterfte. Cursus Perinatale Zorg. Master of Science in de verpleegkunde en de vroedkunde.

1. Fertility and pregnancy outcome after fetoscopic surgery. L.Lewi, G.Vandenberghe, J.Deprest. Am J Obstet Gynecol 2003, 189(suppl) 6, 5597. Abstract

2. Lymphoblastic lymphoma presenting as bilateral gigantomastia in pregnancy. G.Vandenberghe, F.Claerhout, F. Amant. Int J Gynaecol Obstet 2005 Dec, 91930, 252-3.

3. First trimester screening for intrauterine growth restriction and early-onset pre- eclampsia. Vandenberghe G., Mensink I. , Blankenstein A.M., Twisk J.W.R, Vugt J.M.G. Prenatal Diagnosis 2011 Oct;31(10):955-61. doi: 10.1002/pd.2807. Epub 2011 Jun 30.

4. Placental Retention in Late First and Second Trimester Pregnancy Termination with Misoprostol: A Retrospective Analysis. Van der Knoop B., Vandenberghe G., Bolte A.C., Go A.T.J.I. J Matern Fetal Neonatal Med. 2012 Aug;25(8):1287-91. doi: 10.3109/14767058.2011.629257. Epub 2012 Jan 25.

5. Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance system. Vandenberghe G, De Blaere M, Van Leeuw V, Roelens K, Englert Y, Hanssens M, Verstraelen H. BMJ Open 2016;6:5 e010415 doi:10.1136/bmjopen-2015-010415

6. Nationwide population-based cohort study of peripartum hysterectomy and arterial embolisation in Belgium: Results of the Belgian Obstetric Surveillance System. Vandenberghe G, Guisset M, Janssens I, Van Leeuw V, Roelens K, Hanssens M, Russo E, Van Keirsbilck J, Englert Y, Verstraelen H. BMJ Open 2017;7:e016208. doi: 10.1136/bmjopen-2017-016208

7. Results of the INOSS study of uterine rupture: a descriptive multi-country population based study. Vandenberghe G, Berlage S, Colmorn L, Deneux-Tharaux C, Gissler M, Knight M, Lindqvist PG, Oberaigner W, Zwart J, Roelens K. On behalf of INOSS. Accepted for publication by BJOG, in press April 2018..

1. The Belgian Obstetric Surveillance System to monitor Severe Maternal Morbidity. Vandenberghe G, Roelens K, Van Leeuw V, Englert Y, Hanssens M, Verstraelen H. Accepted for publication by Facts, Views and Visions in ObGyn. Feburary 2018.

1. Maternale sterfte. Kristien Roelens, Griet Vandenberghe, Marleen Temmerman. Perinatale activiteiten in Vlaanderen: SPE (2011), volume 14, pp 89-97.

1. Handboek Verloskunde 2015. R Devlieger, Y Jacquemyn, M Laubach, K Roelens (ed) Acco ISBN 978-94-6292-300-3 Hoofdstuk 10. Stoornissen van het nageboortetijdperk en de postplacentaire periode. Hoofdstuk 29. Maternale en perinatale sterfte.

2. Syllabus Verloskunde Deel II Pathologie 2015. Roelens K, Roets E, Temmerman M Academia Press ISBN 2220150001315 Hoofdstuk 5. Hypertensieve aandoeningen tijdens de zwangerschap. Hoofdstuk 8. Dermatologische aandoeningen. Hoofdstuk 11. Endocrinologische aandoeningen.

1. Belgian Obstetric Surveillance System (B.OSS): eerste resultaten en vergelijking met de LEMMON-studie. G Vandenberghe, M Hanssens, K Roelens. Proceedings van het 19e Nederlands-Vlaams Doelencongres Infertiliteit, Gynaecologie en Obstetrie. 17, 18 en 19 april 2013.

2. Meten is weten, het belang van registraties en evaluaties. G Vandenberghe, M Hanssens, K Roelens. Proceedings van het 20e Nederlands-Vlaams Doelencongres Infertiliteit, Gynaecologie en Obstetrie. 22, 23 en 24 april 2015.

3. Belgian Obstetric Surveillance System: een stand van zaken. G Vandenberghe, M Hanssens, K Roelens, H Verstraelen.

Proceedings van het 21e Nederlands-Vlaams Doelencongres Infertiliteit, Gynaecologie en Obstetrie. 5,6 en 7 april 2017.

1. Lymphoblastic lymphoma presenting as bilateral gigantomastia in pregnancy. G.Vandenberghe, F. Claerhout, F. Amant. Poster. 14th International meeting of ESGO, Istanbul September 2005.

2. Urgent Caesarean Section for Intrapartum Fetal Distress: the role of Tocolysis in Fetal Resuscitation. R. Devlieger, G. Vandenberghe, M. Van de Velde. Poster. 8th World Congress of Perinatal Medicine, Florence, September 2007.

3. Placental growth factor as a screening marker for early preeclampsia. Vandenberghe G., Mensink I., Blankenstein A.M., Heymans M., Vugt JMG. Oral presentation. EBCOG European Congress of Obstetrics and Gynaecology, , May 2010.

4. Cervical teratoma at 20-weeks scan: value of US and MRI. Poster. Vandenberghe G, Linskens IH, Verbeke JIML, Kummerlin IPED, Vugt v JMG. ISPD 15th International Conference. Amsterdam, July 2010.

5. Registration and evaluation of three rare complications of pregnancy: the Belgian Obstetric Surveillance System (B.OSS). M. Guisset, A. Moens, J. Van Keirsbilck, M. De Blaere, G. Vandenberghe, M. Hanssens. Poster. 6th meeting of the International Society of Obstetric Medicine, Oxford, UK. July 2012.

6. Belgian Obstetric Surveillance system (B.OSS): eerste resultaten en vergelijking met de LEMMoN-studie. G. Vandenberghe, K. Roelens, M. Hanssens. Oral presentation 19e IGO Doelencongres, Rotterdam, april 2013.

7. Eclampsia in Flanders: a comparative study between Flanders, United Kingdom and the Netherlands. A. Langedock, M. Hanssens, G. Vandenberghe. Poster. ISSHP European Congress, Tromsø, Norway. June 2013.

8. Rare cases of uterine rupture, study proposition. G. Vandenberghe. On behalf of INOSS.

Oral presentation. Advanced course in Obstetric Emergencies, Copenhagen, Denmark. November 2013.

9. Meten is weten: het nut van registraties en evaluaties. G. Vandenberghe, M. Hanssens, K. Roelens. Oral presentation 20e IGO Doelencongres, Rotterdam. April 2015.

10. The International Network of Obstetric Survey Systems (INOSS): an international study of uterine rupture. Vandenberghe G. On behalf of INOSS. Oral presentation XXI FIGO World Congress of Gynecology and Obstetrics. Vancouver, Canada. 4-9 October 2015.

11. Belgian Obstetric Surveillance System, een stand van zaken. G Vandenberghe, M Hanssens, K Roelens, H Verstraelen. Oral presentation 21e IGO Doelencongres, Rotterdam. 6 April 2017.

12. International study of uterine rupture. G. Vandenberghe. On behalf of INOSS. Oral presentation. ‘Severe obstetric complications – state of the art – Conference’. Copenhagen, Denmark. 18 May 2017.

1. Uterusruptuur: registratie en evaluatie naar het model van UKOSS en LeMMON- studie. G. Vandenberghe, M. Hanssens. Oral presentation postuniversitaire studiedag, Leuven. May 2012.

2. Uterusruptuur in België, preliminaire resultaten van het Belgian Obstetric Surveillance System. M. De Blaere, G. Vandenberghe, Y. Englert, V. Van Leeuw, K. Roelens, M. Hanssens. Oral presentation. Assistentendag VVOG, Leuven. March 2014.

3. De eerste resultaten van de Belgian Obstetric Surveillance system (B.OSS). Vandenberghe G. Oral presentation. Congres MIC-NIC AZ St Jan, Brugge. October 2014.

4. Peripartum hysterectomy and arterial embolisation in Belgium. Results of the Belgian Obstetric Surveillance System.

M. Guisset, I. Janssens, G Vandenberghe, V. Van Leeuw, Y. Englert, K. Roelens, H Verstraelen, M Hanssens. Oral presentation. Assistentendag VVOG, Antwerpen. March 2016.

5. Antenatale longembolie in België: preliminaire resultaten van het Belgian Obstetric Surveillance System. A. Huybrechts, G. Vandenberghe, G. Page, V. Van Leeuw, Y. Englert, M. Hanssens, K. Roelens. Oral presentation. Assistentendag VVOG, Antwerpen. March 2016.

6. Aanbevelingen voor het verloskundig handelen bij uterusruptuur. Vandenberghe G. Oral presentation. Verloskunde in beweging III: de kracht van de consensus. 1 December 2016.

7. Eclampsie in België: resultaten van het Belgian Obstetric Surveillance System. A. Langedock, G. Vandenberghe, M. Hanssens, K. Van Calsteren. Oral presentation. Assistentendag VVOG, Brussel. April 2017.

1. Uterine rupture in Belgium: results of the Belgian Obstetric Surveillance System. G. Vandenberghe, H. Verstraelen, Y. Englert, M. Hanssens, K. Roelens. Oral presentation “Research Day 2015”, Ghent University, Faculty of Medicine and Health Sciences. March 2015.

2. Uterine rupture in Belgium: results of the Belgian Obstetric Surveillance System. Oral presentation Opleidingscluster Gynaecologie-Verloskunde UZ Gent, Gent. 4 March 2016.

3. Belgian Obstetric Surveillance system (B.OSS). Oral presentation Student Onderzoekssymposium, bezoek van Staatssecretaris Elke Sleurs. 26 April 2016.

4. Belgian Obstetric Surveillance system (B.OSS). Oral presentation Obstetrische Club, Gent. 17 May 2016

5. Peripartum hysterectomy and arterial embolisation in Belgium: results of the Belgian Obstetric Surveillance System. Oral presentation Opleidingscluster Gynaecologie-Verloskunde UZ Gent, Gent. 10 March 2017

1. Belgian Obstetric Surveillance system (B.OSS). Lok Vergadering ZOL Genk. 20/03/2014

2. Belgian Obstetric Surveillance system (B.OSS). Lok Vergadering KULeuven. 18/03/2016

Academic year 2011-2012 1. Literatuurstudie naar de etnische verschillen in het gebruik van contraceptie in België. Co-promotor masterproef Eline Marin, Master in de Geneeskunde.

2. De vroedvrouw in internationaal en intercultureel perspectief. In kaart brengen van maternale sterfte in Oeganda, met het accent op onveilige abortus. Externe promotor Bachelorproef Lore De Meutter, Bachelor in de verloskunde.

Academic year 2012-2013 3. Literatuurstudie naar de neonatale en lange termijn complicaties voor het kind van het gebruik van antidepressiva door de moeder in de zwangerschap Co-promotor scriptie Lenny Van den Durpel, Master in de Geneeskunde.

Academic year 2013-2014 4. De ontwikkeling van een zorgpad voor vrouwen met gestationele diabetes mellitus. Co-promotor masterproef Regine Goemaes, Master of Science in de Verpleegkunde en Vroedkunde.

Academic year 2013-2014 5. Implications of the implementation of the IADPSG-criteria in the screening for gestational and pregestational diabetes mellitus: a retrospective analysis. Co-promotor masterproef Amber Demey, Master of Science in de Verpleegkunde en Vroedkunde.

Academic year 2013-2014 6. Literatuuronderzoek: Eclampsie, een zeldzame complicatie van de zwangerschap. Incidentie, risicofactoren, uitkomst en beleid. Co-promotor masterproef Anne Vierin, Master in de Geneeskunde.

7. Literatuuronderzoek: Vruchtwaterembolie, een zeer zeldzame levensgevaarlijke complicatie van de zwangerschap. Incidentie, risicofactoren, uitkomst en beleid. Co-promotor masterproef Elise Landuyt, Master in de Geneeskunde.

8. Literatuuronderzoek: AFLP of acute vetlever syndroom, een zeer zeldzame levensgevaarlijke complicatie van de zwangerschap. Incidentie, risicofactoren, uitkomst en beleid. Co-promotor masterproef Tineke van Poucke, Master in de Geneeskunde.

9. Literatuuronderzoek: Rol van aspirine in en rond de zwangerschap. Bewezen en (nog) niet bewezen nut. Co-promotor masterproef Ilse van Baar, Master in de Geneeskunde.

10. Literatuuronderzoek: Stuitbevalling in Europa voor en na de Hannah-trial. Consequenties van een ophefmakende studie over de bevallingswijze bij stuitligging. Co-promotor masterproef Laura Baetens, Master in de Geneeskunde.

Academic year 2014-2015 11. Experimenteel onderzoek: Uterusruptuur als enstige zwangerschapscomplicatie: voorkomen in Vlaanderen. Marlies May en Emilie Vandevoorde. Co-promotor scriptie Master in de Geneeskunde.

12. Peripartum hysterectomie en embolisatie omwille van ernstige postpartumhemorrhagie: voorkomen in Vlaanderen. Iris Janssens Co-promotor scriptie Master in de Geneeskunde.

Academic year 2015-2016 13. Experimenteel onderzoek: Pre-eclampsie en tweelingen: is er een verband met zygositeit en chorioniciteit? Justine Gyssels. Co-promotor scriptie Master in de Geneeskunde.

Academic year 2016-2017 14. Uterine rupture in Belgium. Marlies De Blaere. Co-promotor masterproef Master in de Specialistische geneeskunde.

15. Antenatale longembolie in België: preliminaire resultaten van de Belgian Obstetric Surveillance System. Annejo Huybrechts. Co-promotor masterproef Master in de Specialistische geneeskunde.

ADDENDA

Nationwide population-based cohort study of uterine rupture in Belgium: results from the Belgian Obstetric Surveillance System. Supplementary files

Supplementary figure 1. Surgical risk factors in cases of uterine rupture. Supplementary figure 2. Uterine rupture cases by labour, by gestational age and by previous caesarean section (CS). Supplementary figure 3. Management in 90 cases of uterine rupture. Supplementary table 1. Risk of severe outcome of mother and neonate according to surgical risk factors.

Supplementary figure 1. Surgical risk factors in cases of uterine rupture.

Supplementary figure 2. Uterine rupture cases by labour, by gestational age and by previous caesarean section (CS).

Supplementary figure 3. Management in 90 cases of uterine rupture

Supplementary Table1. Relative risk of severe outcome of mother and neonate according to surgical risk factors: previous CS and/or previous other uterine surgery.

Previous CS No previous CS Relative Risk

Previous CS Previous CS and Other uterine No previous Surgical risk N, (%) other uterine surgery CS or other factors surgery N, (%) uterine N, (%) RR (95% CI) surgery RR (95% CI) N, (%) RR (95% CI)

Maternal outcome N= 90 N = 59 N=14 N=8 N=9 Neonatal N = 55 N =13 N=8 N=8 outcome N=84

Complete 57 (78) 17 (100) 1.2 (1.1-1.4) § rupture

§ Significant, 45 (76.2) 12 (85.7) 8 (100) 9 (100) p < 0.05 1 1.1 (0.8-1.4) 1.3 (1.1-1.4) § 1.3 (1.1-1.4) §

Rupture not in 18 (24.6) 15 (88.2) 3.5 (2.3-5.5) * lower segment * Significant, 14 (23.7) 4 (28.5) 8 (100) 7 (77.7) p < 0.05 1 1.0 (0.4-2.7) 4.2 (2.6-6.6) * 3.2 (1.8-5.8) *

21 (30.8) 6 (37.5) 1.2 (0.5-2.5) Infant in abdomen 19 (34.5) 2 (14.2) 3 (37.5) 3 (37.5) 1 0.4 (0.1-1.7) 1.1 (0.4-2.9) 1.1 (0.4-2.9)

Severe outcome 28 (41.1) 11 (68.7) 1.6 (1.0-2.5) $ infant $ Significant, 22 (40) 6 (46.1) 5 (62.5) 6 (75) p < 0.05 1 1.1 (0.5-2.2) 1.5 (0.8-2.9) 1.8 (1.1-3.1) $

Hysterectomy 3 (4.1) 5 (29.4) 7.1 (1.8-27) £ £ Significant, 1 (1.7) 2 (14.2) 3 (37.5) 2 (22.2) p < 0.05 1 8.4 (0.8-86) 22 (2.6-187) £ 13 (1.3-130) £

Mother 10 (13.6) 9 (52.9) 3.8 (1.8-8.0) # admitted to ICU 5 (55.5) 7 (11.8) 3 (21.4) 4 (50) # Significant, 4.6 (1.8-11.6) 1 1.8 (0.5-6.0) 4.2 (1.5-11.0) p < 0.05 #

Transfusion > 4 8 (10.9) 6 (35.2) 3.2 (1.2-8.0) ° PC ° Significant, 4 (6.7) 4 (28.5) 3 (37.5) 3 (33.3) p < 0.05 1 4.2 (1.1-14.8) ° 5.5 (1.5-10) ° 4.9 (1.3-18) °

Nationwide population-based cohort study of peripartum hysterectomy and arterial embolisation in Belgium: results from the Belgian Obstetric Surveillance System. Supplementary files

Supplementary figure 1. Antenatal suspicion and preventive measures in women with AIP. AIP, abnormally invasive placenta; FFP, fresh frozen plasma; IR, interventional radiology; RBC, red blood cells.

The INOSS study of uterine rupture: a descriptive multi-country population based study. Supplementary files

Table S1. The definitions of uterine rupture used by the different participating countries. Table S2. The set of 37 variables and codebook. Table S3. Registration criteria of live births and stillbirths in the Civil Birth Register and/or Perinatal Register per country. Table S4. Estimated prevalence (and 95% confidence interval) of complete uterine rupture per 10,000 deliveries in the different participating countries: overall and ≥ 24 weeks. Table S5. Estimated prevalence per 10,000 deliveries of uterine rupture in women with previous caesarean section and labouring, compared to caesarean section rate, trial of labour after caesarean section rate and vaginal birth after caesarean section rate in the different participating countries. Table S6. Estimated prevalence (and 95% confidence interval) of complete uterine rupture per 10,000 deliveries. Table S7. Estimated prevalence (and 95% confidence interval) per 10,000 deliveries of complete uterine rupture in women with and without previous caesarean section. Table S8. Management and maternal outcome in women with complete uterine rupture. Table S9. Absolute rate of severe maternal and perinatal outcome according to surgical risk factors: previous CS versus no previous CS. Table S10. Characteristics of women with complete uterine rupture and their fetuses/neonates, per country. Table S11. Perinatal outcome in neonates of women with complete uterine rupture. International comparison. Table S12. Ethics Committee of the participating population-based studies per country.

Figure S1. Possible pathways in parturient women with previous caesarean section. Figure S2. Correlation between uterine rupture rates in women with previous caesarean section and CS rates in the background population (rho=-0.917). Figure S3. Correlation between uterine rupture rates in women with previous caesarean section and trial of labour after caesarean section rates in the background population (rho=0.600)

Figure S4. Correlation between trial of labour after caesarean section rates in the background population and previous CS rates in the background population (rho=- 0.633).

Table S2. The set of 37 variables and codebook.

Previous obstetric history (NB: excluding index pregnancy)

1. Number of pregnancies with ≥24 ( ≥ 22) .. (continuous) completed weeks 2. Has the woman had previous C/S? Yes=1, No=2, Not known=3

3. Number of previous C/S .. (continuous)

4. Date of last C/S mm/yyyy

5. Youngest GA previous C/S (weeks) ww

6. In labour at the time of previous C/S Yes=1, No=2, Both=3, Not known=4

7. Type(s) of uterine incision previous C/S All low transverse=1, Any non-low- transverse=2, Not known=3

8. Type(s) of uterine closure previous C/S All double=1, All single=2, Mixture of double/single or other closure types=3, Not known=4

Previous medical history

9. Other uterine surgery (e.g. D&C, septal Yes=1, No=2, Not known=3 resection, myomectomy, polypectomy, endometrial resection/ablation, other) 10. Previous uterine perforation Yes=1, No=2, Not known=3

Current pregnancy

11. Final estimated date of delivery dd/mm/yyyy

Labour and uterine rupture

12. Was labour induced? Yes=1, No=2, Not known=3

13. Was prostaglandin used? Yes=1, No=2, Not known =3 , Not applicable=4

14. Specify type of prostaglandin used? Misoprostol (Cytotec ®)=1, Dinoproston (Prostin ®, Prepidil ®, Propess®)=2, iv PG=3, Not known=4, Not applicable=5, 1&2 = 6, 1&3=7, 2&3=8

15. Mechanical induction (intracervical Foley Yes=1, No=2, Not known=3 catheter, double balloon catheter, other)

16. Did the woman labour? Yes=1, No=2, Not known=3 (Specify definition of labour used..)

17. Syntocinon used Yes=1, No=2, Not known=3

a. for induction of labour Yes=1, No=2, Not known=3

b. for augmentation of labour Yes=1, No=2, Not known=3 (following spontaneous onset of labour) 18. Duration of syntocinon hh:mm

19. Position of rupture Fundal=1, Lower segment=2, Other=3, Not known=4

20. Other position of rupture specified Lateral=1, Posterior=2, Cornual=3, Longitudinal to fundus=4, Longitudinal to cervix/vagina=5, high transverse=6, not known=7, anterior wall of uterus=9, into broad ligament, extending to vagina=10; cervix = 11

(avoid use of : along old scar = 8)

21. Position of fetus at time of laparotomy Abdomen=1, Uterus=2, Other=3, Delivered=4, Not known=5

Management of uterine rupture

22. Hysterectomy Yes=1, No=2, Not known=3

23. Total units of packed red cells .. (continuous)

Outcome mother

24. Admitted to intensive care unit? Yes=1, No=2, Not known=3

25. Duration of stay at intensive care unit? days

26. Did the woman die? Yes=1, No=2, Not known=3

Outcome infant

(please provide information for all siblings in case of twins / triplets)

27. Date of birth dd/mm/yyyy

28. Mode of delivery C/S=1, Vaginal=2, Not known=3

29. Admitted to a neonatal unit? Yes=1, No=2, Not known=3, Not (please specify if NICU versus SCBU or other) applicable=4

30. PH umbilical artery ≤≤7 Yes=1, No=2, Not known=3, Not applicable=4

31. A. Apgar score at 5 min < 7 Yes=1, No=2, Not known=3, Not applicable=4

B. Apgar score at 5 min < 4 Yes=1, No=2, Not known=3, Not applicable=4

32. a. Neonatal encephalopathy Yes=1, No=2, Not known=3, Not applicable=4

32. b. Intubation / ventilation Yes=1, No=2, Not known=3, Not applicable=4

33. Did the infant die? Yes=1, No=2, Not known=3, N/A=4 (immature)

34. When did infant death occur? Antepartum=1, Intrapartum=2, Postpartum=3, Not known=4, Not applicable=5

35. Did infant have a major congenital anomaly? Yes=1, No=2, Not known=3

36. Was uterine rupture the direct cause of Yes=1, No=2, Not known=3, Not death? applicable=4

Table S12. Ethics Committee of the participating population-based studies per country.

The study was approved by the Ethikkomission der Mediznischen Austria Universität Innsbruck: no ethics committee approval is required for AuOSS retrospective studies.

The study was approved by the Medical Ethics Committee of the University Hospital Ghent (EC UZG 2012/734 B670201215359 and EC Belgium UZ Gent 2015/1470 B670201526875) and by the Medical Ethics B.OSS Committee of the University Hospital Brussels (EC ULB 2012/111; B406201213660)

France The study was approved by the Commission nationale de EPIMOMS l’Informatique et des Libertés (EGY/FLR/AR125069).

The study was approved by the Ethikkommission der Ärztekammer, Germany Niedersachsen, Körperschaft des öffentlichen Rechts, AZ = GerOSS (Bo/05/2011).

The The study was centrally approved by the Medical Ethics Committee of Netherlands Leiden University Medical Centre (P04-020;8 March 2004). LEMMoN

Ethical approval was obtained in each country according to national legislation. Denmark, In Denmark, the study was approved by the Danish Data Protection Finland, Agency. Sweden In Finland by the Ministry of Social Affairs and Health (Dnro NOSS THL/1697/5.05.00/2012). In Sweden by the Central Ethical Board for Medical Research (Ö 8- 2011).

United The study was approved by the London Research Ethics Committee (ref Kingdom 09/H0718/8) UKOSS

Figure S1. Possible pathways in parturient women with previous caesarean section (CS). ERCS: elective repeat caesarean section; TOLAC: trial of labour after caesarean section; VBAC: vaginal birth after caesarean section; EmCS: Emergency caesarean section when TOLAC failed.

Figure S2. Correlation between the uterine rupture rate in women with previous caesarean section and the previous CS rate in the background population (rho=-0.917)

Figure S3. Correlation between the uterine rupture rate in women with previous caesarean section and the trial of labour after caesarean section rate in the background population (rho=0.600).

Figure S4. Correlation between the trial of labour after caesarean section rate in the background population and the previous CS rate in the background population (rho=-0.633).

THANK YOU

Dank Hans. “Een doctoraat moet een plezier zijn “ Je leerde me de etiquette van het researchwereldje kennen en hoe ‘de puntjes op de i’ te zetten. Dank voor je enthousiaste reacties op papers waar ik vaak nauwelijks trots op was. Dank Kristien. “Het is allemaal een spelleke “ Dank om op het juiste moment de juiste dingen te zeggen. Al lijkt je hoofd soms in de wolken, je meningen zijn steeds nuchter en correct. Dank voor de fijne ritjes naar Brussel, eerst samen in de auto en later heel relaxt ‘in 1ste klas’. Dank Professor Hanssens, beste Myriam. Ik ben en zal u altijd zeer erkentelijk zijn voor een groots geschenk dat u me gaf: ik mocht mij ontfermen over uw B.OSS en heb deze baby ook stilaan mogen adopteren. Dank dat u er steeds was bij elke bespreking, of het nu in Gent of in Brussel was. En dank voor uw immer kritische visie. Merci Professeur Englert. Merci pour votre temps qui était énormément précieux. J’ai beaucoup apprecié votre contribution constructive et pragmatique pendant les reunions. Merci.

Merci Virginie, pour ta vision clairvoyante et ta franchise. Grâce à toi B.OSS a été mise sur la bonne voie. Merci Charlotte, pour ta gentilesse et tes réactions rapides. Merci Fatima, pour tes efforts pour B.OSS. Dank Stijn en Vincent van Nestor, de website was en is nog steeds een verademing. Dank voor de ‘vermakelijke’ communicatie. Dank Marlies De Blaere, het knalgroene logo heeft B.OSS aan jou te danken. Dank voor alle hulp en steun. Nu gaan we die richtlijn nog tot een goed einde brengen samen. Dank Iris Janssens, “met een grote omweg”, maar we zijn er geraakt. Ik vond het heel fijn samenwerken. Dank Anne Langedock, je eigen stijl siert je. Dank Annejo Huybrechts, voor je korte ‘price-winning’ bijdrage. Dank Caroline Roelens, om dat goeie werk verder te zetten. Dank Marine Guisset, Emilie Vandevoorde en Marlies Maly voor jullie B.OSS werk.

Thank you INOSS friends, for being such an inspiring, ambitious and easygoing group. It is always a pleasure to meet you all. Thank you Pelle, Mika, Maija, Jens, Lotte, Silvia, Nick, Willi, Joost, Jos, Kitty, Catherine and Marian, for giving me the chance to conduct the INOSS uterine rupture study and for giving me so many helping hands. Dank “Holland House”. Jos en Joost, dank voor jullie heerlijk positieve mailtjes, ze geven heel veel energie. Kitty, jouw enthousiasme is aanstekelijk. Het was een eer om met jou het bed in de Queen room te mogen delen. Timme en Thomas, dankzij jullie wordt elke meeting een gezellig feestje. En ik zweer: nooit verlaat ik nog voortijdig een Holland House om de rust op te zoeken.

Dank Anne De Haan, “aanvaard de chaos” zijn jouw wijze woorden die ik vaak luidop herhaal. Dank Ann-Sofie Van Parys, “omring je met mensen die het goed met je voor hebben”, daar heb ik oprecht veel aan gehad.

Dank Isabelle, mijn bureaumaatje -al wordt er weinig gewerkt als we er samen zitten- en supporter. Je bent bewonderswaardig efficiënt en een ontembare optimist. Dank om die lat hoog te leggen. Dank Ellen en Walter, voor de steun en de mogelijkheid om extra tijd te nemen om dit te kunnen afronden. Dank Tjalina, we zaten in hetzelfde schuitje en op dezelfde FWO-cruise. Dank voor tips en goeie raad. Dank Menekse, jammer dat je er vandaag niet bij kan zijn omdat je alweer in de bres sprong. We maken dit goed. Dank Magaly, je bent een gouden meid. Planning is heel belangrijk en je hebt me er fantastisch mee geholpen. Dank vroedvrouwen op de verloskamer, materniteit en poli, en in het bijzonder aan het babysittersclubje, om Renée te knuffelen terwijl haar PhD- moeder op de computer tokkelde (Johanna, Ine, Els en Valerie) Dank aan collega’s, arts-assistenten (oud en nieuw) en secretariaatsmedewerkers, voor alle steun.

Dank Roland, Je weet altijd het geknipte antwoord op mijn vaak banale problemen. Er kan weer gefeest worden. Dank Bart, Meike, Marieke, Lieven, Nathalie, Bram, Inge, Marleen, Suus, Priska, Leen, Isabel, Miet, Katrien, Delphi, Tim, Lieve, Sarah, Liesbeth, Anneleen, Anke, en nog meer vrienden en buren, voor steun en stoef en de goeie dingen van het leven. Dank Jean-Yves, een nanny als jij wens ik iedereen toe. Dank opa en oma, Jozef en Jeannine, Onze kinderen zijn dol op jullie (zeggen ze letterlijk) en jullie bezorgen ze een mooie tijd.

Dank mama en papa, “je moet afmaken waar je aan begint”, “niet te vele trunten” en vooral ook “niet te vele stoefen”. “Het kan altijd nog wat beter” heb ik volgens mij ook van jullie? Dankjewel om me alle kansen te geven. En dank voor de wekelijkse bezoekjes op woensdag en de vele andere keren dat jullie me uit de nood kwamen helpen, steeds denkende dat het ‘nu wel af zou zijn’. Danku Frannie, voor je luisterend oor en voor het mooie boekje. Je bent een topper. Danku Goelie, voor het binnenvallen af en toe en om op Frans en zijn zusjes te letten. Danku Eva, voor wijze raad. Komen logeren bij jullie is voor Frans en Riks een zaligheid. Danku Elke, voor het stijladvies. Het was niet simpel, maar het is goedgekomen. Danku Hanne, immer bereikbaar, en o zo lovend. Heerlijk om naar je lofzang te luisteren en dan te weerleggen. Danku Maarten, dat je er vandaag bij bent ondanks je drukke artiestenleven.

Dankjewel mijn lieve Wim, “Het is tijd om af te ronden” Ik heb je oeverloze geduld lang op de proef gesteld, maar eindelijk is het zover. Dank om me te temperen als het te dol werd en me aan te moedigen als het me teveel werd. Jij brengt me tegelijk rust en energie. Blij dat jij bij mij wil zijn. Frans, Riks en Renée, Straks zullen we “mama’s werk” ombouwen tot een grote speelkamer, de tekeningen mogen er allemaal blijven hangen.