University of Groningen. Pregnancy in women with congenital heart disease Kampman, Marlies Aleida Maria

Transcription

1 University of Groningen Pregnancy in women with congenital heart disease Kampman, Marlies Aleida Maria IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2016 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Kampman, M. A. M. (2016). Pregnancy in women with congenital heart disease: Complications and mechanisms [Groningen]: Rijksuniversiteit Groningen Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date:

2 Pregnancy in women with congenital heart disease Complications and Mechanisms Marlies A.M. Kampman

3 Financial support by Graduate school of Medical Sciences, University of Groningen, Amgen B.V., Servier Farma Nederland and Chipsoft B.V. for the publication of this thesis is gratefully acknowledged. Kampman, MAM Pregnancy in women with congenital heart disease: complications and mechanisms. ISBN: ISBN (electronic version): Cover photograph: Maxanel fotografie, Hoogeveen, the Netherlands ( Cover design: Optima Grafische Communicatie, Rotterdam, the Netherlands. Layout and printing: Optima Grafische Communicatie, Rotterdam, the Netherlands Marlies A.M. Kampman All rights are reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the written permission of the author.

4 Pregnancy in women with congenital heart disease Complications and mechanisms Proefschrift ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen op gezag van de rector magnificus prof. dr. E. Sterken en volgens besluit van het College voor Promoties. De openbare verdediging zal plaatsvinden op woensdag 16 maart 2016 om uur door Marlies Aleida Maria Kampman geboren op 28 mei 1985 te Ommen

5 Promotores Prof. dr. D.J. van Veldhuisen Prof. dr. C.M. Bilardo Copromotor Dr. P.G. Pieper Beoordelingscommissie Prof. dr. S.A. Scherjon Prof. dr. T. Ebels Prof. dr. A.H.E.M. Maas

6 Paranimfen Mw. E.H.A. Loeters Mw. J. Wolf The research described in this thesis was supported by a grant of the Dutch Heart Foundation (DHF-2007b75). Financial support by the Dutch Heart Foundation for the publication of this thesis is gratefully acknowledged.

7

8 Table of Contents Chapter 1 Introduction 9 Part I Complications Chapter 2 Chapter 3 Chapter 4 N-terminal pro-b-type natriuretic peptide predicts cardiovascular complications in pregnant women with congenital heart disease. European Heart Journal 2014;35: Cardiac adaption during pregnancy: comparison between women with congenital heart disease and healthy women. Submitted Cardiac function and cardiac events one year post-partum in women with congenital heart disease. American Heart Journal 2015;169: Part II Mechanisms Chapter 5 Chapter 6 Chapter 7 Uteroplacental Blood Flow, Cardiac Function and Pregnancy Outcome in Women with congenital heart disease. Circulation 2013;128: Uteroplacental Doppler flow and pregnancy outcome in women with Tetralogy of Fallot. Submitted Maternal cardiac function, uteroplacental Doppler flow parameters and pregnancy outcome: a systematic review. Ultrasound in Obstetrics and Gynecology 2015;46: Chapter 8 Summary and future perspectives 137 Appendices Summary in Dutch Nederlandse samenvatting 153 Acknowledgements Dankwoord 157 Bibliography Publicatielijst 163 Curriculum Vitae 167

9

10 Chapter 1 Introduction A

11

12 Introduction 11 Epidemiology Congenital heart defects (CHD) are the most common congenital defects in newborns, with a birth prevalence of 9/1000 live births worldwide 1,2. Due to improved surgical and medical treatment, the majority of these children survive until adulthood. This resulted in a large cohort of adult survivors, mostly still young and of child bearing age 3. It is important to realize that these women are not cured and that pregnancy can have deleterious effects on cardiac function and patient wellbeing since pregnancy requires major hemodynamic changes. 1 Physiological changes during normal pregnancy Pregnancy induces profound changes in the cardiovascular system, in order to meet the increased maternal and fetal metabolic demands. During normal pregnancy, cardiac output increases 30-50% above baseline, starting in early pregnancy and continues to rise until 24 weeks gestation. This augmented cardiac output is maintained until term (figure 1) 4. The increase in cardiac output results from an increase in preload (circulating volume), a decrease in afterload (peripheral vascular resistance) and an increase in heart rate. During labor, delivery and the post-partum period again profound changes in cardiac output occur, because of pain, anxiety, uterine contraction and uterine involution 4-6. The hemodynamic changes described result in progressive left ventricular remodeling. Left ventricular end-diastolic and end-systolic dimensions increase during pregnancy, where ejection fraction remains unchanged 7,8. The myocardium becomes hypertrophic as gestation advances and as a result of that, left ventricular mass increases. Despite the morphological changes, diastolic function is not impaired during pregnancy and left ventricular diastolic filling pressures remain unchanged 7,9. Studies describing right ventricular remodeling during pregnancy are extremely scarce. The available data suggests that right ventricular systolic function remains stable during pregnancy 10,11. Right ventricular diameter appears to increase towards the third trimester of pregnancy 10. Figure 1: Hemodynamic changes during pregnancy according to gestational week (modified from Robson et al. 4 ).

13 12 Chapter 1 Pregnancy in women with congenital heart disease Thirteen years of ZAHARA (Zwangerschap bij Aangeboren HARtAfwijkingen, pregnancy in congenital heart disease) research provided valuable insights to the knowledge of pregnancy in women with CHD and the research line evolved from epidemiology focused to more fundamental research focusing on the underlying pathophysiological mechanism of complications. The first ZAHARA study was conducted between 2002 and 2007 aiming to investigate pregnancy outcome in women with CHD and to identify predictors of cardiac complications during pregnancy retrospectively. Pregnancy in women with congenital heart disease appeared to be associated with increased incidence of cardiovascular, obstetric and neonatal complications In approximately 7-13% (depending on the underlying heart disease) of the pregnancies in women with CHD, cardiovascular complications occur, with heart failure and arrhythmia being most frequently observed 14, Siu et al. developed the CARPREG (Cardiac disease in pregnancy) risk score, in order to estimate the risk of cardiac complications during pregnancy in women with cardiac disease 24. The CARPREG risk score underestimated the risk of cardiac events in a cohort of women with solely CHD, which suggested that modifications in the risk score were needed 26. The ZAHARA study resulted in the development of a new risk estimation model for the occurrence of cardiovascular complications during pregnancy in women with CHD, and also identified new predictors of adverse neonatal outcome 23. The study resulted in the thesis that was defended by Wim Drenthen ( ). The first ZAHARA study, as well as other reports, revealed that especially placenta-related complications are more frequently observed in women with CHD (ie. hypertensive disorders of pregnancy and fetal growth restriction) 12-14,27. The ZAHARA II study was primarily designed to shed light on a possible underlying mechanism for the increased incidence. In patients with chronic heart failure, worsening cardiac function is associated with dysfunction of other organs 28,29 and a similar mechanism might be present during pregnancy, during which cardiac dysfunction hampers placental development and function leading to adverse pregnancy outcome. We postulated that maternal cardiac dysfunction is responsible for abnormal placental development and placental dysfunction. In pregnant women without underlying heart disease, inadequate adaption of the uteroplacental circulation is responsible for several obstetric and offspring complications. The pathophysiological mechanism is poor trophoblast invasion of the spiral arteries during the placentation process 30, causing failure of the placental-bed arteries to transform from high to low-resistance vessels. Uteroplacental flow investigations (resistance and pulsatility indices of the uterine artery and umbilical artery) provide insight in the placentation process and are commonly used as screening tool to predict the future development of pre-eclampsia, fetal growth restriction, still birth and placental abruption 31,32. In ZAHARA II we investigated the effect of impaired cardiac function

14 Introduction 13 on the uteroplacental circulation and its relationship with the occurrence of adverse obstetric and offspring outcome 33. The ZAHARA II study was conducted between March 2008 and August This study also provided the opportunity to validate the ZAHARA prediction model and to compare it with the other risk prediction models used. It turned out to perform better in a population with CHD compared to the CARPREG risk score 34. These findings, as well as the study design of ZAHARA II were described in the thesis of Ali Balci, defended at The increased volume load is thought to play an important role in the pathogenesis of the cardiovascular complications and therefore natriuretic peptides (B-type natriuretic peptide (BNP) and amino terminal probnp (Nt-proBNP)) may be an important predictor of cardiovascular events. Natriuretic peptides are well established predictors of adverse outcome in various cardiac diseases and predict outcome independent of ejection fraction Data on the predictive role for cardiac event during pregnancy in women with CHD are scarce. Tanous et al. reported a clear association between high BNP values and cardiovascular events during pregnancy, but could not determine the role of BNP in predicting adverse cardiovascular events 40. Natriuretic peptides can provide a valuable tool for the clinicians caring for these patients in order to identify high risk patients, needing close follow up. The prospective nature of ZAHARA II allowed a comparison of Nt-proBNP between women with CHD and healthy controls. In addition, the role of Nt-proBNP in predicting cardiovascular events during pregnancy could be assessed. In women with congenital heart disease the progressive cardiac remodeling during pregnancy might have serious repercussions for cardiac function after pregnancy, and therefore might influence their prognosis. Longitudinal data on cardiac remodeling during and after pregnancy in women with CHD is scarce and most of the available research focuses on left ventricular parameters. Data examining longitudinal changes in right ventricular parameters have never been reported in pregnant women with CHD. Several studies indicate that pregnancy can be associated with permanent deterioration in cardiac function and impaired event-free survival The secondary objective of ZAHARA II was to investigate the incidence of permanent post-partum cardiovascular deterioration in women with CHD, since data is scarce and existing reports mainly describe small retrospective study populations.

15 14 Chapter 1 Aims of the thesis In this thesis we describe the incidence of cardiovascular complications and we report new predictors for cardiovascular complications during and after pregnancy, furthermore we assess the impact of pregnancy on cardiac function and remodeling. Finally, we investigate whether cardiac dysfunction plays a role in the pathogenesis of obstetric and neonatal complications. In part I of this thesis, we focus on the cardiovascular complications during and after pregnancy and we assess the impact of pregnancy on cardiac function and remodeling. In chapter 2 the independent role of Nt-proBNP levels during pregnancy in women with CHD in predicting the occurrence of cardiovascular events will be assessed. Chapter 3 will focus on cardiac adaption during pregnancy in women with CHD compared to healthy pregnant women. The incidence of cardiovascular complications one year post-partum will be described and cardiac function parameters pre-pregnancy and one year post-partum will be compared in chapter 4. In part II of this thesis we investigate the role of cardiac dysfunction in the pathogenesis of obstetric and neonatal complications. Chapter 5 describes the main outcomes of the ZA- HARA II study, which assesses the differences in uteroplacental Doppler flow (UDF) patterns as well as the differences in outcome in pregnant women with CHD and healthy pregnant women. The relationship between cardiac dysfunction and uteroplacental Doppler flow patterns will be investigated. Chapter 6 will focus on pregnancy outcome and complications in women with Tetralogy of Fallot. The outcome of the ZAHARA II study will be verified in a homogeneous population. The final chapter of this thesis, chapter 7, reviews the literature systematically in order to investigate all existing evidence for a link between maternal cardiac function, abnormal uteroplacental flow and poor perinatal outcome in women with and without known cardiac disease.

16 Introduction 15 References 1 1. European Surveillance of Congenital Anomalies (EUROCAT) Working Group. 2011; Available at: Accessed 03/14, van der Linde D, Konings EE, Slager MA, Witsenburg M, Helbing WA, Takkenberg JJ, et al. Birth prevalence of congenital heart disease worldwide: a systematic review and meta-analysis. J Am Coll Cardiol 2011 Nov 15; 58(21): Marelli AJ, Mackie AS, Ionescu-Ittu R, Rahme E, Pilote L. Congenital heart disease in the general population: changing prevalence and age distribution. Circulation 2007 Jan 16; 115(2): Robson SC, Hunter S, Boys RJ, Dunlop W. Serial study of factors influencing changes in cardiac output during human pregnancy. Am J Physiol 1989 Apr; 256(4 Pt 2): H Robson SC, Dunlop W, Boys RJ, Hunter S. Cardiac output during labour. Br Med J (Clin Res Ed) 1987 Nov 7; 295(6607): Robson SC, Dunlop W, Hunter S. Haemodynamic changes during the early puerperium. Br Med J (Clin Res Ed) 1987 Apr 25; 294(6579): Estensen ME, Beitnes JO, Grindheim G, Aaberge L, Smiseth OA, Henriksen T, et al. Altered maternal left ventricular contractility and function during normal pregnancy. Ultrasound Obstet Gynecol 2013 Jun; 41(6): Savu O, Jurcut R, Giusca S, van Mieghem T, Gussi I, Popescu BA, et al. Morphological and functional adaptation of the maternal heart during pregnancy. Circ Cardiovasc Imaging 2012 May 1; 5(3): Fok WY, Chan LY, Wong JT, Yu CM, Lau TK. Left ventricular diastolic function during normal pregnancy: assessment by spectral tissue Doppler imaging. Ultrasound Obstet Gynecol 2006 Nov; 28(6): Ducas RA, Elliott JE, Melnyk SF, Premecz S, dasilva M, Cleverley K, et al. Cardiovascular magnetic resonance in pregnancy: insights from the cardiac hemodynamic imaging and remodeling in pregnancy (CHIRP) study. J Cardiovasc Magn Reson 2014 Jan 3; 16(1): 1-429X Vogt M, Muller J, Kuhn A, Elmenhorst J, Muhlbauer F, Oberhoffer A. Cardiac Adaptation of the Maternal Heart During Pregnancy: A Color-Coded Tissue Doppler Imaging Study - Feasibility, Reproducibility and Course during Pregnancy. Ultraschall Med 2014 Apr Drenthen W, Pieper PG, Ploeg M, Voors AA, Roos-Hesselink JW, Mulder BJ, et al. Risk of complications during pregnancy after Senning or Mustard (atrial) repair of complete transposition of the great arteries. Eur Heart J 2005 Dec; 26(23): Drenthen W, Pieper PG, Roos-Hesselink JW, van Lottum WA, Voors AA, Mulder BJ, et al. Pregnancy and delivery in women after Fontan palliation. Heart 2006 Sep; 92(9): Drenthen W, Pieper PG, Roos-Hesselink JW, van Lottum WA, Voors AA, Mulder BJ, et al. Outcome of pregnancy in women with congenital heart disease: a literature review. J Am Coll Cardiol 2007 Jun 19; 49(24): Drenthen W, Pieper PG, Roos-Hesselink JW, Schmidt AC, Mulder BJ, van Dijk AP, et al. Noncardiac complications during pregnancy in women with isolated congenital pulmonary valvar stenosis. Heart 2006 Dec; 92(12): Drenthen W, Pieper PG, Roos-Hesselink JW, Zoon N, Voors AA, Mulder BJ, et al. Fertility, pregnancy, and delivery after biventricular repair for pulmonary atresia with an intact ventricular septum. Am J Cardiol 2006 Jul 15; 98(2):

17 16 Chapter Drenthen W, Pieper PG, Zoon N, Roos-Hesselink JW, Voors AA, Mulder BJ, et al. Pregnancy after biventricular repair for pulmonary atresia with ventricular septal defect. Am J Cardiol 2006 Jul 15; 98(2): Yap SC, Drenthen W, Meijboom FJ, Moons P, Mulder BJ, Vliegen HW, et al. Comparison of pregnancy outcomes in women with repaired versus unrepaired atrial septal defect. BJOG 2009 Nov; 116(12): Yap SC, Drenthen W, Pieper PG, Moons P, Mulder BJ, Klieverik LM, et al. Outcome of pregnancy in women after pulmonary autograft valve replacement for congenital aortic valve disease. J Heart Valve Dis 2007 Jul; 16(4): Yap SC, Drenthen W, Pieper PG, Moons P, Mulder BJ, Vliegen HW, et al. Pregnancy outcome in women with repaired versus unrepaired isolated ventricular septal defect. BJOG 2010 May; 117(6): Yap SC, Drenthen W, Pieper PG, Moons P, Mulder BJ, Mostert B, et al. Risk of complications during pregnancy in women with congenital aortic stenosis. Int J Cardiol 2008 May 23; 126(2): Balci A, Drenthen W, Mulder BJ, Roos-Hesselink JW, Voors AA, Vliegen HW, et al. Pregnancy in women with corrected tetralogy of Fallot: Occurrence and predictors of adverse events. Am Heart J 2011 Feb; 161(2): Drenthen W, Boersma E, Balci A, Moons P, Roos-Hesselink JW, Mulder BJ, et al. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010 Sep; 31(17): Siu SC, Sermer M, Colman JM, Alvarez AN, Mercier LA, Morton BC, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001 Jul 31; 104(5): Roos-Hesselink JW, Ruys TP, Stein JI, Thilen U, Webb GD, Niwa K, et al. Outcome of pregnancy in patients with structural or ischaemic heart disease: results of a registry of the European Society of Cardiology. Eur Heart J 2013 Mar; 34(9): Khairy P, Ouyang DW, Fernandes SM, Lee-Parritz A, Economy KE, Landzberg MJ. Pregnancy outcomes in women with congenital heart disease. Circulation 2006 Jan 31; 113(4): Siu SC, Colman JM, Sorensen S, Smallhorn JF, Farine D, Amankwah KS, et al. Adverse neonatal and cardiac outcomes are more common in pregnant women with cardiac disease. Circulation 2002 May 7; 105(18): Damman K, van Deursen VM, Navis G, Voors AA, van Veldhuisen DJ, Hillege HL. Increased central venous pressure is associated with impaired renal function and mortality in a broad spectrum of patients with cardiovascular disease. J Am Coll Cardiol 2009 Feb 17; 53(7): van Deursen VM, Damman K, Hillege HL, van Beek AP, van Veldhuisen DJ, Voors AA. Abnormal liver function in relation to hemodynamic profile in heart failure patients. J Card Fail 2010 Jan; 16(1): Prefumo F, Sebire NJ, Thilaganathan B. Decreased endovascular trophoblast invasion in first trimester pregnancies with high-resistance uterine artery Doppler indices. Hum Reprod 2004 Jan; 19(1): Cnossen JS, Morris RK, ter Riet G, Mol BW, van der Post JA, Coomarasamy A, et al. Use of uterine artery Doppler ultrasonography to predict pre-eclampsia and intrauterine growth restriction: a systematic review and bivariable meta-analysis. CMAJ 2008 Mar 11; 178(6): Harrington K, Cooper D, Lees C, Hecher K, Campbell S. Doppler ultrasound of the uterine arteries: the importance of bilateral notching in the prediction of pre-eclampsia, placental abrup-

18 Introduction 17 tion or delivery of a small-for-gestational-age baby. Ultrasound Obstet Gynecol 1996 Mar; 7(3): Balci A, Sollie KM, Mulder BJ, de Laat MW, Roos-Hesselink JW, van Dijk AP, et al. Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study. Am Heart J 2011 Feb; 161(2): e Balci A, Sollie-Szarynska KM, van der Bijl AG, Ruys TP, Mulder BJ, Roos-Hesselink JW, et al. Prospective validation and assessment of cardiovascular and offspring risk models for pregnant women with congenital heart disease. Heart 2014 Sep; 100(17): van Veldhuisen DJ, Linssen GC, Jaarsma T, van Gilst WH, Hoes AW, Tijssen JG, et al. B-type natriuretic Peptide and prognosis in heart failure patients with preserved and reduced ejection fraction. J Am Coll Cardiol 2013 Apr 9; 61(14): Kirk V, Bay M, Parner J, Krogsgaard K, Herzog TM, Boesgaard S, et al. N-terminal probnp and mortality in hospitalised patients with heart failure and preserved vs. reduced systolic function: data from the prospective Copenhagen Hospital Heart Failure Study (CHHF). Eur J Heart Fail 2004 Mar 15; 6(3): Kragelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-terminal pro-b-type natriuretic peptide and long-term mortality in stable coronary heart disease. N Engl J Med 2005 Feb 17; 352(7): Linssen GC, Bakker SJ, Voors AA, Gansevoort RT, Hillege HL, de Jong PE, et al. N-terminal pro-btype natriuretic peptide is an independent predictor of cardiovascular morbidity and mortality in the general population. Eur Heart J 2010 Jan; 31(1): Omland T, Persson A, Ng L, O Brien R, Karlsson T, Herlitz J, et al. N-terminal pro-b-type natriuretic peptide and long-term mortality in acute coronary syndromes. Circulation 2002 Dec 3; 106(23): Tanous D, Siu SC, Mason J, Greutmann M, Wald RM, Parker JD, et al. B-type natriuretic peptide in pregnant women with heart disease. J Am Coll Cardiol 2010 Oct 5; 56(15): Uebing A, Arvanitis P, Li W, Diller GP, Babu-Narayan SV, Okonko D, et al. Effect of pregnancy on clinical status and ventricular function in women with heart disease. Int J Cardiol 2010 Feb 18; 139(1): Guedes A, Mercier LA, Leduc L, Berube L, Marcotte F, Dore A. Impact of pregnancy on the systemic right ventricle after a Mustard operation for transposition of the great arteries. J Am Coll Cardiol 2004 Jul 21; 44(2): Balint OH, Siu SC, Mason J, Grewal J, Wald R, Oechslin EN, et al. Cardiac outcomes after pregnancy in women with congenital heart disease. Heart 2010 Oct; 96(20): Silversides CK, Colman JM, Sermer M, Farine D, Siu SC. Early and intermediate-term outcomes of pregnancy with congenital aortic stenosis. Am J Cardiol 2003 Jun 1; 91(11): Tzemos N, Silversides CK, Colman JM, Therrien J, Webb GD, Mason J, et al. Late cardiac outcomes after pregnancy in women with congenital aortic stenosis. Am Heart J 2009 Mar; 157(3): Zentner D, Wheeler M, Grigg L. Does pregnancy contribute to systemic right ventricular dysfunction in adults with an atrial switch operation? Heart Lung Circ 2012 Aug; 21(8): Kamiya CA, Iwamiya T, Neki R, Katsuragi S, Kawasaki K, Miyoshi T, et al. Outcome of pregnancy and effects on the right heart in women with repaired tetralogy of fallot. Circ J 2012; 76(4):

19

20 Part I 8 Complications A

21

22 Chapter 2 N-terminal pro-b-type natriuretic peptide predicts cardiovascular complications in pregnant women with congenital heart disease Marlies A.M. Kampman Ali Balci Dirk J. van Veldhuisen Arie P.J. van Dijk Jolien W. Roos-Hesselink Krystyna M. Sollie-Szarynska Marieke Ludwig-Ruitenberg Joost P. van Melle Barbara J.M. Mulder Petronella G. Pieper on behalf of the ZAHARA II investigators 7 8 A European Heart Journal 2014;35:

23 22 Chapter 2 Abstract Aims In women with congenital heart disease (CHD), cardiovascular complications during pregnancy are common, but the risk assessment of these patients remains difficult. This study sought to determine the independent role of N-terminal pro B-type natriuretic peptide (NtproBNP) levels in addition to other parameters in predicting adverse cardiovascular events during pregnancy in women with CHD. Methods We conducted a national, prospective multicenter cohort study. Follow-up with clinical evaluation and echocardiography and Nt-proBNP measurement was performed at 20 week gestation. Results Adverse cardiovascular events occurred in 10.3% of 213 pregnancies. N-terminal pro B-type natriuretic peptide levels > 128 pg/ml at 20 weeks gestation, the presence of a mechanical valve and subpulmonary ventricular dysfunction before conception were independently associated with events (odds ratio (OR) 10.6 (p=0.039), OR 12.0 (p=0.016) and OR 4.2 (p=0.041), respectively). The negative predictive value of Nt-proBNP levels < 128 pg/ml was 96.9%. N-terminal pro B-type natriuretic peptide levels > 128 pg/ml at 20 weeks of gestation had an additional value in predicting the occurrence of adverse cardiovascular events on top of the other identified predictors (area under the curve 0.90 versus 0.78, p=0.035). Conclusions Increased Nt-proBNP levels at 20 weeks of gestation are an independent risk predictor of cardiovascular events during pregnancy in women with CHD. Key words: congenital heart disease, pregnancy, natriuretic peptides, right ventricular function.

24 Nt-proBNP in pregnant women with CHD 23 Introduction Nowadays, the majority of patients with congenital heart disease (CHD) reach childbearing age. Although many women with CHD experience uncomplicated pregnancies, a substantial sub-group suffers from cardiovascular, obstetric and offspring complications during pregnancy 1. Many predictors of complications have been identified and several prediction models have been developed, but the identification of women who will experience cardiovascular complications remains difficult 1-7. As the increased volume load during pregnancy is thought to play a central role in the pathogenesis of cardiovascular events, natriuretic peptides may be an important predictor of cardiovascular complications during pregnancy. B-type natriuretic peptide (BNP) and N-terminal pro B-type natriuretic peptide (Nt-proBNP) are well established predictors of adverse outcome in various cardiac diseases 8-11 and predict outcome in patients with heart failure independent of ejection fraction 12. Data on their role in risk evaluation for adverse pregnancy outcome are scarce. The only study considering BNP levels during pregnancy in women with congenital and acquired heart disease found a clear association between high BNP levels and cardiovascular events, but was not able to determine the role of BNP in predicting adverse cardiovascular events 13. The primary objectives of the ZAHARA II (Zwangerschap bij Aangeboren HARtAfwijkingen; Pregnancy in CHD) study are described previously 14. The aims of the present study were (i) to determine the independent role of Nt-proBNP levels during pregnancy in predicting the occurrence of cardiovascular complications during pregnancy in women with CHD; (ii) to identify other independent predictors of cardiovascular complications in pregnant women with CHD; and (iii) to describe the nature and incidence of cardiovascular complications during pregnancy in this population. 2 Methods Study design As reported in detail elsewhere, the ZAHARA II study is a prospective observational multicentre cohort study 14. All consecutive pregnant women with structural CHD, aged 18, presenting in one of the eight participating centres between March 2008 and August 2011, were eligible for enrolment. Miscarriages or pregnancies terminated before 20 weeks gestation were excluded. To determine normal Nt-proBNP levels during pregnancy, healthy pregnant women, without chronic medication use, were recruited from midwife practices in Groningen and Rotterdam, the Netherlands. Women with known illicit drug or alcohol abuse were excluded. The Research Ethics Committee of all participating centers approved the study protocol and all participating women gave written informed consent.

25 24 Chapter 2 Baseline characteristics Medical records were used to collect baseline data at the first prenatal visit (at 20 weeks gestation). Baseline data included underlying congenital anomaly, prior interventions, cardiac sequelae, previous cardiac events, comorbidity and obstetric history. Data related to cardiac function prior to pregnancy (including data on New York Heart Association functional class, physical examination, medication use, laboratory values, echocardiographic and electrocardiographic registrations and modified WHO classification of pregnancy risk 7 ) were also recorded. During follow-up visits at 20 and 32 weeks of gestation, clinical evaluation, echocardiography, electrocardiography and 24 hour electrocardiographic registrations, as well as Nt-proBNP measurements were performed. Serum creatinine was measured in all patients to exclude renal dysfunction. Healthy pregnant women underwent identical evaluation. All echocardiographic recordings were evaluated offline in one of the participating centers by 4 experienced cardiologists, blinded to the endpoints. Chamber quantification, valvular function, systolic and diastolic ventricular function were assessed according to the current recommendations We defined values > 95 th percentile of the Nt-proBNP values in healthy women as elevated, since no generally accepted normal values for Nt-proBNP during pregnancy exist. In this paper, data of 20 week gestation are evaluated for their value in predicting the occurrence of cardiovascular complications during pregnancy. Cardiovascular events Cardiovascular events were recorded during pregnancy and up to 6 months post-partum. Primary cardiovascular events were defined as any of the following: need for an urgent invasive cardiovascular procedure, heart failure (according to the guidelines of the European Society of Cardiology and documented by the attending physician 19 ), new onset or symptomatic tachy- or brady arrhythmia requiring new or extended treatment, thromboembolic events, myocardial infarction, cardiac arrest, cardiac death, endocarditis and aortic dissection. New York Heart Association (NYHA) class deterioration 2 points during pregnancy or within the first six months post-partum was defined as a secondary cardiovascular event 14. Statistical analysis Descriptive statistics for categorical data were expressed in absolute numbers and percentages. Mean and standard deviations or median with interquartile ranges were presented for continuous variables, depending on their distribution. For intergroup comparisons, the Student s t-test, Mann-Whitney U test, Chi-square and Fisher s exact test were used as appropriate. Univariable logistic regression was performed to identify variables associated with the composite endpoint of primary cardiovascular events. The following preconception variables were assessed: disease complexity, modified WHO risk classification 6,7, the presence of a systemic right ventricle, NYHA functional class, maternal age, parity, body surface area,

26 Nt-proBNP in pregnant women with CHD 25 resting heart rate, the presence of sinus rhythm, mean arterial pressure, smoking, use of cardiac medication, prior cardiac events, valve regurgitation and stenosis, systemic ventricular size and function, left atrial volume index, subpulmonary ventricular size and function and Nt-proBNP levels. Systemic right ventricles were excluded from quantitative analysis of echocardiographic chamber function measurements only (tricuspid annular plane systolic excursion (TAPSE) and ventricular ejection fraction determination), since these measurements are invalid in systemic right ventricles. Because of the limited amount of endpoints, only the five variables most significantly associated with an increased incidence of the composite endpoint entered the multivariable model. Bootstrap selection methods were used to avoid an over-fit model. Two hundred bootstrap samples were selected and predictors selected in > 130 models (65%) were included in the final model. The final model was constructed by backward deletion of the least significant variable until all remaining variables were significantly associated with the endpoint. To identify the additional role of Nt-proBNP levels at 20 weeks gestation as a discriminator between patients at risk for complications, receiver operating curves (ROCs) of the initial multivariable model and the initial multivariable model plus Nt-proBNP at 20 weeks gestation were constructed and compared. Cardiovascular complications occurring before 20 weeks gestation and valve thrombosis were excluded from this analysis. Statistical analysis was performed using SPSS (IBM SPSS statistics, version 20.0, Chicago, IL, USA) and STATA software package (version 11, College Station, TX, USA). A two-tailed p-value < 0.05 was considered significant. 2 Results Baseline characteristics We identified 234 women who were potentially eligible for participation. Twenty-one women were excluded, because of miscarriage (n=11), serious protocol violation (n=6) and withdrawal of informed consent (n=4). A total of 213 pregnancies in 203 women were observed (209 singleton and 4 twin pregnancies). Maternal baseline characteristics are shown in table 1; underlying CHD is summarized in table 2. Results are reported per pregnancy. The majority of patients had a mildly or moderately increased risk of cardiovascular complications during pregnancy, as indicated by the modified WHO class. Most patients were in NYHA class I or II. The majority of the patients did not use any medication 6 months prior to pregnancy; two patients who were on ACE-inhibitors discontinued use before conception. Systemic ventricular ejection fraction < 45% was seen in 7.9% of the patients; 14.3% had systolic dysfunction of the subpulmonary ventricle.

27 26 Chapter 2 Table 1. Maternal baseline characteristics (prior to pregnancy; N = 213 pregnancies) N (%) Demographics Maternal age at conception (years ± SD) 28.7 (± 4.4) Parity status Smoking prior to pregnancy NYHA class I II III Modified WHO class* I II III IV Past medical history Mechanical valve prosthesis Sustained symptomatic brady- or tachyarrhythmia requiring treatment Pacemaker Subacute bacterial endocarditis Congestive heart failure Cerebrovascular accident Hypertension Medication use prior to pregnancy None ACE-inhibitor Beta-blocker Calciumchannel blocker Digoxin Anticoagulants Echocardiographic parameters Systemic AV valve regurgitation Pulmonary AV valve regurgitation Pulmonary valve stenosis Pulmonary valve regurgitation Aortic valve stenosis Aortic valve regurgitation Systemic ventricular systolic dysfunction Subpulmonary ventricular systolic dysfunction** * Modified World Health Organisation class according to ESC guidelines.7 Moderate or severe regurgitation; peak gradient 36 mmhg; Ejection fraction < 45%; ** TAPSE < 16 mm.

28 Nt-proBNP in pregnant women with CHD 27 Cardiovascular events We observed primary cardiovascular events in 22 (10.3%) pregnancies (table 2). Six of these patients (27.3%) had a history of previous cardiovascular events. Five patients (22.7%) had > 1 cardiovascular event during pregnancy. Two patients (0.9%) required an urgent invasive procedure. One patient with compromised right ventricular function needed valve replacement two weeks post-partum because of mechanical valve thrombosis and a patient with a subvalvular aortic stenosis received a pacemaker because of a second degree atrioventricular block two months after pregnancy. Eight patients (3.8%) developed heart failure (including the patient with pacemaker implantation); seven of these patients (87.5%) developed heart failure during the third trimester or post-partum and one patient during the second trimester. Fourteen patients (6.6%) had arrhythmias, mainly supraventricular arrhythmias, occurring from the second trimester until the post-partum period. Four patients (1.9%) had thrombosis of their bileaflet mechanical valve; two during the first trimester, one during the 18 th week of pregnancy and one in the post-partum period. Three of these patients developed valve thrombosis while on low molecular weight heparin treatment. One patient (0.5%) developed a type B aortic dissection two weeks post-partum. No mortality, cardiac arrest, endocarditis or myocardial infarction occurred. Thirty-nine patients (18.3%) had a decline of 2 points in NYHA functional class during pregnancy or within 6 months post-partum compared with pre-pregnancy. 2 N-terminal pro-b-type natriuretic peptide during pregnancy N-terminal pro-b-type natriuretic peptide was available in 169 of 213 patients. Figure 1 shows the distribution of patients and cardiovascular complications by Nt-proBNP level. The 95 th percentile of the Nt-proBNP values at 20 weeks gestation in healthy women was 128 pg/ ml. Of the patients with cardiovascular complications, 82.4% had Nt-proBNP > 128 pg/ml Women with Nt-proBNP concentration < 100 pg/ml (77/169) had no cardiovascular complications. Patients with cardiovascular events had significantly higher Nt-proBNP at 20 weeks gestation compared with patients without events (173 ( ) vs. 100 ( ), p< 0.001). N-terminal pro-b-type natriuretic peptide levels >128 pg/ml at 20 weeks of gestation had a negative predictive value (NPV) of 96.9% for the occurrence of cardiovascular events after 20 weeks gestation; the positive predictive value (PPV) was 18.3%. The sensitivity of Nt-proBNP > 128 pg/ml was 81.3%; the specificity was 61.8%. There was no difference in Nt-proBNP levels between patients with heart failure compared to patients with arrhythmia and all women with elevated Nt-proBNP values had normal serum creatinine (48.7 ± 7.1 µmol/l). 8 patients had BMI > 30. Four of these patients had Nt-proBNP < 128 pg/ml, but none of the patients with BMI > 30 had cardiovascular events. The presence of a mechanical valve and subpulmonary ventricular dysfunction (tricuspid annular plane systolic excursion (TAPSE) < 16 mm) were identified as independent preconception predictors of primary cardiovascular events (table 3). In addition, an elevated Nt-proBNP

29 28 Chapter 2 Table 2. Distribution of cardiovascular events by primary type of congenital heart disease (N = 213 pregnancies). Maternal congenital lesion N % Pregnancies with primary cardiovascular events Cardiovascular events (N (%)) NICP HF AR TE Aortic dissection NYHA Left sided lesions 57 Aortic valve stenosis / Bicuspid aortic valve Surgically repaired Aortic coarctation Other* Right sided lesions 66 Tetralogy of Fallot after repair Pulmonary valve stenosis Ebstein s anomaly Shunt lesions 61 Ventricular septal defect Atrial septum secundum defect Atrioventricular septal defect Abnormal pulmonary venous return Connective tissue disorder Marfan syndrome Loeyz-Dietz syndrome 1 Complex CHD 20 Complete transposition of great arteries (TGA) D-TGA with Mustard or Senning D-TGA with arterial switch 2 0

30 Nt-proBNP in pregnant women with CHD 29 Table 2. Distribution of cardiovascular events by primary type of congenital heart disease (N = 213 pregnancies). (continued) Cardiovascular events (N (%)) NICP HF AR TE Aortic dissection NYHA Pregnancies with primary cardiovascular events Maternal congenital lesion N % Congenital corrected TGA 1 0 Single ventricle with Fontan physiology Other complex cyanotic heart disease Total (10.3) 2 (0.9) 8 (3.8) 14 (6.6) 4 (1.9) 1 (0.5) 39 (18.3) NICP need of invasive cardiovascular procedure; HF heart failure; AR arrhythmia; TE thrombo-embolic event; NYHA deterioration of 2 New York Heart Association (NYHA) functional classes during pregnancy or until 6 months post-partum. * 1 patient with a right sided aortic arch and 1 patient with a cleft mitral valve. 1 patient with a corrected truncus arteriosus, type A; 1 patient with pulmonary atresia, atrial septal defect and intact intraventricular septum. Secondary cardiovascular event. 2

31 30 Chapter Number of patients NT-proBNP (pg/ml) >400 A Patients with cardiovascular event (%) NT-proBNP (pg/ml) Figure 1: N-terminal pro-b-type natriuretic peptide (pg/ml) at 20 weeks of gestation in women with congenital heart disease. A: distribution of patients by N-terminal pro-b-type natriuretic peptide level; B: Percentage of cardiovascular events by N-terminal pro-b-type natriuretic peptide level. B Table 3. Maternal preconceptional predictors of cardiovascular events during pregnancy. OR 95% CI P-value Number of times selected # Univariate predictor preconception WHO class III or IV < Mechanical valve prosthesis < Subpulmonary ventricular dysfunction (TAPSE < 16 mm) Prior cardiovascular event 4, History of pacemaker implantation Right systemic ventricle Multivariable predictor preconception* Mechanical valve prosthesis Subpulmonary ventricular dysfunction (TAPSE < 16 mm) * Adjusted for preconception WHO risk class. # Explanation can be found in the statistical method section. level at 20 weeks gestation (> 128 pg/ml) was an independent predictor of cardiovascular events, OR 10.6 (95% CI , p=0.039) (adjusted for the preconception variables in the model, after exclusion of patients with complications before 20 weeks gestation only and considering only complications other than valve thrombosis). The area under the ROC increased from 0.78 to 0.90 (p=0.035) when Nt-proBNP at 20 weeks gestation was added to the multivariable model with the identified independent preconception predictors. The

32 Nt-proBNP in pregnant women with CHD 31 absence of all three risk factors had a high NPV for the occurrence of cardiovascular events (94.3%) in the entire study population. The positive predictive value of the presence of all three predictors was 50.0% for the entire study population. In a selected population with none or one of the preconception predictors present, the NPV of Nt-proBNP levels < 128 pg/ ml is 100%; the PPV of Nt-proBNP > 128 pg/ml is poor with 10.0% and 33.3% respectively. We could not calculate PPV and NPV in selected patients with both preconception risk factors, due to the absence of low Nt-proBNP in these patients. 2 Discussion The main new finding of the present study is the identification of elevated Nt-proBNP levels at 20 weeks gestation as an independent predictor of cardiovascular events during pregnancy, in addition to the presence of a mechanical valve and subpulmonary ventricular dysfunction before conception. The absence of these predictors had a high NPV for the occurrence of events. Our study is the first to report on the role of Nt-proBNP to predict events during pregnancy in women with CHD. All women survived and cardiovascular events occurred in 10.3% of all pregnancies. This is in line with most previous studies 1-5,20,21. Differences in the prevalence of cardiovascular events probably reflect differences in study design and in the composition of the study populations. In the present study, arrhythmias and heart failure were the most frequently observed adverse events, as consistently described before 2,3,5, The presence of a mechanical valve and subpulmonary ventricular dysfunction before pregnancy were identified as independent predictors of cardiovascular events. Mechanical valves are known for their association with pregnancy complications, mainly valve thrombosis resulting in embolism, heart failure, re-operation and maternal death However, mechanical valves were only reported once before, in a previous study of our group, as an independent predictor and are not incorporated in the CARPREG prediction model 2,4. The risk of valve thrombosis depends strongly on anticoagulation regimen, monitoring and patient adherence, which must be taken into account. Therefore we did not include mechanical valve thrombosis in the analysis of the predictive value of Nt-proBNP for cardiovascular events. On the other hand, the risk of valve thrombosis may be higher in patients with ventricular dysfunction due to lower flow velocities, which may link elevated Nt-proBNP levels to valve thrombosis. Indeed, our only patient with valve thrombosis of a mechanical valve after 20 weeks gestation had compromised right ventricular function and elevated Nt-proBNP. The present study identified subpulmonary ventricular dysfunction as an independent predictor of events. Two previous studies did identify subpulmonary ventricular dysfunction as a predictor, but only in combination with other predictors 3,26. The cohort described here included a considerable amount of patients with tetralogy of Fallot and right sided obstruc-

33 32 Chapter 2 tive lesions, which could have contributed to identifying subpulmonary ventricular dysfunction as an independent risk factor. Tricuspid annular plane systolic excursion as a marker of RV function has been scarcely validated in patients with CHD, but the available evidence indicates that TAPSE is associated with RV function and symptoms in patients with CHD and pressure- or volume loaded right ventricles Even though it did not emerge as an independent predictor, the modified WHO class was highly associated with the occurrence of cardiovascular events in our study 7. This is in line with results of the European Registry on pregnancy and cardiac disease 21. The results reported here underline that the WHO class can be a useful tool in preconception risk assessment and counseling 7. The presence of a systemic right ventricle was also associated with cardiovascular complications in the univariable analysis, in line with previous literature in which the complication rate in these patients was high 2,30. An elevated Nt-proBNP level at 20 weeks gestation was an independent predictor of cardiovascular events during pregnancy in this study and had additional value in identifying patients at risk of cardiovascular events after 20 weeks gestation. The NPV of Nt-proBNP levels < 128 pg/ml at 20 weeks for the development of cardiovascular events is high. This is in line with the results of Tanous et al. for BNP, but they measured BNP only during the event in 3/8 cases and therefore could not calculate the value of BNP for prediction of events 13. N-terminal pro-b-type natriuretic peptide appears helpful in identifying women with low risk of cardiovascular events during pregnancy, since a low Nt-proBNP has a very high NPV even when one other pre-pregnancy predictor of events is present. As a consequence, less frequent follow up during the second and third trimester may be acceptable for women with low Nt-proBNP levels at 20 weeks. It should be kept in mind though, that positive and negative predictive values are strongly dependent on the prevalence of complications in a cohort of patients for a given sensitivity and specificity of a test (table 4). Table 4. Estimates of negative and positive predictive values depending on disease prevalence. Reference Prevalence of cardiovascular complications NPV* PPV* Kampman et al. (present study) 10.3% 96.9% 18.3% Drenthen et al (2) 7.6% 97.8% 14.9% Khairy et al (3) 19.4% 93.2% 33.9% *Sensitivity and specificity values presented in the results section were used for the calculations. Patients with cardiovascular events had significantly higher Nt-proBNP levels than women without cardiovascular events. This may suggest that women with CHD who develop cardiovascular complications do not have the required cardiac reserve to adapt to the hemodynamic changes of pregnancy. Compared to the values found in clinical overt heart failure, our cutoff value of Nt-proBNP > 128 pg/ml for predicting cardiovascular events is relatively low. Two recent reviews concern-

34 Nt-proBNP in pregnant women with CHD 33 ing natriuretic peptides in patients with CHD describe relatively low Nt-proBNP values in asymptomatic patients, although values are higher than in healthy controls 31,32. Most of the patients described here were asymptomatic before pregnancy and had good cardiac function, explaining the relatively low Nt-proBNP values in our study. Although Nt-proBNP levels are relatively low, they have meaning in predicting cardiovascular events. The results of the present study are in line with previous studies in non-pregnant patients with CHD, which reported that relatively low levels of natriuretic peptides are associated with adverse outcome, ventricular dysfunction and exercise capacity Giannakoulas et al. found elevated BNP levels (> 78 pg/ml) to be strongly associated with mortality and Nt-proBNP is known to be a marker for early detection of right ventricular dysfunction in patients with right heart disease 11,33,35. Although the prognostic role of natriuretic peptides in patients with heart failure is studied extensively, its value in patients with CHD is still understudied. Large prospective studies are missing and additional research on this topic is clearly warranted. In our study, a considerable amount of patients without cardiovascular events did have elevated Nt-proBNP levels at 20 weeks gestation, which was reflected by low PPV. Comparable results were found by Tanous et al. for BNP levels 13. Exercise increases BNP values in patients with tetralogy of Fallot, which is associated with impaired contractile reserve of the right ventricle 34. Since pregnancy can be considered a stress test due to its hemodynamic changes, the elevated NT-pro-BNP levels in our population with CHD may indicate subclinical compromised cardiac function even in patients who have an uneventful pregnancy. To observe a possible relation with deterioration after pregnancy, long term follow up is necessary. 2 Strengths and limitations This relatively large prospective study is the first study evaluating the role of Nt-proBNP to predict cardiovascular events during pregnancy in women with CHD. Because all women with structural CHD were eligible for inclusion, our cohort was heterogeneous. Individual diseases may be underrepresented. Because of the design of our protocol (inclusion < 20 weeks gestation) pre-pregnancy data were collected retrospectively and missing data were inevitable. N-terminal pro-b-type natriuretic peptide values were scarcely available prior to pregnancy; therefore comparison with pre-pregnancy data was not possible. The exclusion of systemic right ventricles from quantitative analysis of chamber function measurements must be kept in mind when interpreting the reported results. Since NPV and PPV are strongly dependent on prevalence of cardiovascular complications, caution is warranted when extrapolating these results to other populations.

35 34 Chapter 2 Because of the limited amount of cardiovascular events, we could not correct for all the known independent predictors. Therefore, the presented results should be interpreted with caution and additional studies are required. The limited number of events prevented a separate analysis with heart failure as the only endpoint. Conclusions Women with CHD are prone to cardiovascular events during pregnancy. Increased Nt-proBNP levels at 20 weeks gestation are an independent predictor of adverse cardiovascular events during pregnancy in women with CHD and have additional value in predicting the risk of adverse cardiovascular events during pregnancy on top of preconception predictors. Normal levels of Nt-proBNP have a high NPV for adverse maternal cardiovascular events and determination of Nt-proBNP levels during pregnancy can therefore be helpful in clinical evaluation and follow-up of a pregnant woman with CHD. Further study is recommended to establish the role of Nt-proBNP values in subgroups with specific cardiac lesions. Acknowledgements The authors thank Dr. J.P.M. Hamer for his contribution in evaluating the echocardiograms. Funding This work is supported by a grant from The Netherlands Heart Foundation (2007B75). D.J.V. is a clinically established investigator of The Netherlands Heart Foundation (D97.015). Conflicts of interest D.J.V. has received Board Membership Fees from Amgen, BG Medicine, Biocontrol, Johnson & Johnson, Novartis, Sorbent and Vifor.

36 Nt-proBNP in pregnant women with CHD 35 References 1. Drenthen W, Pieper PG, Roos-Hesselink JW, van Lottum WA, Voors AA, Mulder BJ, van Dijk AP, Vliegen HW, Yap SC, Moons P, Ebels T, van Veldhuisen DJ, ZAHARA Investigators. Outcome of pregnancy in women with congenital heart disease: a literature review. J Am Coll Cardiol 2007; 49: Drenthen W, Boersma E, Balci A, Moons P, Roos-Hesselink JW, Mulder BJ, Vliegen HW, van Dijk AP, Voors AA, Yap SC, van Veldhuisen DJ, Pieper PG, ZAHARA Investigators. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010; 31: Khairy P, Ouyang DW, Fernandes SM, Lee-Parritz A, Economy KE, Landzberg MJ. Pregnancy outcomes in women with congenital heart disease. Circulation 2006; 113: Siu SC, Sermer M, Colman JM, Alvarez AN, Mercier LA, Morton BC, Kells CM, Bergin ML, Kiess MC, Marcotte F, Taylor DA, Gordon EP, Spears JC, Tam JW, Amankwah KS, Smallhorn JF, Farine D, Sorensen S, Cardiac Disease in Pregnancy (CARPREG) Investigators. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001; 104: Jastrow N, Meyer P, Khairy P, Mercier LA, Dore A, Marcotte F, Leduc L. Prediction of complications in pregnant women with cardiac diseases referred to a tertiary center. Int J Cardiol 2011; 151: Thorne S, MacGregor A, Nelson-Piercy C. Risks of contraception and pregnancy in heart disease. Heart 2006; 92: Regitz-Zagrosek V, Blomstrom Lundqvist C, Borghi C, Cifkova R, Ferreira R, Foidart JM, Gibbs JS, Gohlke-Baerwolf C, Gorenek B, Iung B, Kirby M, Maas AH, Morais J, Nihoyannopoulos P, Pieper PG, Presbitero P, Roos-Hesselink JW, Schaufelberger M, Seeland U, Torracca L, ESC Committee for Practice Guidelines, Bax J, Auricchio A, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Knuuti J, Kolh P, McDonagh T, Moulin C, Poldermans D, Popescu BA, Reiner Z, Sechtem U, Sirnes PA, Torbicki A, Vahanian A, Windecker S, ESC Guidelines on the management of cardiovascular diseases during pregnancy: the Task Force on the Management of Cardiovascular Diseases during Pregnancy of the European Society of Cardiology (ESC). Eur Heart J 2011; 32: Kragelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-terminal pro-b-type natriuretic peptide and long-term mortality in stable coronary heart disease. N Engl J Med 2005; 352: Linssen GC, Bakker SJ, Voors AA, Gansevoort RT, Hillege HL, de Jong PE, van Veldhuisen DJ, Gans RO, de Zeeuw D. N-terminal pro-b-type natriuretic peptide is an independent predictor of cardiovascular morbidity and mortality in the general population. Eur Heart J 2010; 31: Omland T, Persson A, Ng L, O Brien R, Karlsson T, Herlitz J, Hartford M, Caidahl K. N-terminal pro-b-type natriuretic peptide and long-term mortality in acute coronary syndromes. Circulation 2002; 106: Kirk V, Bay M, Parner J, Krogsgaard K, Herzog TM, Boesgaard S, Hassager C, Nielsen OW, Aldershvile J, Nielsen H. N-terminal probnp and mortality in hospitalised patients with heart failure and preserved vs. reduced systolic function: data from the prospective Copenhagen Hospital Heart Failure Study (CHHF). Eur J Heart Fail 2004; 6: van Veldhuisen DJ, Linssen GC, Jaarsma T, van Gilst WH, Hoes AW, Tijssen JG, Paulus WJ, Voors AA, Hillege HL. B-type natriuretic Peptide and prognosis in heart failure patients with preserved and reduced ejection fraction. J Am Coll Cardiol 2013; 61:

37 36 Chapter Tanous D, Siu SC, Mason J, Greutmann M, Wald RM, Parker JD, Sermer M, Colman JM, Silversides CK. B-type natriuretic peptide in pregnant women with heart disease. J Am Coll Cardiol 2010; 56: Balci A, Sollie KM, Mulder BJ, de Laat MW, Roos-Hesselink JW, van Dijk AP, Wajon EM, Vliegen HW, Drenthen W, Hillege HL, Aarnoudse JG, van Veldhuisen DJ, Pieper PG. Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study. Am Heart J 2011; 161: e Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS, Solomon SD, Spencer KT, Sutton MS, Stewart WJ, Chamber Quantification Writing Group, American Society of Echocardiography s Guidelines and Standards Committee, European Association of Echocardiography. Recommendations for chamber quantification: a report from the American Society of Echocardiography s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005; 18: Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A, Griffin BP, Iung B, Otto CM, Pellikka PA, Quinones M, American Society of Echocardiography, European Association of Echocardiography. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. J Am Soc Echocardiogr 2009; 22: 1-23; quiz Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA, Waggoner AD, Flachskampf FA, Pellikka PA, Evangelista A. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. J Am Soc Echocardiogr 2009; 22: Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, Chandrasekaran K, Solomon SD, Louie EK, Schiller NB. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010; 23: ; quiz Dickstein K, Cohen-Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, Stromberg A, van Veldhuisen DJ, Atar D, Hoes AW, Keren A, Mebazaa A, Nieminen M, Priori SG, Swedberg K, ESC Committee for Practice Guidelines (CPG). ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 2008; 29: Stangl V, Schad J, Gossing G, Borges A, Baumann G, Stangl K. Maternal heart disease and pregnancy outcome: a single-centre experience. Eur J Heart Fail 2008; 10: Roos-Hesselink JW, Ruys TP, Stein JI, Thilen U, Webb GD, Niwa K, Kaemmerer H, Baumgartner H, Budts W, Maggioni AP, Tavazzi L, Taha N, Johnson MR, Hall R, on behalf of the ROPAC Investigators. Outcome of pregnancy in patients with structural or ischaemic heart disease: results of a registry of the European Society of Cardiology. Eur Heart J 2012; 22. Siu SC, Colman JM, Sorensen S, Smallhorn JF, Farine D, Amankwah KS, Spears JC, Sermer M. Adverse neonatal and cardiac outcomes are more common in pregnant women with cardiac disease. Circulation 2002; 105: Pieper PG, Balci A, Van Dijk AP. Pregnancy in women with prosthetic heart valves. Neth Heart J 2008; 16:

38 Nt-proBNP in pregnant women with CHD Sadler L, McCowan L, White H, Stewart A, Bracken M, North R. Pregnancy outcomes and cardiac complications in women with mechanical, bioprosthetic and homograft valves. BJOG 2000; 107: Elkayam U, Bitar F. Valvular heart disease and pregnancy: part II: prosthetic valves. J Am Coll Cardiol 2005; 46: Greutmann M, Von Klemperer K, Brooks R, Peebles D, O Brien P, Walker F. Pregnancy outcome in women with congenital heart disease and residual haemodynamic lesions of the right ventricular outflow tract. Eur Heart J 2010; 31: Koestenberger M, Nagel B, Avian A, Ravekes W, Sorantin E, Cvirn G, Beran E, Halb V, Gamillscheg A. Systolic right ventricular function in children and young adults with pulmonary artery hypertension secondary to congenital heart disease and tetralogy of Fallot: tricuspid annular plane systolic excursion (TAPSE) and magnetic resonance imaging data. Congenit Heart Dis 2012; 7: Koestenberger M, Nagel B, Ravekes W, Everett AD, Stueger HP, Heinzl B, Sorantin E, Cvirn G, Fritsch P, Gamillscheg A. Systolic right ventricular function in pediatric and adolescent patients with tetralogy of Fallot: echocardiography versus magnetic resonance imaging. J Am Soc Echocardiogr 2011; 24: Fang F, Henein MY, Yu CM, Li W, Kaya MG, Coats AJ, Lam YY. Right ventricular long-axis response to different chronic loading conditions: Its relevance to clinical symptoms. Int J Cardiol 2012; 30. Drenthen W, Pieper PG, Ploeg M, Voors AA, Roos-Hesselink JW, Mulder BJ, Vliegen HW, Sollie KM, Ebels T, van Veldhuisen DJ, ZAHARA Investigators. Risk of complications during pregnancy after Senning or Mustard (atrial) repair of complete transposition of the great arteries. Eur Heart J 2005; 26: Eindhoven JA, van den Bosch AE, Boersma E, Roos-Hesselink JW. The usefulness of brain natriuretic peptide in simple congenital heart disease - a systematic review. Cardiol Young 2012; : Eindhoven JA, van den Bosch AE, Jansen PR, Boersma E, Roos-Hesselink JW. The usefulness of brain natriuretic peptide in complex congenital heart disease: a systematic review. J Am Coll Cardiol 2012; 60: Giannakoulas G, Dimopoulos K, Bolger AP, Tay EL, Inuzuka R, Bedard E, Davos C, Swan L, Gatzoulis MA. Usefulness of natriuretic Peptide levels to predict mortality in adults with congenital heart disease. Am J Cardiol 2010; 105: Ishii H, Harada K, Toyono M, Tamura M, Takada G. Usefulness of exercise-induced changes in plasma levels of brain natriuretic peptide in predicting right ventricular contractile reserve after repair of tetralogy of Fallot. Am J Cardiol 2005; 95: Lemmer J, Heise G, Rentzsch A, Boettler P, Kuehne T, Dubowy KO, Peters B, Lemmer B, Hager A, Stiller B, German Competence Network for Congenital Heart Defects. Right ventricular function in grown-up patients after correction of congenital right heart disease. Clin Res Cardiol 2011; 100: Eindhoven JA, van den Bosch AE, Ruys TP, Opic P, Cuypers JA, McGhie JS, Witsenburg M, Boersma E, Roos-Hesselink JW. N-terminal probrain Natriuretic Peptide and its Relation with Cardiac Function in Adult Patients with Congenital Heart Disease. J Am Coll Cardiol 2013; Sep 24; 62(13):

39

40 Chapter 3 Cardiac adaption during pregnancy: comparison between women with congenital heart disease and healthy women Marlies A.M. Kampman Mattia A. E. Valente Joost P. van Melle Ali Balci Jolien W. Roos-Hesselink Barbara J.M. Mulder Arie P.J. van Dijk Martijn A. Oudijk Monique R.M. Jongbloed Dirk J. van Veldhuisen Petronella G. Pieper on behalf of the ZAHARA II investigators 7 8 A Submitted.

41 40 Chapter 3 Abstract Background Pregnancy in women with congenital heart disease (CHD) is associated with cardiovascular complications and with deterioration in cardiac function. Still data on longitudinal changes during pregnancy are scarce. Data on right ventricular function parameters are not yet reported. We aimed to describe serial changes in cardiac dimension and cardiac function during pregnancy in women with CHD and compare these to changes seen in healthy women. We focus on both left and right ventricular parameters. Methods We performed a prospective multicenter cohort study. Follow-up with clinical evaluation and standardized echocardiography at 20 and 32 weeks gestation and one year postpartum was performed. In women with CHD, pre-pregnancy echocardiograms were also evaluated. Results We studied 125 women with CHD and 49 healthy women. The absolute level of ventricular function parameters and diameters differs clearly between women with CHD and healthy women. No changes occurred in right and left ventricular function parameters and ventricular dimensions during pregnancy in the total population of women with CHD. However, women with right-sided CHD had a different profile of TAPSE over time compared to healthy women (p=0.043). Women with left-sided CHD had a different profile of LVEDD over time compared to healthy women (p=0.045). Conclusion Absolute levels of ventricular function parameters and diameters differ between women with CHD and healthy women. The different patterns over time seen for TAPSE and LVEDD in women with right-sided and left-sided CHD respectively, compared to healthy women indicate the importance of echocardiographic follow-up during pregnancy in women with CHD. Key words: congenital heart disease, pregnancy, cardiac function, echocardiography.

42 Cardiac adaption during pregnancy 41 Introduction Pregnancy in women with congenital heart disease (CHD) is associated with increased incidences of cardiovascular, obstetric and neonatal complications 1-5. Cardiac complications, such as arrhythmia and heart failure, are thought to be due to the hemodynamic changes of pregnancy. Depending on the specific underlying congenital defect, pregnancy can be associated with persisting structural cardiac remodeling and deterioration in function, such as dilatation of the subpulmonary ventricle after pregnancy, deterioration of valvular dysfunction and worsening ventricular function However, the observations in most of these studies were not based on longitudinal data. The majority of data on cardiovascular changes over time during pregnancy is based on studies in healthy women and only a few reports describe longitudinal cardiovascular changes in women with heart disease 11,12. Most of the available research focuses on left ventricular parameters 13. Data examining longitudinal changes in right ventricular function and dimension are scarce in healthy pregnant women and have never been reported in pregnant women with congenital heart disease 14,15. Therefore, we aimed to describe serial measurements in cardiac dimensions and systolic and diastolic function of the right and left ventricle during pregnancy in women with congenital heart disease. In addition, we compared the serial measurements in women with congenital heart disease to the changes seen in healthy pregnant women. 3 Methods Study population The ZAHARA II study (Zwangerschap bij Aangeboren HARtAfwijkingen; Pregnancy in CHD) is a national, prospective multicenter cohort study. All consecutive pregnant women with structural CHD, aged 18 years, pregnancy duration < 20 weeks and presenting in one of the eight participating centers were eligible for enrollment. Healthy pregnant women (nonsmokers, no medication use, aged 18 years) were recruited from low risk midwife practices in Groningen and Rotterdam, the Netherlands. The study design and primary results have been published previously 16,17. Only women from the ZAHARA II study with singleton pregnancies and complete echocardiographic follow-up (preconception (only applicable for women with CHD), 20 and 32 weeks gestation, one year post-partum) were included for the current study and only the first pregnancy during the study period was taken into account. Women with a systemic right ventricle or Fontan physiology were excluded, since various echocardiographic measurements are not validated for these types of congenital heart disease.

43 42 Chapter 3 The Research Ethics Committee of all participating centers approved the study protocol and all participating women gave written informed consent. The ZAHARA II study was supported by a grant from the Netherlands Heart foundation (2007B75). Baseline characteristics and echocardiography Baseline characteristics were collected using medical records during the first ante-partum visit and included maternal age, underlying congenital heart disease, previous interventions, prior cardiac events, cardiac medication use, New York Heart Association (NYHA) functional class, echocardiography data, co-morbid conditions and obstetric history. Patients and healthy pregnant women visited the outpatient clinic at 20 and 32 weeks gestation and one year post-partum for clinical evaluation (including NYHA class assessment) and standardized echocardiography. All echocardiograms were evaluated offline by four experienced cardiologists at the University Medical Center Groningen, Groningen, the Netherlands. Transthoracic echocardiographic evaluation was performed according to current guidelines and recommendations, and adapted to the structural defect when necessary Left ventricular end-diastolic and end-systolic diameters (LVEDD, LVESD), as well as the left ventricular outflow tract (LVOT) diameter, were assessed on the parasternal long axis view (PLAX). Left ventricular outflow tract velocity time integral was derived from the apical five chamber view. Cardiac output was calculated by multiplying LVOT area with the LVOT velocity time integral and heart rate. Left ventricular ejection fraction was determined using Simpson s biplane method using the apical four chamber view and the apical two chamber view where possible, otherwise Simpson s monoplane or the eyeballing method were used. Diastolic function was assessed by pulsed wave Doppler of the mitral inflow (E, A, E/A ratio) and color tissue Doppler of the septal and lateral mitral annulus (E/E ). Right ventricular function was measured using the Tricuspid Annular Plane Systolic Excursion (TAPSE) in the apical four chamber view and the peak systolic color tissue velocity Doppler of the right ventricular lateral wall assessed at the tricuspid annulus (S RV). Maximal right ventricular end-diastolic diameter was measured using the modified apical four chamber view. Left ventricular systolic dysfunction was defined as an ejection fraction < 45%. Right ventricular systolic dysfunction was defined as TAPSE < 16 mm. Statistical analysis Continuous variables are presented as mean with standard deviation (SD) or medians with interquartile ranges as appropriate. Absolute numbers and percentages are displayed for categorical data. To investigate changes in the serial echocardiographic parameters over time in pregnant women with CHD and to compare the serial changes over time with healthy pregnant women, random slope, random intercept linear mixed-effects models were used, adjusted for age, race and parity. These hierarchical regression models include fixed and random

44 Cardiac adaption during pregnancy 43 (subject-specific) effects, allowing for within subject correlation between repeated measurements. As the evolution of parameters during pregnancy might not be linear, patterns were investigated graphically, and polynomial transformations were considered where appropriate. Best fit transformations were selected via combined assessment of Akaike s Information Criterium (AIC, lower is better) for fixed effects and likelihood ratio tests of nested models for random effects. Second degree polynomial transformations for changes over time were selected for the final models. Interaction terms for group membership (pregnant women with CHD vs. healthy pregnant women; pregnant women with right-sided CHD vs. healthy pregnant women; pregnant women with left-sided CHD vs. healthy pregnant women) were introduced as a fixed effect to check for differences in course over time. Women with Tetralogy of Fallot, atrial septal defects, partial atrioventricular septal defects, pulmonary atresia with intact ventricular septum, Ebstein s anomaly, abnormal pulmonary venous return, sinus venosus defects and pulmonary valve stenosis were classified as having right-sided CHD. Women with ventricular septal defects, aortic valve abnormalities, aortic coarctation, Marfan syndrome and with a cleft mitral valve were considered to have left-sided CHD. All statistical analyses were performed using STATA software package (version 11, college station, Texas, USA) and R: A language and environment for statistical computing (version 3.1.0, R Foundation for statistical computing, Vienna, Austria). A two-tailed p-value < 0.05 was considered significant. 3 Results During the study period, 213 women with congenital heart disease and 70 healthy women were included. Eighty-eight pregnancies in women with CHD were excluded because of a twin pregnancy (n=4), Fontan physiology or a systemic right ventricle (n=15), second pregnancy in the study period (n=9) or incomplete echocardiographic data (n=60), rendering 125 patients available for analysis. In one healthy woman a previously unknown atrial septal defect type II was found and 20 women did not have complete echocardiographic follow-up, resulting in 49 healthy pregnant women included in this analysis. Baseline characteristics and underlying congenital defects are displayed in table 1. Mean age was comparable between women with CHD (29.4 ± 4.5 years) and healthy women (30.1 ± 4.1 years; p=0.31). The majority of patients and healthy controls was nulliparous (62.4% vs. 61.2%, p=0.89) and most of them were in NYHA functional class I (75.2% vs. 98.0%, p< 0.001). Left ventricular systolic dysfunction was seen in 4 (3.2%) women with CHD and 15 (12.0%) had right ventricular systolic dysfunction. Serial means over time of the ventricular function parameters and the ventricular dimensions of right and left ventricle are reported in table 2 and displayed in figure 1 and 2. No statisti-

45 44 Chapter 3 Table 1. Maternal baseline characteristics (prior to pregnancy). Patients (N=125) N (%) Healthy women (N=49) N (%) Demographics and clinical data Maternal age at conception (years ± SD) 29.4 ± ± 4.1 Parity status 0 78 (62.4) 30 (61.2) 1 47 (37.6) 19 (38.8) NYHA class I 94 (75.2) 48 (98.0) II 30 (24.0) 1 (2.0) III 1 (0.8) Modified WHO class* I 15 (12.0) II 81 (63.7) III 27 (21.6) IV 2 (1.6) Mechanical valve prosthesis 10 (8.0) Arrhythmia 10 (8.0) Pacemaker 2 (1.6) Congestive heart failure 4 (3.2) Hypertension 10 (8.0) Type of congenital lesion Abnormal pulmonary venous return 1 (0.8) Aortic valve stenosis / Bicuspid aortic valve 23 (18.4) Atrial septum defect 14 (11.2) Atrioventricular septal defect 7 (5.6) Ebstein s anomaly 1 (0.8) Loeyz-Dietz syndrome 1 (0.8) Marfan syndrome 7 (5.6) Other* 1 (0.8) Other complex cyanotic heart disease 2 (1.6) Pulmonary valve stenosis 10 (8.0) Surgically repaired Aortic coarctation 19 (15.2) Tetralogy of Fallot after repair 28 (22.4) Transposition of great arteries with arterial switch 2 (1.6) Ventricular septal defect 9 (7.2) Echocardiographic parameters Left ventricular systolic dysfunction (LVEF < 45%) 4 (3.2) Right ventricular systolic dysfunction (TAPSE < 16 mm) 15 (12.0%) *Patient with cleft mitral valve. 1 patient with a corrected truncus arteriosus, type A; 1 patient with pulmonary atresia, atrial septal defect and intact intraventricular septum.

46 Cardiac adaption during pregnancy 45 Table 2. Longitudinal echocardiographic parameters in pregnant women with congenital heart disease and healthy women (mean ± SD). Prepregnancy 20 weeks 32 weeks 1-yr postpartum Right ventricular parameters Entire CHD population Tricuspid annular plane systolic excursion (TAPSE (mm)) 21.2 ± ± ± ± 5.5 Systolic tissue velocity of the lateral wall (S RV (cm/s)) 8.4 ± ± ± ± 2.9 Right ventricular end-diastolic diameter (RVEDD (mm)) 38.2 ± ± ± ± 8.0 Women with right-sided CHD Tricuspid annular plane systolic excursion (TAPSE (mm)) 21.0 ± ± ± ± 5.5 Systolic tissue velocity of the lateral wall (S RV (cm/s)) 8.7 ± ± ± ± 2.7 Right ventricular end-diastolic diameter (RVEDD (mm)) 41.0 ± ± ± ± 8.3 Women with left-sided CHD Tricuspid annular plane systolic excursion (TAPSE (mm)) 21.9 ± ± ± ± 5.1 Systolic tissue velocity of the lateral wall (S RV (cm/s)) 7.8 ± ± ± ± 2.0 Right ventricular end-diastolic diameter (RVEDD (mm)) 35.2 ± ± ± ± 6.2 Healthy women Tricuspid annular plane systolic excursion (TAPSE (mm)) ± ± ± 3.1 Systolic tissue velocity of the lateral wall (S RV (cm/s)) ± ± ± 2.3 Right ventricular end-diastolic diameter (RVEDD (mm)) ± ± ± 5.0 Left ventricular parameters Entire CHD population Left ventricular ejection fraction (LVEF (%)) 57.6 ± ± ± ± 9.3 Mean systolic tissue velocity of the septal and lateral wall (S LV (cm/s)) 6.2 ± ± ± ± 1.7 Left ventricular end-systolic diameter (LVESD (mm)) 29.8 ± ± ± ± 5.5 Left ventricular end-diastolic diameter (LVEDD (mm)) 46.3 ± ± ± ± 5.7 Women with right-sided CHD Left ventricular ejection fraction (LVEF (%)) 56.1 ± ± ± ± 9.4 Mean systolic tissue velocity of the septal and lateral 6.3 ± ± ± ± 1.7 wall (S LV (cm/s)) Left ventricular end-systolic diameter (LVESD (mm)) 29.1 ± ± ± ± 5.0 3

47 46 Chapter 3 Table 2. Longitudinal echocardiographic parameters in pregnant women with congenital heart disease and healthy women (mean ± SD). (continued) Prepregnancy 20 weeks 32 weeks 1-yr postpartum Left ventricular end-diastolic diameter (LVEDD (mm)) 44.9 ± ± ± ± 5.4 Women with left-sided CHD Left ventricular ejection fraction (LVEF (%)) 58.8 ± ± ± ± 9.2 Mean systolic tissue velocity of the septal and lateral wall (S LV (cm/s)) 6.1 ± ± ± ± 1.7 Left ventricular end-systolic diameter (LVESD (mm)) 30.7 ± ± ± ± 5.7 Left ventricular end-diastolic diameter (LVEDD (mm)) 47.9 ± ± ± ± 5.6 Healthy women Left ventricular ejection fraction (LVEF (%)) ± ± ± 5.6 Mean systolic tissue velocity of the septal and lateral wall (S LV (cm/s)) ± ± ± 1.1 Left ventricular end-systolic diameter (LVESD (mm)) ± ± ± 4.0 Left ventricular end-diastolic diameter (LVEDD (mm)) ± ± ± TAPSE (mm) S' RV (cm/s) A 7 B Preconception 20 wk 32 wk 1-yr post partum Preconception 20 wk 32 wk 1-yr post partum RVEDD (mm) C Preconception 20 wk 32 wk 1-yr post partum Figure 1: Serial changes (means with 95% confidence interval) in right ventricular function parameters for the entire population of women with CHD (A, Tricuspid Annular Plane Systolic Excursion (TAPSE (mm)(n=121)); B, Systolic tissue velocity of the right ventricle lateral wall (S RV (cm/s)(n=96)) and right ventricular end diastolic diameter (RVEDD (mm)(n=113)) (C).

48 Cardiac adaption during pregnancy LVEF (%) S' LV (cm/s) A 5.0 B Preconception wk 32 wk 1-yr post partum Preconception wk 32 wk 1-yr post partum LVEDD (mm) LVESD (mm) C 26 D Preconception 20 wk 32 wk 1-yr post partum Preconception 20 wk 32 wk 1-yr post partum Figure 2: Serial changes (means with 95% confidence interval) in left ventricular function parameters for the entire population of women with CHD (A, Left ventricular ejection fraction (LVEF (%)(n=124)); B, Mean systolic tissue velocity of the septal and left ventricular lateral wall (S LV (cm/s)(n=91)) and left ventricular end diastolic (C (n=125)) and end systolic (D (n=125)) diameter (mm) for women with CHD. cally significant changes occurred during pregnancy in either of the parameters assessed in the entire population of women with CHD. Figure 3 compares the fitted longitudinal profiles over time of the right ventricular function parameters and right ventricular dimension. Women with CHD have lower values of right ventricular function parameters (TAPSE and systolic tissue velocity of the right ventricle) compared to healthy women. Right ventricular end-diastolic diameter was larger throughout pregnancy compared to healthy women. The changes observed in the entire cohort of women with CHD were comparable to the changes seen in healthy women; no statistically significant differences were found in the slope of any of the longitudinal profiles, indicating similar patterns of change over time for both populations. In women with solely right-sided CHD, the fitted longitudinal profile of TAPSE over time was significantly different from that in healthy women (p=0.043), with TAPSE remaining unchanged in women with right-sided CHD (table 2). The fitted longitudinal profiles over time of the left ventricular function parameters and left ventricular dimensions are shown in figure 4. Women with CHD have a worse left ventricular systolic function during pregnancy compared to healthy pregnant women, as displayed by lower LVEF and systolic tissue velocity. Left ventricular dimensions are comparable between

49 48 Chapter 3 TAPSE (mm) s' RV (cm/s) wk 32 wk 1-yr post partum A 7 20 wk 32 wk 1-yr post partum B RVEDD (mm) wk 32 wk 1-yr post partum C Figure 3: Fitted longitudinal profiles of right ventricular function parameters for women with CHD ( ), and healthy women ( ). (A, Tricuspid Annular Plane Systolic Excursion (TAPSE (mm) (n=116 vs. 47)); B, Systolic tissue velocity of the right ventricle lateral wall (S RV (cm/s) (n=91 vs. 39)) and right ventricular Figuren mixed Figuren models (2).pzf:LV mixed models parameters (2).pzf:LV - Tue parameters Jan 05 13:33:53 - Tue Jan :33: end diastolic diameter (RVEDD (mm) (n=105 Figuren mixed vs. models 41)) (2).pzf:LV (C) parameters Error - Tue bars Jan 05 13:33:53 represent % confidence interval LVEF (%) (%) LVEF (%) Figuren mixed models (2).pzf:LV parameters - Tue Jan 05 13:33: S' S' LV LV (cm/s) S' LV (cm/s) 7 6 LVEDD (mm) S' LV (cm/s) LVEDD (mm) LVESD (mm) wk 20 wk wk 20 wk 20 wk 32 wk 32 wk 20 wk 32 wk 32 wk 1-yr post 32 wk partum 1-yr post partum 32 wk 1-yr post partum 1-yr post partum A A 1-yr post partum C C 1-yr post partum A A C LVESD (mm) LVESD (mm) 5 20 wk 20 wk wk 20 wk 20 wk 32 wk 32 wk 20 wk 32 wk 32 wk 32 wk 1-yr post partum 1-yr post partum 32 wk 1-yr post partum 1-yr post partum B B 1-yr post partum D D 1-yr post partum Figure 4: Fitted longitudinal profiles of left ventricular function parameters for women with CHD ( ) and healthy women ( ). (A, Left ventricular ejection fraction (LVEF (%) (n=116 vs. 49)); B, Mean systolic tissue velocity of the septal and left ventricular lateral wall (S LV (cm/s)(n=87 vs. 44)) and left ventricular end diastolic (C 31 (n=118 vs 49)) and end systolic (D (n=118 vs 49)) diameter (mm) Error bars represent 95% confidence interval wk 20 wk 32 wk 32 wk 1-yr post partum 1-yr post partum B D B D

50 Cardiac adaption during pregnancy 49 both groups. We found no statistically significant differences in the slope of the longitudinal profiles between women with CHD and healthy controls, indicating similar patterns of change over time for both populations. The fitted longitudinal profile of LVEDD in women with left-sided CHD over time was significantly different from that in healthy women (p=0.045) with LVEDD tending to increase over time in women with left-sided CHD (table 2). Discussion This is the first study that compares the serial changes in ventricular function parameters and ventricular dimensions during pregnancy in women with CHD to healthy pregnant women. In addition, this is the first study that describes the serial changes in right ventricular parameters seen during and after pregnancy in women with CHD. For the serial changes in the heterogeneous population of women with CHD, we did not find any statistical significant effect of time and the serial changes in echocardiographic parameters during pregnancy in women with CHD were comparable to healthy pregnant women, indicating similar patterns in both populations. However, the absolute levels of ventricular function and dimensions did clearly differ between women with CHD compared to healthy women. For women with right-sided CHD fitted profiles of TAPSE over time differed from the pattern seen in healthy women. Women with left-sided CHD had a different profile of LVEDD over time compared to healthy pregnant women. Comparison of our results to other studies is difficult since data on cardiac adaption during pregnancy in women with CHD or heart disease in general are very rare. Cornette et al. described a time effect during pregnancy towards a larger LVESD, lower fractional shortening and lower left ventricular ejection fraction 12. In addition, they found a parabolic effect for E/E, stroke volume and cardiac output. In our study, visual patterns and model fit criteria suggest non-linear variation in parameters over time, however there was no statistically significant effect of time for any of these parameters. The heterogeneity of the population is most probably due to this observation, since differences might cancel out when women with right-sided and left-sided defects are considered together in one population. The fact that we did find differences in fitted profiles of TAPSE over time in women with right-sided CHD and of LVEDD over time in women with left-sided CDH compared to healthy controls also strengthens this argument. Our results on Doppler peak systolic velocity of the left ventricle were comparable to those described by Bamfo et al. in healthy pregnant women 22. Their results support our finding that the pattern of longitudinal changes does not differ essentially from healthy women. Compared to healthy women, women with CHD in our cohort show lower fitted values for left ventricular ejection fraction (figure 4). Vasapollo et al. described significant lower LVEF in 3

51 50 Chapter 3 healthy women with fetal growth restriction compared to healthy women with uncomplicated pregnancy outcome 23. The pattern we observe in our study is comparable, with lower LVEF during pregnancy in women with CHD compared to healthy women. Although we did not find any deterioration in systolic function during pregnancy, this finding suggests that there might be an impaired potential in women with CHD to provide the required cardiac adaptions necessary to meet the increased metabolic demands of pregnancy. This may contribute to the higher obstetric and offspring complication rate in women with CHD, since cardiac dysfunction is related to impaired uteroplacental circulation and offspring complications 17. The observed difference in fitted values for LVEDD over time in women with left-sided CHD compared to healthy women may suggest that the volume load of pregnancy is not well tolerated in patients with these type of lesions; however, lesion-specific data are warranted. Data on right ventricular function and dimensions during pregnancy are extremely scarce and have never been described in a longitudinal manner for women with CHD. Vogt and colleagues were the first to report on systolic tissue velocities of the right ventricle (S ) during pregnancy in healthy women and did not find a significant change in systolic velocity comparing the first with the third trimester 14. Ducas et al. described, in a study using magnetic resonance imaging (MRI), no change in TAPSE and systolic myocardial velocity of the right ventricle lateral wall in healthy pregnant women during the third trimester compared to postpartum values (which were used as baseline measurement) 15. Our results on right ventricular systolic function in women with CHD are comparable to these findings, but the absolute values in our patients are considerably lower compared to the healthy women. TAPSE is related to impaired uteroplacental circulation and offspring complications 17. It might be that the absolute level of right ventricular systolic function is not sufficient to meet the increased demands of pregnancy, leading to adverse obstetric and offspring outcome. The observation that the evolution of TAPSE in women with right-sided CHD is significantly different from the pattern seen in healthy women, may also point into that direction. It might be that this subgroup has insufficient capacity to increase TAPSE, in order to accommodate cardiac output, which may contribute to the increased incidence rates of obstetric and offspring complications. It is known for specific congenital lesions, i.e. systemic right ventricles and Tetralogy of Fallot, that pregnancy can be associated with persistent deterioration in cardiac function 6,8 and women with cardiovascular complications during pregnancy are at risk for persistent dilatation of the right ventricle 10. Close follow up of high risk patients is still warranted, since small changes in cardiac function in the individual patient might be clinical relevant, although we did not find any statistical significant changes in our study cohort.

52 Cardiac adaption during pregnancy 51 Strengths & Limitations This is the first study that assessed the serial changes in ventricular function and dimensions during pregnancy in women with congenital heart disease and compared these to changes in healthy pregnant women. The comparison with healthy pregnant women is unique and makes the results more valuable. Due to the study design echocardiographic data before pregnancy from patients were collected retrospectively and in healthy women preconception echocardiography was not preformed. This hampered the comparison of the serial changes in women with CHD and healthy controls, since preconception data were not included in this analysis. Several highrisk congenital lesions (Fontan physiology and systemic right ventricles) were excluded from analysis. These are the most vulnerable patient groups during pregnancy for cardiac complications and deterioration in cardiac function during and after pregnancy. Excluding these types of patients may underestimate the effects of pregnancy on maternal cardiac function. In our study, the visual patterns and model fit criteria suggest non-linear variation in parameter values over time, however there was no statistically significant effect of time found for any of the parameters. This is most probably due to the heterogeneity of the study cohort. With subgroups analyses for right and left-sided CHD we attempted to compensate for that; however, insufficient power hampers analyses. Additional, lesion specific research is clearly warranted. 3 Conclusion In this study, we showed that absolute levels of ventricular function parameters and ventricular dimensions differ between women with CHD and healthy controls. The patterns of change over time seen during pregnancy are comparable between women with CHD and healthy pregnant women. However, fitted profiles of TAPSE over time in women with solely right-sided CHD differed significantly from healthy women. In women with left-sided CHD, fitted profiles of LVEDD over time were significantly different compared to healthy controls. These findings indicate that serial follow-up of cardiac function and dimensions during pregnancy in women with CHD is an important part of the management of pregnancy in women with congenital heart disease. Acknowledgements The authors thank dr. J.P.M. Hamer for his contribution in evaluating the echocardiograms.

53 52 Chapter 3 References 1. Drenthen W, Boersma E, Balci A, et al. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010; 31: Drenthen W, Pieper PG, Roos-Hesselink JW, et al. Outcome of pregnancy in women with congenital heart disease: A literature review. J Am Coll Cardiol 2007; 49: Siu SC, Sermer M, Colman JM, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001; 104: Siu SC, Colman JM, Sorensen S, et al. Adverse neonatal and cardiac outcomes are more common in pregnant women with cardiac disease. Circulation 2002; 105: Roos-Hesselink JW, Ruys TP, Stein JI, et al. Outcome of pregnancy in patients with structural or ischaemic heart disease: Results of a registry of the european society of cardiology. Eur Heart J 2013; 34: Uebing A, Arvanitis P, Li W, et al. Effect of pregnancy on clinical status and ventricular function in women with heart disease. Int J Cardiol 2010; 139: Kamiya CA, Iwamiya T, Neki R, et al. Outcome of pregnancy and effects on the right heart in women with repaired tetralogy of fallot. Circ J 2012; 76: Bowater SE, Selman TJ, Hudsmith LE, et al. Long-term outcome following pregnancy in women with a systemic right ventricle: Is the deterioration due to pregnancy or a consequence of time? Congenit Heart Dis 2013; 8: Guedes A, Mercier LA, Leduc L, et al. Impact of pregnancy on the systemic right ventricle after a mustard operation for transposition of the great arteries. J Am Coll Cardiol 2004; 44: Kampman MA, Balci A, Groen H, et al. Cardiac function and cardiac events 1-year postpartum in women with congenital heart disease. Am Heart J 2015; 169: Lesniak-Sobelga A, Tracz W, KostKiewicz M, et al. Clinical and echocardiographic assessment of pregnant women with valvular heart diseases--maternal and fetal outcome. Int J Cardiol 2004; 94: Cornette J, Ruys TP, Rossi A, et al. Hemodynamic adaptation to pregnancy in women with structural heart disease. Int J Cardiol 2013; 168: Demirkol S, Balta S, Cakar M, et al. Do we just assess the left ventricle in pregnant women with structural heart disease? Int J Cardiol 2013; 168: Vogt M, Muller J, Kuhn A, et al. Cardiac adaptation of the maternal heart during pregnancy: A color-coded tissue doppler imaging study - feasibility, reproducibility and course during pregnancy. Ultraschall Med 2014;. 15. Ducas RA, Elliott JE, Melnyk SF, et al. Cardiovascular magnetic resonance in pregnancy: Insights from the cardiac hemodynamic imaging and remodeling in pregnancy (CHIRP) study. J Cardiovasc Magn Reson 2014; 16: 1,429X Balci A, Sollie KM, Mulder BJ, et al. Associations between cardiovascular parameters and uteroplacental doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the zwangerschap bij aangeboren hartafwijking (ZAHARA) II study. Am Heart J 2011; 161: 269,275.e Pieper PG, Balci A, Aarnoudse JG, et al. Uteroplacental blood flow, cardiac function, and pregnancy outcome in women with congenital heart disease. Circulation 2013; 128: Baumgartner H, Hung J, Bermejo J, et al. Echocardiographic assessment of valve stenosis: EAE/ ASE recommendations for clinical practice. J Am Soc Echocardiogr 2009; 22: 1,23; quiz

54 Cardiac adaption during pregnancy Easterling TR, Carlson KL, Schmucker BC, et al. Measurement of cardiac output in pregnancy by doppler technique. Am J Perinatol 1990; 7: Robson SC, Hunter S, Boys RJ, et al. Serial study of factors influencing changes in cardiac output during human pregnancy. Am J Physiol 1989; 256: H Rudski LG, Lai WW, Afilalo J, et al. Guidelines for the echocardiographic assessment of the right heart in adults: A report from the american society of echocardiography endorsed by the european association of echocardiography, a registered branch of the european society of cardiology, and the canadian society of echocardiography. J Am Soc Echocardiogr 2010; 23: 685,713; quiz Bamfo JE, Kametas NA, Nicolaides KH, et al. Maternal left ventricular diastolic and systolic longaxis function during normal pregnancy. Eur J Echocardiogr 2007; 8: Vasapollo B, Valensise H, Novelli GP, et al. Abnormal maternal cardiac function and morphology in pregnancies complicated by intrauterine fetal growth restriction. Ultrasound Obstet Gynecol 2002; 20:

55

56 Chapter 4 Cardiac function and cardiac events 1-year post-partum in women with congenital heart disease Marlies A. M. Kampman Ali Balci Henk Groen Arie P. J. van Dijk Jolien W. Roos-Hesselink Joost P. van Melle Krystyna M. Sollie-Szarynska Elly M. C. J. Wajon Barbara J. M. Mulder Dirk J. van Veldhuisen Petronella G. Pieper on behalf of the ZAHARA II investigators 7 8 A American Heart Journal 2015;169:

57 56 Chapter 4 Abstract Background Pregnancy is increasingly common in women with congenital heart disease (CHD), but little is known about long-term cardiovascular outcome after pregnancy in these patients. We studied the incidence of cardiovascular events 1-year post-partum and compared cardiac function pre-pregnancy and 1-year post-partum in women with CHD. Methods From our national, prospective multicenter cohort study, 172 women were studied. Follow-up with clinical evaluation and echocardiography and Nt-proBNP measurement was performed during pregnancy and 12 months post-partum. Cardiovascular events were defined as need for an urgent invasive cardiovascular procedure, heart failure, arrhythmia, thromboembolic events, myocardial infarction, cardiac arrest, cardiac death, endocarditis and aortic dissection. Results Cardiovascular events were observed after 11 pregnancies (6.4%). Women with cardiovascular events post-partum had significant higher Nt-proBNP values at 20 weeks gestation (191 ( ) vs (57-167), p=0.049) and 1-year post-partum compared with women without cardiovascular events post-partum (306 ( ) vs. 105 (54-187) pg/ml, p=0.014). Women with cardiovascular events during pregnancy were at higher risk for late cardiovascular events (HR 7.1, 95% CI , p=0.003). In women with cardiovascular events during pregnancy, subpulmonary end-diastolic diameter had significantly increased 1-year post-partum (39.0 ( ) to 44.0( ),p=0.028). No other significant differences were found in cardiac function or size 1-year post-partum compared with preconception values. Conclusions Cardiovascular events are relatively rare 1-year after pregnancy in women with CHD. Women with cardiovascular events during pregnancy are prone to develop cardiovascular events 1-year post-partum and have increased subpulmonary ventricular diameter compared with preconception values. Key words: Congenital heart disease, pregnancy, cardiovascular events, post-partum.

58 Cardiac function & cardiac events one year post-partum. 57 Introduction Women with congenital heart disease (CHD) are prone to develop cardiovascular complications during pregnancy 1-4. Until now, research has mainly focused on risk prediction of cardiovascular complications during pregnancy, and several prediction models have been developed 1-4. However, limited data exist regarding the effects of pregnancy on maternal cardiac function and prognosis after pregnancy. Data from a small number of (mostly retrospective) studies have suggested that pregnancy may have an adverse effect on maternal cardiac outcome (with each subsequent pregnancy causing progression of cardiac dysfunction) and that women with adverse cardiac events during pregnancy are at increased risk for cardiac events > 6 months after pregnancy Prospective research data are very limited and concern only small study populations. Therefore, a large prospective multicenter study was conducted to observe the incidence of cardiovascular complications in the first year post-partum and to compare cardiac function parameters pre-pregnancy and 1-year post-partum in women with CHD. 4 Methods Patient selection The ZAHARA II study (Zwangerschap bij Aangeboren HARtAfwijkingen; pregnancy in CHD) is a prospective multicenter cohort study, conducted between March 2008 and August All consecutive pregnant women with structural CHD, aged 18 years, and presenting in one of the eight participating centers were eligible for enrollment. The methods and primary results have been described elsewhere 13,14. Women again pregnant before the post-partum follow-up visit were excluded from determination of cardiac outcome, and only the first pregnancy during the study period was included in this article. Baseline characteristics, pregnancy data and 1-year post-partum follow-up Pre-pregnancy baseline characteristics were collected from medical records during the first antepartum visit. Baseline data included maternal age, underlying congenital anomaly, prior interventions, previous cardiovascular events, medication use, New York Heart Association (NYHA) functional class, echocardiographic recordings, cardiopulmonary aerobic capacity test results (< 2 years prior to conception), 12-lead electrocardiogram, laboratory results, co-morbid conditions, and obstetric history. Patients visited the outpatient clinic at 20 and 32 weeks of gestation and 1-year post-partum for clinical evaluation (including NYHA class assessment), standardized echocardiograms (according to study protocol), 12-lead electrocardiogram, and Nt-proBNP measurement. In addition, when possible, cardiopulmonary aerobic capacity testing was performed 1-year post-partum.

59 58 Chapter 4 All echocardiographic recordings were evaluated off-line by four experienced cardiologists, blinded to the endpoints. Chamber quantification, valvular function, and systolic and diastolic ventricular function assessment were performed according to the current recommendations as previously described 13. Cardiac function and cardiovascular events 1- year post-partum Cardiovascular events (> 6 months after delivery) were assessed during the follow-up visit 1-year post-partum. Primary cardiovascular events as previously defined include any of the following: need for an urgent invasive cardiovascular procedure, heart failure (according to the guidelines of the European Society of Cardiology and documented by the attending physician 15 ), new onset or symptomatic tachyarrhythmia or brady arrhythmia requiring new or extended treatment, thromboembolic events, myocardial infarction, cardiac arrest, cardiac death, endocarditis, and aortic dissection 13. Echocardiographic changes were defined as deterioration in size or function of subpulmonary or sub-aortic ventricle, new onset or aggravation of valve regurgitation or stenosis 1 grade (mild to moderate or severe or moderate to severe). Statistical analysis For continuous data, means and standard deviations (SD) or medians with interquartile range (IQR) were calculated, depending on their distribution. Absolute numbers and percentages were presented for categorical data. The student s T-test, Mann-Whitney U test, Chi-square, or Fisher s exact test was used for intergroup comparison as appropriate. For comparison of cardiac function parameters before and after pregnancy, paired Student T-test or Wilcoxon signed rank test for related samples was used, depending on their distribution. For categorical data, McNemar test for related samples was used. Differences in the rates of late cardiac events were determined using log-rank tests. The current study concerns an analysis of the secondary outcomes of the ZAHARA II study, for which a formal power analysis was not performed. Statistical analysis was performed using STATA software package (version 11, college station, TX). A 2-tailed p-value < 0.05 was considered significant. The Research Ethics Committee of all participating centers approved the study protocol, and all participating women gave written informed consent. This work is supported by a grant from the Netherlands Heart Foundation (2007B75).The authors are solely responsible for the design and conduct of this study, all study analyses and drafting and editing of the paper. Results During the study period, 213 pregnancies (209 singleton and 4 twin pregnancies) in 202 women were observed. Twenty-three women had a second pregnancy during the study

60 Cardiac function & cardiac events one year post-partum. 59 period (12 before the post-partum follow-up visit) and 18 women did not return for followup. None of the women lost to follow-up died; these patients were all in modified World Health Organisation (WHO) risk class 1 or 2, except one (modified WHO class 3). The final study included 172 pregnancies in 172 women. Maternal baseline characteristics and underlying CHD are shown in table 1. Mean age at conception was 28.9 years. Most patients were in NYHA functional class I or II and had a mildly or moderately increased risk of cardiovascular complications during pregnancy as indicated by the modified WHO risk class (16). Most did not use cardiac medication before pregnancy, and the two patients who were on angiotensin-converting enzyme inhibitors discontinued the use because of its teratogenic effects. Systemic ventricular dysfunction (ejection fraction < 45%) was present in 6.5% of the women, and subpulmonary ventricular dysfunction (tricuspid annular plane systolic excursion (TAPSE) < 16 mm) was seen in 13.8%. Table 1. Maternal baseline characteristics (before pregnancy). N=172 women. N (%) Demographics and clinical data Maternal age at conception (years ± SD) 28.9 (± 4.3) Parity status (66.9) 1 43 (25.0) 2 14 (8.1) Smoking prior to pregnancy 32 (18.6) NYHA class I 127 (73.8) II 44 (25.6) III 1 (0.6) Modified WHO class* I 32 (18.6) II 94 (54.7) III 44 (25.6) IV 2 (1.2) Mechanical valve prosthesis 10 (5.8) Sustained symptomatic brady- or tachyarrhythmia requiring treatment 16 (9.3) Pacemaker 5 (2.9) Congestive heart failure 5 (2.9) Hypertension 11 (6.4) Cardiac medication 35 (20.4) ACE-inhibitor 2 (1.2) Beta-blocker 21 (12.2) Calcium channel blocker 3 (1.7) Anticoagulants 14 (8.1) 4

61 60 Chapter 4 Table 1. Maternal baseline characteristics (before pregnancy). N=172 women. (continued) N (%) Type of congenital lesion Left sided lesions 49 Aortic valve stenosis / Bicuspid aortic valve 27 (15.7) Surgically repaired Aortic coarctation 21 (12.2) Other 1 (0.6) Right sided lesions 51 Tetralogy of Fallot after repair 33 (19.2) Pulmonary valve stenosis 16 (9.3) Ebstein s anomaly 2 (1.2) Shunt lesions 46 Ventricular septal defect 17 (9.9) Atrial septum defect 17 (9.9) Atrioventricular septal defect 7 (4.1) Abnormal pulmonary venous return 5 (2.9) Connective tissue disorder 8 (4.7) Marfan syndrome 7 Loeyz-Dietz syndrome 1 Complex CHD Complete transposition of great arteries 12 (7.0) D-TGA with Mustard or Senning 10 D-TGA with arterial switch 2 Congenital corrected TGA 1 (0.6) Single ventricle with Fontan physiology 3 (1.7) Other complex cyanotic heart disease 2 (1.2) Echocardiographic parameters Systemic atrioventricular valve regurgitation** 4 (2.7) Pulmonary atrioventricular valve regurgitation** 10 (6.9) Pulmonary valve stenosis 10 (7.6) Pulmonary valve regurgitation** 27 (19.9) Aortic valve stenosis 12 (8.3) Aortic valve regurgitation** 8 (5.4) Systemic ventricular systolic dysfunction # 9 (6.5) Subpulmonary ventricular systolic dysfunction** 16 (13.8) Pregnancy outcome Primary cardiac event during pregnancy 18 (10.5) ACE, angiotensin-converting enzyme. TGA, transposition of the great arteries. *Modified World Health Organization class according to ESC guidelines. 16 Patient with cleft mitral valve. 1 patient with a corrected truncus arteriosus, type A; 1 Patient with pulmonary valve atresia, atrial septum defect and intact ventricular septum. Missing data excluded from analysis. **Moderate or severe regurgitation; peak gradient 36 mmhg; # Ejection fraction < 45%; **TAPSE < 16 mm.

62 Cardiac function & cardiac events one year post-partum. 61 Cardiovascular events one year post-partum Cardiovascular events were seen after 11 pregnancies (6.4 %). The median follow up time was 1.07 years (IQR years). Women with cardiovascular events 1-year post-partum had higher modified WHO risk classifications prior to pregnancy (p=0.010). They had more often a mechanical valve prosthesis (27.3% vs. 4.4%; p=0.024) or a history of arrhythmia (36.4% vs. 7.5%; p=0.021). Women with cardiovascular events post-partum were more likely to use cardiac medication preconception (54.6% vs. 18.0%; p=0.010), in particular beta-blockers (36.4% vs. 10.6%; p=0.031) and anticoagulation therapy (45.5% vs. 5.6%; p=0.001). No differences were observed in systemic or subpulmonary ventricular function prior to pregnancy (0.0% vs. 6.9%; p=1.00)(28.6% vs. 12.8; p= 0.25). Table 2 shows the cardiovascular events post-partum and underlying cardiac lesions. Of the women with cardiovascular events post-partum, 63.6% had no history of cardiovascular events prior to pregnancy and 71.4% developed new arrhythmia. 4 Table 2. Characteristics of patients with cardiovascular events after pregnancy (N=11). Patient number Type of congenital lesion Corrective procedure Cardiovascular event after pregnancy 1 Aortic valve stenosis --- Heart failure 2 Aortic arch atresia type A Correction with end-to-side anastomosis Ventricular tachycardia 3 Double inlet left ventricle - hypoplastic right ventricle. Fontan circulation Atrial flutter 4 Hypoplastic right ventricle with Ebstein s anomaly and tricuspid valve stenosis Fontan circulation Atrial tachycardia Protein losing enteropathy. 5 Ebstein s anomaly Tricuspid valve plasty Atrial fibrillation 6 Tetralogy of Fallot Total correction with transannulair patch Atrial fibrillation 7 Tetralogy of Fallot Total correction with transannulair patch Atrial tachycardia 8 Bicuspid aortic valve 9 Bicuspid aortic valve Mechanical aortic valve prosthesis Ross-procedure, mechanical aortic valve and mechanical pulmonary valve prosthesis Mechanical mitral valve prosthesis Symptomatic ventricular ectopia Atrial flutter 10 Cleft mitral valve Atrial fibrillation 11 Ventricular septal defect --- Infective endocarditis

63 62 Chapter 4 Women with cardiovascular events during pregnancy had a higher risk of cardiovascular events post-partum compared with women without cardiovascular events during pregnancy (HR 7.1; 95% CI ; p=0.003)(figure 1). The nature of cardiovascular events postpartum was identical to the events during pregnancy in 50% of the cases. Multivariable regression analysis was not performed because of the low incidence of cardiovascular events after pregnancy. Freedom from late cardiovascular events 'No cardiac event during pregnancy' 'Cardiac event during pregnancy' Follow up duration (years) Number at risk No cardiac event during pregnancy Cardiac event during pregnancy Figure 1: Freedom from late cardiovascular events after pregnancy in women with cardiac events during pregnancy ( ) and women without cardiac events during pregnancy ( ). Nt-proBNP 1-year post-partum Nt-proBNP values 1-year post-partum were available in 90 out of 172 women and ranged between 5 and 1044 pg/ml. Median Nt-proBNP values did not differ between 20 weeks of gestation and 1-year postpartum (106 ( ) vs ( ) pg/ml; p=0.13). Women with cardiovascular events post-partum had significant higher Nt-proBNP values at 20 weeks gestation and 1-year post-partum compared with women without cardiovascular events post-partum (191 ( ) vs (57-167) (p=0.049) and 306 ( ) vs. 105 (54-187) pg/ml (p=0.014), respectively). In addition, women with cardiovascular events during pregnancy had significantly higher Nt-proBNP values 1-year post-partum compared with women without cardiovascular events during pregnancy (303.5 ( ) vs (54-183) pg/ml; p=0.0077). Nt-proBNP level < 128 pg/ml at 20 weeks of gestation had a negative predictive value of 98.7% for the occurrence of cardiovascular events 1-year post-partum.

64 Cardiac function & cardiac events one year post-partum. 63 Cardiac function and aerobic exercise capacity one year post-partum Table 3 provides an overview of NYHA functional class, cardiac function parameters, and aerobic exercise capacity parameters before and 1-year after pregnancy. New York Heart Association class did not change significantly, although 2 women experienced NYHA functional class deterioration ( 2 functional class) post-partum (1.2 %). In 5 women (33.3%) with NYHA class deterioration ( 2 functional classes) during pregnancy, NYHA class 1-year postpartum did not completely recover. One year post-partum, no significant differences were seen in NYHA functional class, systemic ventricular end-diastolic diameter, systemic ventricular ejection fraction, and subpulmonary Table 3. Comparison of cardiac function parameters preconception and one year post-partum. N Preconception 1-year post-partum P-value General parameters NYHA class NYHA class I 127 (73.8%) 128 (74.9%) NYHA class II 44 (25.6%) 39 (22.8%) NYHA class III 1 (0.6%) 3 (1.8%) NYHA class IV 0 (0.0%) 1 (0.6%) Systemic ventricular size, mass and systolic function* Mean difference (95% CI) Systolic ventricular end-diastolic diameter (mm) ( ) 47.0 ( ) ( ) Systemic ventricular mass/bsa (g/m 2 ) ( ) 70.9 ( ) ( ) Systemic ventricular ejection fraction ± ± ( ) Systemic ventricular diastolic function* Left atrial volume (ml) ( ) ( ) ( ) E/A ratio ( ) 1.5 ( ) ( ) Mean E (septal-lateral) (cm/s) ( ) 10.2 ( ) ( ) E/E ( ) 8.8 ( ) ( ) Right ventricular size and function* Right ventricular end-diastolic diameter (mm) ( ) 36.0 ( ) ( ) Tricuspid Annular Plane Systolic Excursion (TAPSE (mm)) ± ± ( ) Right ventricular S (cm/s) ( ) 8.9 ( ) ( ) Systemic atrioventricular valve regurgitation (2.9%) 4 (2.9%) 0.53 Pulmonary atrioventricular valve regurgitation (7.3%) 11 (7.9%) 0.21 Pulmonary valve regurgitation (20.3%) 30 (22.4%) 0.43 Pulmonary valve stenosis** (8.1%) 7 (5.7%)

65 64 Chapter 4 Table 3. Comparison of cardiac function parameters preconception and one year post-partum. ( continued) Mean N Preconception 1-year post-partum P-value difference (95% CI) Aortic valve regurgitation (5.0%) 8 (5.7%) 0.44 Aortic valve stenosis** (8.0%) 11 (8.0%) 0.46 Aerobic exercise capacity VO2 max (ml/kg/min) ± ± ( ) Anaerobic threshold (ml/kg/min) ( ) 15.5 ( ) ( ) BSA, body surface area; VO2 max, oxygen consumption per unit time. * Excluding systemic right ventricles. Early diastolic tissue Doppler velocity of systemic ventricular annular ring; Systolic tissue Doppler velocity of subpulmonary ventricular annular ring; Moderate or severe regurgitation; **peak gradient 36 mmhg. ventricular end-diastolic diameter or function. Valvular function had not changed. Aerobic exercise capacity testing prior to pregnancy was available in 18.6% of the patients, and no differences were observed post-partum. In patients with systemic right ventricles (RV), systolic function of the systemic RV (qualitative assessment) remained unchanged; the diameter of the systemic RV increased (34.0 mm ( ) vs mm ( ) (p=0.035)). Women with cardiovascular events during pregnancy had significantly larger subpulmonary ventricular end-diastolic diameter (apical 4-chamber view) post-partum compared with pre-pregnancy (39.0 ( ) mm vs ( ) mm; p=0.028; mean difference -4.8 mm (95% CI )). This could not be attributed to an increased incidence of right-sided valve insufficiency or tetralogy of Fallot in this group. In addition, women with cardiovascular events during pregnancy had significantly more often aggravation of aortic valve regurgitation 1-year post-partum (p=0.046). The clinical and echocardiographic data of the patients with subpulmonary ventricular dilation are summarized in table 4.

66 Cardiac function & cardiac events one year post-partum. 65 Table 4. Clinical and echocardiographic characteristics of women with an increase in subpulmonary ventricular diameters 1-year post-partum and cardiovascular events during pregnancy (n=10/18). NYHA functional class RV diameter (mm) 1-year post-partum 1-year post-partum preconception preconception Tricuspid valve regurgitation * Pulmonary valve stenosis * Pulmonary valve regurgitation * Diagnosis - corrective surgery Patient number 1 ToF repaired Severe Mild Absent II I ToF - repaired Moderate Moderate Mild II II pavsd - repaired Absent Absent Mild II II pavsd - repaired Absent Absent Absent I I PS valvulotomy Severe Absent Absent I I PA - intact ventricular septum repaired Severe Mild Moderate I II Aortic arch atresia type A repaired Absent Absent Mild II II Bicuspid aortic valve Ross procedure. Severe aortic and pulmonary valve insufficiency: 2x mechanical valve prosthesis Absent Absent Severe II III Sinus venosus defect - repaired Absent Mild Mild II I Cleft mitral valve - mechanical valve prosthesis Absent Absent Absent I II * Valve dysfunction preconception. ToF, tetralogy of Fallot; pavsd, partial atrioventricular septal defect; PS, pulmonary valve stenosis; PA, pulmonary valve atresia; BiAoV, bicuspid aortic valve. 4

67 66 Chapter 4 Discussion The present study shows that the incidence of cardiovascular events in the first year postpartum is relatively low and that the majority of patients go through pregnancy without permanent damage to their cardiac function. Women with cardiovascular complications during pregnancy are at higher risk of developing cardiovascular complications 1-year postpartum and are prone to dilatation of the subpulmonary ventricle 1-year post-partum. In this study, the incidence of cardiovascular events 1-year post-partum was 6.4%. This incidence is lower than previously reported in a large retrospective study by Balint and colleagues (12%) 8. This is probably due to differences in study population, study design and follow-up duration. Most of the cardiovascular events post-partum were atrial arrhythmias. In the general CHD population, the overall 20-year risk of developing atrial arrhythmia is 7% in a 20-year old CHD patient 17. It is not clear if pregnancy as such or the number of pregnancies influences this risk. Women with cardiovascular events during pregnancy have higher risk of developing cardiovascular events 1-year post-partum. This finding is consistent with Balint et al 8. Probably, these women have limited cardiac reserve, and the hemodynamic requirements of pregnancy cannot be met, which results in complications such as cardiac failure and arrhythmia. This is reflected by high Nt-proBNP levels during pregnancy (which was identified before as predictor of cardiovascular events during pregnancy 18 ) and similarly by the higher Nt-proBNP levels after pregnancy in women with cardiovascular events during pregnancy or in the first year post-partum. The current study shows that cardiac function and chamber dimensions did not change after pregnancy compared with pre-pregnancy and that valve dysfunction did not aggravate. However, in women with cardiovascular events during pregnancy, subpulmonary ventricular dilatation was seen 1-year post-partum. This could not be attributed to an increased incidence of patients with Tetralogy of Fallot or right-sided valve insufficiency in this group. This is, partly, in contrast to the findings Uebing et al, where this effect was attributable to patients with Tetralogy of Fallot 5. Interestingly and in line with results of Uebing et al., we found no association between subpulmonary ventricular dilatation and presence of pulmonary valve insufficiency. The increasing number of studies describing subpulmonary ventricular dilatation after pregnancy in women with Tetralogy of Fallot 11,19 and the identification of subpulmonary ventricular dysfunction (TAPSE < 16 mm) as a predictor of cardiovascular events during pregnancy suggests that the subpulmonary ventricle may be vulnerable for consequences of the physiological volume overload of pregnancy 18. The finding that Tetralogy of Fallot patients with more severe disease needing medication are prone for complications during pregnancy is also in line with these findings 20. Data on pregnancy related changes in subpulmonary ventricular function and dimension in healthy women are scarce, which makes interpretation

68 Cardiac function & cardiac events one year post-partum. 67 of subpulmonary ventricular changes after pregnancy in women with CHD difficult 21. As is known from a small study in healthy women, the subpulmonary ventricular volume increases far more during pregnancy than the systemic ventricle 22. This physiological dilatation may be aggravated in women with CHD because of abnormal myoarchitecture, damage from prior surgery, and pre-existent volume or pressure overload, leading to persistent subpulmonary ventricle dilatation 23. The clinical implications of the findings presented here are not completely clear yet. Nevertheless, the observations made in this study are important in counseling women in their wish to pursue pregnancy. The present study indicates that women with CHD should have regular check-ups after pregnancy. This is particularly important for women with cardiovascular events during pregnancy because they appear to be prone for late cardiovascular events and subpulmonary ventricular dilatation. Strengths and limitations 4 This is the first large prospective study concerning cardiac function after pregnancy in women with CHD. Because of our protocol design (inclusion before 20 weeks of gestation) missing data were inevitable because pre-pregnancy data were collected retrospectively. Especially pre-pregnancy aerobic exercise capacity tests were only limited available. Because of the limited number of cardiovascular events late after pregnancy, it was not feasible to perform multivariable Cox-regression models to investigate independent predictors of late cardiovascular events and cardiac function deterioration. Because this was an analysis of the secondary outcomes of the ZAHARA II study, it should be regarded as exploratory, and potentially relevant results may have failed to reach significance. Because all women with structural CHD were eligible for inclusion, our cohort is heterogeneous. Some individual lesions, in particular some high risk lesions, are underrepresented, which may partly be the result of well-organized pregnancy counseling in the Netherlands. These limitations should be kept in mind when extrapolating the findings, and the results should be interpreted with caution. Conclusion and clinical implications In this cohort, most women with congenital heart disease go through pregnancy without permanent damage to their cardiac function. Women with congenital heart disease who experience cardiovascular events during pregnancy or have high modified WHO risk classification were prone to develop cardiovascular events after pregnancy. In addition, an increase in subpulmonary ventricular diameter post-

69 68 Chapter 4 partum was seen in women with cardiovascular events during pregnancy. These findings indicate that regular follow-up after pregnancy, especially of women with cardiovascular events during pregnancy, is important. Large, lesion-specific, prospective studies are warranted to investigate the long-term effect of pregnancy on cardiac function and prognosis. This information is needed for effective counseling of patients regarding their wish to pursue pregnancy. In addition, to make advances in research concerning risk stratification and prognosis for women with CHD contemplating pregnancy, consequent guideline based preconception evaluation is clearly warranted 16.

70 Cardiac function & cardiac events one year post-partum. 69 References 1. Drenthen W, Pieper PG, Roos-Hesselink JW, et al. Outcome of pregnancy in women with congenital heart disease: A literature review. J Am Coll Cardiol. 2007; 49: Drenthen W, Boersma E, Balci A, et al. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J. 2010; 31: Siu SC, Sermer M, Colman JM, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation. 2001; 104: Khairy P, Ouyang DW, Fernandes SM, et al. Pregnancy outcomes in women with congenital heart disease. Circulation. 2006; 113: Uebing A, Arvanitis P, Li W, et al. Effect of pregnancy on clinical status and ventricular function in women with heart disease. Int J Cardiol. 2010; 139: Guedes A, Mercier LA, Leduc L, et al. Impact of pregnancy on the systemic right ventricle after a mustard operation for transposition of the great arteries. J Am Coll Cardiol. 2004; 44: Silversides CK, Colman JM, Sermer M, et al. Early and intermediate-term outcomes of pregnancy with congenital aortic stenosis. Am J Cardiol. 2003; 91: Balint OH, Siu SC, Mason J, et al. Cardiac outcomes after pregnancy in women with congenital heart disease. Heart. 2010; 96: Tzemos N, Silversides CK, Colman JM, et al. Late cardiac outcomes after pregnancy in women with congenital aortic stenosis. Am Heart J. 2009; 157: Zentner D, Wheeler M, Grigg L. Does pregnancy contribute to systemic right ventricular dysfunction in adults with an atrial switch operation? Heart Lung Circ. 2012; 21: Kamiya CA, Iwamiya T, Neki R, et al. Outcome of pregnancy and effects on the right heart in women with repaired tetralogy of fallot. Circ J. 2012; 76: Sliwa K, Hilfiker-Kleiner D, Petrie MC, et al. Current state of knowledge on aetiology, diagnosis, management, and therapy of peripartum cardiomyopathy: A position statement from the heart failure association of the european society of cardiology working group on peripartum cardiomyopathy. Eur J Heart Fail. 2010; 12: Balci A, Sollie KM, Mulder BJ, et al. Associations between cardiovascular parameters and uteroplacental doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the zwangerschap bij aangeboren hartafwijking (ZAHARA) II study. Am Heart J. 2011; 161: 269,275.e Pieper PG, Balci A, Aarnoudse JG, et al. Uteroplacental blood flow, cardiac function, and pregnancy outcome in women with congenital heart disease. Circulation. 2013; 128: McMurray JJ, Adamopoulos S, Anker SD, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The task force for the diagnosis and treatment of acute and chronic heart failure 2012 of the european society of cardiology. developed in collaboration with the heart failure association (HFA) of the ESC. Eur Heart J. 2012; 33: Regitz-Zagrosek V, Blomstrom Lundqvist C, Borghi C, et al. ESC guidelines on the management of cardiovascular diseases during pregnancy: The task force on the management of cardiovascular diseases during pregnancy of the european society of cardiology (ESC). Eur Heart J. 2011; 32: Bouchardy J, Therrien J, Pilote L, et al. Atrial arrhythmias in adults with congenital heart disease. Circulation. 2009; 120:

71 70 Chapter Kampman MA, Balci A, van Veldhuisen DJ, et al. N-terminal pro-b-type natriuretic peptide predicts cardiovascular complications in pregnant women with congenital heart disease. Eur Heart J. 2014; 35: Egidy Assenza G, Cassater D, Landzberg M, et al. The effects of pregnancy on right ventricular remodeling in women with repaired tetralogy of fallot. Int J Cardiol. 2013; 168: Balci A, Drenthen W, Mulder BJ, et al. Pregnancy in women with corrected tetralogy of fallot: Occurrence and predictors of adverse events. Am Heart J. 2011; 161: Cornette J, Ruys TP, Rossi A, et al. Hemodynamic adaptation to pregnancy in women with structural heart disease. Int J Cardiol. 2013; 168: Campos O. Doppler echocardiography during pregnancy: Physiological and abnormal findings. Echocardiography. 1996; 13: Sanchez-Quintana D, Anderson RH, Ho SY. Ventricular myoarchitecture in tetralogy of fallot. Heart. 1996; 76:

72

73

74 Part II 8 Mechanisms A

75

76 Chapter 5 Uteroplacental blood flow, cardiac function and pregnancy outcome in women with congenital heart disease Petronella G. Pieper Ali Balci Jan G. Aarnoudse Marlies A.M. Kampman Krystyna M. Sollie-Szarynska Henk Groen Barbara J.M. Mulder Martijn A. Oudijk Jolien W. Roos-Hesselink Jerome Cornette Arie P.J. van Dijk Marc E. Spaanderman Willem Drenthen Dirk J. van Veldhuisen on behalf of the ZAHARA II investigators A Circulation 2013; 128:

77 76 Chapter 5 Abstract Background Pregnant women with congenital heart disease (CHD) are susceptible to cardiovascular, obstetric, and offspring complications. In women with CHD cardiac dysfunction may compromise uteroplacental flow and contribute to the increased incidence of obstetric and offspring events. Methods and Results We performed a prospective multicenter cohort study of pregnant women with CHD and healthy pregnant women. We compared clinical, laboratory, echocardiographic and uteroplacental Doppler flow (UDF) parameters at 20 and 32 weeks gestation, and pregnancy outcome. We related cardiovascular parameters to UDF parameters and pregnancy outcome in women with CHD. We included 209 women with CHD and 70 healthy women. Cardiovascular parameters (N-terminal pro-b-type natriuretic peptide (Nt-proBNP), left and right ventricular function) differed between both groups. UDF parameters were impaired in CHD women (umbilical artery pulsatility and resistance index at 32 weeks in CHD versus healthy women, p= and p=0.017). The following cardiovascular parameters prepregnancy and at 20 weeks gestation were associated with UDF (umbilical artery resistance index) at 32 weeks at multivariable analysis: (1) right ventricular function (tricuspid annular plane systolic excursion) (p=0.002), (2) high N-terminal pro-b-type natriuretic peptide (p=0.085), (3) systemic (p=0.001), and (4) pulmonary (p=0.045) atrioventricular valve regurgitation. Women with CHD had more obstetric (58.9% versus 32.9%, p<0.0001) and offspring events (35.4% versus 18.6%, p=0.008) than healthy women. Impaired UDF was associated with adverse obstetric and offspring outcome. Conclusions UDF parameters are abnormal in pregnant women with CHD. Cardiovascular function is associated with an abnormal pattern of UDF. Compromised UDF may be a key factor in the high incidence of offspring and obstetric complications in this population. Keywords: complications, congenital, heart diseases, placental circulation, pregnancy.

78 Uteroplacental blood flow in women with CHD. 77 Introduction Congenital heart disease (CHD) occurs in around 1% of newborns, and 50% of these children are female. The extensive evolution of cardiac surgery for CHD has resulted in a large population of adult women with CHD. Many of them pursue pregnancy. Pregnancy in these women is associated with cardiovascular complications, which occur in approximately 10% of pregnancies. Moreover, obstetric and offspring complications are also more prevalent than in healthy pregnant women. 1-5 In women with CHD, offspring complications are related to maternal cardiac function. 5 However, the underlying pathophysiology of this relationship is not completely unraveled. In healthy women with intrauterine growth restriction or hypertensive disorders of pregnancy, the process of placentation is often disturbed, resulting in abnormal uterine and umbilical artery Doppler flow patterns. 6 Such abnormal uteroplacental Doppler flow (UDF) patterns are validated markers of adverse offspring outcome. Moreover, co-existing maternal cardiac and vascular function abnormalities have been demonstrated. 7-9 Whether these abnormalities may be explained by damage caused by circulating angiogenic factors secreted by the placenta, or abnormal placentation and offspring outcome are caused by (subtle) underlying cardiac and vascular disease, is unknown. In women with CHD, the relation between cardiac function (as expressed in N-terminal pro-b-type natriuretic peptide (Nt-proBNP) levels and echocardiographic parameters), UDF patterns, and offspring complications has not been investigated. We hypothesized that preexisting cardiac dysfunction in pregnant women with CHD results not only in cardiovascular complications, but also can lead to disturbed placentation with abnormal UDF patterns, thus compromising normal growth and development of the fetus and contributing to offspring complications in pregnancy. To confirm this hypothesis, we performed a prospective study in women with CHD and healthy women. The primary objectives of this study are (1) to compare the cardiovascular clinical, biochemical, and echocardiographic parameters, and UDF patterns, as well, of pregnant women with CHD with healthy pregnant women, and (2) to relate maternal cardiovascular parameters in women with CHD to UDF patterns. The secondary objective is to relate UDF patterns to obstetric and offspring outcome. This study will give insight in the pathophysiology of offspring complications in women with CHD. 5 Patients and methods Design and setting This prospective observational multicenter cohort study was conducted between March 2008 and August The extensive study design of the Zwangerschap en Aangeboren HARtAfwijkingen (ZAHARA) II study was published previously and is summarized below. 10

79 78 Chapter 5 Patient selection Female patients with structural CHD (aged 18 years) reporting pregnancy with a duration 20 weeks who provided written informed consent and who were followed in 1 of the 8 participating tertiary hospitals participated in the study. In the Netherlands, pregnancies of all healthy women are routinely handled by midwives, regardless of socio-economic status. Therefore, simultaneously, healthy pregnant women were recruited from midwifery practices. Miscarriages or termination before 20 weeks gestation and twin pregnancies were excluded, as were women with known illicit drug or alcohol abuse. The study was approved by the Medical Ethics Committee of all participating hospitals. Sample size calculation One of the primary aims of the ZAHARA II study was to compare the UDF, expressed as pulsatility and resistance indices in the uterine and umbilical artery, during pregnancy between women with CHD and healthy controls. A total sample size of 240 subjects (160 patients and 60 (healthy) controls) achieves 80% power to detect a difference of 0.05 in pulsatility index (25% of the expected standard deviation) among the means versus the alternative of equal means using an independent samples T test with a 0.05 significance level. The common standard deviation within a group is assumed to be The sample size for comparison of pulsatility index (PI) was based on an effect size of For resistance index (RI), we would use the same assumption and therefore arrive at a similar sample size. Preconception characteristics Baseline data were recorded at the first prenatal visit and included: maternal age, obstetric history, cardiovascular history, co-morbidity, prepregnancy cardiac status and echocardiographic recordings (including systemic and pulmonary ventricular function and valvular function), use of medication, and alcohol and smoking history. Evaluation at 20 and 32 weeks At 20 and 32 weeks gestation, participants underwent clinical and laboratory evaluation (including serum hemoglobin and Nt-proBNP), echocardiographic examination, and UDF registration (PI and RI of umbilical artery and of right and left uterine artery, and the presence of early diastolic notching). All echocardiographic recordings were made on commercially available Philips or Vingmed General Electric ultrasound equipment. Echocardiograms were evaluated off-line by 3 experienced cardiologists (each of them reviewed a part of the echocardiograms). A fourth cardiologist checked the consistency and accuracy of the echocardiography data. Chamber quantification and ventricular and valvular function were assessed according to current guidelines Because Tricuspid Annular Plane Systolic Excursion (TAPSE) and ejection fraction by Simpsons rule are not validated in patients with single ventricles and systemic right ventricles, these measurements were not performed in these patient groups.

80 Uteroplacental blood flow in women with CHD. 79 Obstetric and offspring events Extensive definitions of obstetric and offspring events were published previously and are summarized below. 10 Obstetric events were noncardiac death, pregnancy-induced hypertension, preeclampsia, eclampsia, gestational diabetes mellitus, HELLP syndrome (hemolysis, elevated liver enzymes, low platelet syndrome), hyperemesis gravidarum, assisted delivery, postpartum hemorrhage, preterm labor, preterm premature rupture of membranes, and abruptio placentae. Offspring events were fetal death, neonatal death, intraventricular hemorrhage, neonatal respiratory distress syndrome, infections leading to hospital admission, neonatal intensive care unit admission, premature birth, occurrence of CHD, occurrence of other congenital disease, small for gestational age, and low birth weight. Statistical analysis We used SPSS (IBM SPSS Statistics, version 19.0, IBM SPSS Statistics, IBM Corporation, Armonk, NY) and STATA (version 12.0, StatCorp LP, College Station, TX) for statistical analysis. Continuous variables with normal distribution are presented as mean with standard deviation (± standard deviation), non-normally distributed variables as median with interquartile ranges, and dichotomous variables are presented as absolute numbers with percentages. Cardiovascular parameters and UDF parameters at 20 and 32 weeks gestation, and pregnancy outcome, were compared between women with CHD and healthy women. Comparison of continuous variables between groups was performed with the Student t test or Mann- Whitney U test, depending on distribution, with and without logarithmic transformation. Longitudinal comparison of continuous variables within CHD and healthy pregnancy groups at two time points (20 and 32 weeks) was performed using the paired t test. We compared the PI and RI within the groups CHD and healthy women and compared these measurements of the CHD group with measurements of healthy women at 20 weeks and at 32 weeks, as well. For the comparison of dichotomous variables, we used the χ 2 test or Fisher exact test, as appropriate. A P-value <0.05 was considered statistically significant and all P-values are two-sided. Uni- and multivariable linear and logistic regression analyses were performed to assess associations between cardiovascular parameters and UDF parameters during pregnancy and between UDF parameters and obstetric and offspring outcome, as well, in women with CHD. The following predefined variables were assessed in univariable analysis: age, disease complexity 15, risk of cardiovascular complications according to modified World Health Organization class 16, body surface area, body mass index, New York Heart Association functional class, resting heart rate, heart rhythm, mean arterial pressure, smoking during pregnancy, cardiac medication use, pre-pregnancy hypertension, anemia, high Nt-proBNP, valve dysfunction (stenosis and regurgitation), left ventricular diastolic diameter/body surface area, left ventricular mass/body surface area, left ventricular ejection fraction, mean left ventricular systolic tissue velocity (S ) (septal-lateral), left atrial volume, left ventricular early 5

81 80 Chapter 5 to atrial mitral inflow velocity ratio, left ventricular mitral inflow deceleration time, mean left ventricular early diastolic tissue velocity (E ) (septal-lateral), right ventricular diastolic diameter, right ventricular function (TAPSE), and right ventricular systolic tissue velocity (S ). In addition, variables at 20 weeks gestation were adjusted for pre-pregnancy values that were significantly associated with the studied endpoints (p<0.05), and variables at 32 weeks gestation were adjusted for values that were significantly associated with the studied endpoints pre-pregnancy and at 20 weeks gestation. Variables that were strongly associated with the studied endpoints (p<0.10) or variables considered relevant (p> 0.10) entered the multivariable model. The final multivariable model was constructed by backward deletion of the least significant characteristic, with a criterion for deletion of p When performing the multivariable model, we used pairwise deletion of cases to deal with missing variables. Results Pre-pregnancy baseline characteristics We recruited 234 pregnant women with CHD. Twenty-five women were excluded, because of miscarriage (n=11), serious protocol violation (n=6), twin pregnancy (n=4), or withdrawal of informed consent (n=4). Simultaneously, 70 healthy, age and parity-matched pregnant control women with a singleton pregnancy were recruited. No significant difference was observed between women with CHD and healthy pregnant women with respect to maternal age at conception (28.7±4.4 versus 29.2±4.5, p=0.44), parity (64.1% versus 62.9% nulliparous, p=0.46), ethnic origin (95.7% versus 97.1% white, p=0.35), and pre-pregnancy body mass index (23.5±3.9 versus 23.1±3.9, p=0.56). More healthy women smoked pre-pregnancy compared to CHD women (33.3% versus 20.7%, p=0.03). None of the women had impaired glucose tolerance or hypertensive disorder of pregnancy at the time of recruitment. Table 1 shows pre-pregnancy cardiovascular data of the CHD cohort. None of the women had uncorrected cyanotic disease or SpO 2 < 90%; mean oxygen saturation was 98.5±1.5% at 20 weeks gestation. Of patients with shunt lesions, 78% had a history of correction of the defect. Cardiac medication was used before pregnancy by 15.8% of women with CHD; 7.2% were on anticoagulation therapy and 12.4% used a beta-blocker. Sinus rhythm was present in 88% (N=185). Systemic ventricular ejection fraction was known in 161 CHD women and was below 45% in 8.1% of these women. Pre-pregnancy right ventricular (RV) function (tricuspid annular plane systolic excursion (TAPSE) was known in 138 CHD women; RV dysfunction (TAPSE < 16 mm) existed in 14.5% of these women. Three women conceived through intracytoplasmatic sperm injection. Six women had a history of thyroid dysfunction; however thyroid-stimulating hormone was normal preconception.

82 Uteroplacental blood flow in women with CHD. 81 Table 1. Maternal pre-pregnancy characteristics in women with CHD (n=209) N % Underlying CHD Left sided lesions Aortic stenosis / bicuspid aortic valve Aortic coarctation Other Right sided lesions Ebstein s anomaly Pulmonary stenosis Tetralogy of Fallot Shunt lesions Abnormal pulmonary venous return 6 10 Atrial septal defect Atrioventricular septal defect Ventricular septal defect Connective tissue disease Marfan syndrome Loeys-Dietz syndrome Complex CHD Transposition of great arteries (Mustard/Senning operation) Transposition of great arteries (arterial switch operation) Congenitally corrected transposition of great arteries Fontan circulation Other complex CHD Disease complexity* Simple Moderate complex Complex Modified WHO classification (risk of pregnancy) Class 1 (low risk) Class 2 (moderately high risk) Class 3 (high risk) Medical history History of heart failure History of arrhythmia History of hypertension History of diabetes Pacemaker Mechanical valve prosthesis Biological-valve prosthesis

83 82 Chapter 5 Table 1. Maternal pre-pregnancy characteristics in women with CHD (n=209) (continued) N % Medication use pre-pregnancy Cardiac medication Beta-blockers Other cardiac medication Vitamin K-antagonists/Heparin New York Heart Association (NYHA) functional class NYHA functional class I NYHA functional class II NYHA functional class III In underlying heart disease, several groups are mentioned (ie, left sided lesions, right sided lesions, etc). The n and % in roman are then a % for such a group. Within each group, subdiagnoses are mentioned (ie, aortic stenosis, aortic coarctation, other). The n and % of subdiagnoses are shown in italics. CHD indicates congenital heart disease; NYHA, New York Heart Association; and WHO, World Health Organization.* Disease complexity: according to Warnes et al. 15 Comparison of cardiovascular and UDF parameters between pregnant women with CHD and healthy pregnant women New York Heart Association functional class deterioration > 1 class at 32 weeks compared to pre-pregnancy occurred only in CHD and not in healthy women: 10.1% versus 0%, p= We compared laboratory, echocardiographic, and UDF parameters between CHD and healthy cohorts at 20 and 32 weeks gestation (Table 2). Nt-proBNP was higher throughout pregnancy in women with CHD and decreased during pregnancy in both groups; the decrease was significantly greater in CHD women (p=0.04). Systemic ventricular mass corrected for body surface area was higher and increased (p<0.005) only in women with CHD. Systemic ventricular ejection fraction did not change significantly in both groups. Several diastolic systemic ventricular function parameters were significantly worse in CHD women: systemic ventricular annular velocity (E ) was lower and diastolic filling pressure (E/E ) higher; change during pregnancy was comparable between both groups. RV systolic function (represented by TAPSE and systolic annular velocity) was worse in CHD and decreased significantly in CHD women only (p=0.017 and p=0.009, respectively). Figure 1 shows UDF parameters at 20 and 32 weeks. Uterine and umbilical artery PI and RI were higher throughout pregnancy in the CHD group and decreased in both groups. Uterine artery PI and RI were both measured in 139 women at 20 weeks, and umbilical artery PI and RI were both measured in 157 women at 32 weeks, while 51 women did not have any uterine artery UDF measurement at 20 weeks, and 23 women did not have any of umbilical artery measurements at 32 weeks. Missing measurements were mainly attributable to logistic reasons.

84 Uteroplacental blood flow in women with CHD. 83 all p-values < CHD Mean Resistance Index Healthy Healthy CHD p =0.017 umbilical artery uterine artery A Gestational week Figure 1: Uteroplacental Doppler flow parameters: pulsatility index (A) and resistance index (B) of mean of right and left uterine artery and of umbilical artery at 20 and 32 weeks of pregnancy, in women with CHD and healthy women. Mean PI (A) and mean RI (B) differed significantly between 20 and 32 weeks, in both uterine and umbilical artery and in healthy controls and CHD patients (as represented by the horizontal line indicating all p-values < ). Significant differences in separate analyses comparing groups at 20 weeks, and at 32 weeks are indicated by vertical lines with p-values. CHD indicates congenital heart disease; PI, pulsatility index; and RI, resistance index. Relation of cardiovascular parameters and UDF indices in women with CHD We related maternal cardiovascular to UDF parameters. PI and RI were not both measured in all patients. Because results were comparable, we present RI data, in accordance with previous studies presenting data on the relation of UDF parameters with cardiac function in healthy women. 17 RI was available in 141 women for the uterine artery at 20 weeks gestation and in 157 women for the umbilical artery at 32 weeks. Univariable analysis revealed the following baseline (pre-pregnancy) variables to be associated with uterine artery RI (20 weeks): parity, preconception heart rate, systemic atrioventricular valve regurgitation and left atrial volume. Heart rate, use of cardiac medication, and TAPSE at 20 weeks were also associated with uterine artery RI (20weeks). Hypertension was not significantly associated with UDF in our cohort (B=0.028, p=0.28). Multivariable analysis rendered parity (B=0.04, p=0.048), resting heart rate at 20 weeks (B=-0.002, p=0.006), and use of cardiac medication at 20 weeks (B=0.08, p=0.035) significant. Univariable analysis and multivariable models for the prediction of umbilical artery RI (32 weeks) are presented in Tables 3 and B 5

85 84 Chapter 5 Table 2. Comparison of women with CHD (n=209**) with healthy (n=70**) women during pregnancy. Gestational week 20 Gestational week 32 CHD n Healthy n P value CHD n Healthy n P value General parameters Smoking during pregnancy 10.0% 2.9% % 2.9% Cardiac medication 11.0% 0 % % 0% < NYHA class I 53.1% % 37.1% NYHA class II 39.7% 41.4% 46.4% 62.9% NYHA class III 7.2% 0 % 14.0% 0% MAP (mmhg) 81.4± ± ± ± Laboratory parameters Hb (mmol/l) 7.5± ± ± ± NT-proBNP (pg/ml) ( ) ( ) 49 < ( ) ( ) 48 < Systemic ventricular size, mass and systolic function* Systemic ventricular end diastolic diameter 47.4± ± ± ± Systemic ventricular mass/bsa (g/m 2 ) 49.7± ±7.2 < ± ± < Systemic ventricular ejection fraction (%) 57.4± ± < ± ± Systemic ventricular diastolic function* LA volume (ml) 40.2± ± ± ± E/A ratio 1.8 ( ) ( ) ( ) ( ) E deceleration time (ms) ( ) ( ) ( ) ( )

86 Uteroplacental blood flow in women with CHD. 85 Table 2. Comparison of women with CHD (n=209**) with healthy (n=70**) women during pregnancy. (continued) Gestational week 20 Gestational week 32 CHD n Healthy n P value CHD n Healthy n P value E/E 9.2 ( ) ( ) 62 < ( ) ( ) 55 < Right ventricular size and function* 39.0± ± < ± ± Right ventricular diastolic diameter (cm) TAPSE (mm) 22.5± ± < ± ± < Right ventricular S (cm/s) 9.7± ± < ± ± < BSA indicates body surface area; CHD, congenital heart disease; E, early passive filling velocity of systemic ventricular inflow; E, early diastolic tissue Doppler velocity of systemic ventricular annular ring; E/A ratio, early to atrial mitral inflow velocity ratio; Hb, serum hemoglobin; LA, left atrium; MAP, mean arterial pressure; NT-proBNP, N-terminal pro-b-type natriuretic peptide; NYHA, New York Heart Association functional class; Right ventricular S, systolic tissue Doppler velocity of tricuspid annular ring; and TAPSE, tricuspid annular plane systolic excursion.* Women with systemic right ventricle heart were only excluded from this specific analysis. LA = atrium receiving pulmonary venous flow; volume not measured in women with atrial correction of transposition and Fontan patients. ** Numbers different from n = 209 or n=70 are shown separately. 5

87 86 Chapter 5 Table 3. Associations of preconception and 20 week variables with umbilical artery RI 32 weeks entering the multivariable models. n # B 95% CI P Association of 20 weeks variables with umbilical artery RI 32 weeks entering model 1 Age at conception Parity Smoking during pregnancy High NT-proBNP* Systemic AV valve regurgitation LVEF (%) TAPSE (mm) Association of preconception variables with umbilical artery RI 32 weeks entering model 2 Disease complexity Simple (reference) Moderate complex Complex Age at conception Parity Pacemaker Sinus rhythm LVEF (%) TAPSE (mm) <0.001 Aortic stenosis (moderate /severe) Systemic AV valve regurgitation Pulmonary AV valve regurgitation Association of 20 weeks variables with umbilical artery RI 32 weeks entering model 2 Smoking during pregnancy High NT-proBNP* LVEF (%) For the full list of variables that were assessed in univariable analysis, see Methods. AV indicates atrioventricular; CI, confidence interval; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-b-type natriuretic peptide; RI, resistance index; and TAPSE, tricuspid annular plane systolic excursion. #Because the number of measurements of umbilical artery RI at 32 weeks is limited to 157, n cannot exceed this number. * High NT-proBNP: > 95 th percentile of healthy controls (>128 ng/l). Table 4. Multivariable regression analysis for the prediction of umbilical artery RI at 32 weeks of gestation. n # B 95% CI P Model 1: Association of 20 weeks variables with umbilical artery RI 32 weeks (degree of freedom = 128) Age at conception Smoking during pregnancy High NT-proBNP* Systemic AV valve regurgitation Model 2: Association of 20 weeks variables with umbilical artery RI 32 weeks after adjusting for preconception variables (degree of freedom = 99) Age Pulmonary AV valve regurgitation preconception TAPSE pre-conception Systemic AV valve regurgitation 20 weeks AV indicates atrioventricular; CI, confidence interval; NT-proBNP, N-terminal pro B-type natriuretic peptide; RI, resistance index; and TAPSE, tricuspid annular plane systolic excursion. # Because the number of measurements of umbilical artery RI at 32 weeks is limited to 157, n cannot exceed this number.* High NT-proBNP: > 95 th percentile of healthy controls (>128 ng/l).

88 Uteroplacental blood flow in women with CHD. 87 Pregnancy outcome in CHD and healthy women and relation of outcome to UDF Cardiovascular events occurred in 10.0% in the CHD and 0% in the healthy group. UDF parameters were not significantly associated with cardiovascular events. Obstetric events occurred in 58.9% of CHD and 32.9% of healthy women (p<0.005). CHD women had more planned caesarean sections (13.4% versus 1.4%, p=0.003) and assisted vaginal deliveries (47.4% versus 25.7%, p=0.001). The secondary caesarean delivery rate did not differ between both groups (10.0 versus 11.4%). Several obstetric events occurred more often in CHD women without the differences reaching statistical significance: hypertensive disorders of pregnancy (17.7% versus 11.4%), preeclampsia (5.7% versus 1.4%,) and preterm premature rupture of membranes (6.7 versus 2.9%). Postpartum hemorrhage occurred in both groups in 8.6%. In women with CHD, high umbilical artery RI (>90 th percentile of healthy group) at 32 weeks was associated with obstetric events (p=0.049). CHD women had shorter gestational age at delivery than healthy women (38.3 versus 39.7 weeks, p<0.005) and their babies had lower birth weight (3036 versus 3578 gram, p<0.005). More babies of CHD women had an Apgar score of <9 (8.7 versus 0%, p=0.009) 10 minutes after birth. Offspring events occurred more often in CHD women than in healthy women: 35.4% versus 18.6%, (p=0.008); offspring events excluding the small number of women with isolated CHD in the offspring: 34.4% versus 18.6, p= More children of women with CHD were small for gestational age (16.3% versus 4.3%, p=0.008). Congenital heart disease occurred in 4.8% of offspring of CHD women versus 0% of healthy women s offspring (p=0.176). Offspring death occurred in 2.9% of the CHD group and 0% of the healthy group. Causes of death were pregnancy termination because of spina bifida or complex heart disease in 2 pregnancies, intrauterine death in 2 patients because of hydrops fetalis and placental insufficiency, and post-partum death because of respiratory insufficiency in 2 pregnancies. Premature birth occurred in 12.4% versus 5.7% (p=0.18). UDF patterns of women with CHD were associated with offspring events (figure 2). This association was also significant when offspring CHD was excluded from the total number of offspring events. 5

89 88 Chapter 5 A B RR 2.9; 95% CI p = % 80 Offspring events (%) % 20 0 EDN Without EDN C Figure 2: Relation of uteroplacental Doppler flow parameters and offspring outcome. Uterine and umbilical artery PI and offspring events (A); Uterine and umbilical artery RI and offspring events (B) and uterine artery early diastolic notch at 20 weeks gestation and offspring events (C). Reported are the percentage of offspring events within, respectively, the groups uterine artery PI and RI at gestational week 20 and the groups umbilical artery PI and RI at gestational week 32, and within the groups with and without early diastolic notch. CI indicates confidence interval; EDN, early diastolic notch; OR, odds ratio; PI, pulsatility index; RI, resistance index; RR, relative risk; Uma, umbilical artery; and Uta, uterine artery.

90 Uteroplacental blood flow in women with CHD. 89 Discussion Our study is the first to compare UDF parameters of pregnant women with CHD and healthy pregnant women and relate these to cardiovascular parameters in pregnant women with CHD. Our data show that UDF and cardiovascular parameters differ between women with CHD and healthy women. In women with CHD, ventricular function, and valvular function, is related to UDF. As expected (because this is known in the general pregnant population), UDF is associated with obstetric and offspring events. Adequate uteroplacental blood flow is necessary for normal pregnancy outcome. Vascular remodeling of the uteroplacental circulation guarantees sufficient blood flow throughout pregnancy. This remodeling is characterized by vascular widening of the uterine circulation, which is mediated by endovascular trophoblast invasion of uterine spiral arteries, increased shear stress, and angiogenic and humoral factors. 18 The remodeling process results in a low resistance in the uteroplacental circulation. Abnormalities in the placentation process can result in elevated resistance and pulsatility indices, which are associated with adverse maternal and offspring outcome, particularly hypertensive disorders and intrauterine growth restriction. 5,19 In our study, women with CHD had significantly more obstetric and offspring complications than healthy women. This included a four-fold increase in the incidence of preeclampsia and of children born small for gestational age. The increased incidence of these complications in women with CHD is in line with previous studies. 1,3-5,20-22 The association of abnormal UDF patterns and obstetric and offspring outcome, which is well established in the general population, was also present in our women with CHD. More important, UDF indices indicated a higher resistance in the uteroplacental circulation throughout pregnancy in women with CHD than in healthy women. We demonstrated that UDF abnormalities in women with CHD were related to cardiac function, both before and during pregnancy. Cardiac parameters associated with UDF in the multivariable model included preconception right ventricular function but not left ventricular function. The likely explanation is the higher prevalence of RV dysfunction in our population. We used the TAPSE as a measure of right ventricular function, since it is a reproducible simple measurement that is associated with RV function and symptoms in patients with CHD. 23,24 Healthy pregnant women demonstrated a relatively high New York Heart Association functional class during pregnancy, reflecting the normal symptoms of pregnancy that can resemble heart failure. Functional class deteriorated more in women with CHD than in healthy women, which may indicate a less favorable adaptation of women with CHD to the hemodynamic changes of pregnancy. Not surprisingly, Nt-proBNP was higher throughout pregnancy in women with CHD compared to healthy women. Nt-proBNP decreased during pregnancy in both groups, as has been demonstrated previously in healthy women 25, and as can be explained by an increasing 5

91 90 Chapter 5 glomerular filtration rate during pregnancy. We found elevated Nt-proBNP to be weakly associated with abnormal UDF. Nt-proBNP and BNP are well-established biomarkers of heart failure, and BNP is a predictor of maternal cardiovascular pregnancy complications. 26 NtproBNP or BNP have not previously been investigated in relation to UDF in women with heart disease. Pre-pregnancy Nt-proBNP was unfortunately not available. Nt-proBNP may become a useful tool in pregnancy risk estimation in women with heart disease, but its role needs further investigation. Cardiac medication was related to uterine artery RI. The use of cardiac medication is also a predictor of maternal cardiac complications and is probably a marker of disease severity. 5 Most medications were beta-blockers, which are known to be associated with lower birth weight, which may be mediated by a negative effect on placental blood flow. Interestingly, both systemic and pulmonary atrioventricular valve regurgitation were associated with UDF parameters. Atrioventricular valve regurgitation is regarded as relatively harmless for the mother and her child, because the decrease in vascular resistance that accompanies pregnancy may reduce regurgitation. However, recent research indicates that mitral regurgitation does predict maternal cardiovascular complications and induces unfavorable cardiac remodeling. 5,27 A recent study demonstrated that mitral prolapse is associated with preterm delivery. 28 Therefore, atrioventricular valve regurgitation cannot be regarded as completely innocent. Our results indicate that placental flow may be compromised by atrioventricular valve regurgitation. This association may be caused by a direct hemodynamic effect or by a common developmental disorder. Valve stenosis did not predict UDF, which might be explained by a lower prevalence than regurgitant lesions. In addition to cardiac parameters, parity, age, and smoking were also associated with UDF. Our results support the hypothesis that pre-pregnancy cardiac dysfunction is related to UDF abnormalities, which are indicative of abnormal placentation. This finding is linked to the increased incidence of obstetric and offspring complications in women with CHD. Evidence from the literature indicates a relationship in the general population between previous hypertension during pregnancy, preeclampsia or intra-uterine growth restriction and the later occurrence of acquired cardiovascular disease in the mother. 29,30, 31 A recent study revealed an association of uterine artery RI during pregnancy with pre-pregnancy uterine artery blood flow. 32 Based on these data it has been hypothesized that pregnancy complications, particularly preeclampsia and intra-uterine growth restriction, reveal latent cardiovascular abnormalities that may be already present before pregnancy. Our study adds evidence to support this hypothesis, because, in our women with CHD, cardiac function pre-pregnancy related to abnormal UDF and adverse offspring outcome.

92 Uteroplacental blood flow in women with CHD. 91 Strengths and limitations Our study is the first to investigate UDF in pregnant women with cardiac disease. Several limitations must be considered. We designed our study to include pregnant women with various underlying congenital cardiac diseases. The heterogeneity of our population may have caused under-representation not only of individual diseases, but also of specific cardiac dysfunctions. This may have impacted the robustness of our prediction models. Moreover, because the study included women when they were already pregnant, collection of prepregnancy data was retrospective, and missing data were inevitable (mainly pre-pregnancy echocardiography data). Additionally, in this multicenter study, deviation from the protocol sometimes occurred, whereas complex disease often prevented accurate measurements of chamber size and function. Cardiac output could therefore not be measured reliably, and not all data were available in all patients. Technical limitations prevented the digital storage of UDF patterns, which were therefore measured by the different caregivers. Our composite outcome variable combined all offspring events. Because some offspring events (eg, CHD) may not be influenced by UDF or may be influenced through a different mechanism, we repeated the analysis without offspring CHD, which did not significantly alter outcome. We did not have data available on the course of intra-uterine growth and could not report on intra-uterine growth restriction. Therefore, we used small for gestational age as parameter of offspring growth. There might be some inclusion bias because we did not include patients from regional hospitals. Because the composition of our population is comparable with the Dutch national congenital database (CONCOR), this bias can be regarded unimportant. Because of the significant number of missing data from echocardiography in the preconception period, we chose to make a prediction model using data at 20 weeks gestation. Where possible, we assessed the influence of the known pre-pregnancy data. Despite these limitations, we were able to demonstrate that cardiac function in women with CHD is associated with an abnormal pattern of UDF and adverse pregnancy outcome. Our study results lead to an improved understanding of the pathophysiology of offspring events in women with CHD, and may also contribute to a better insight in the pathophysiology of offspring complications in the general population. 5 Sources of funding This work is supported by a grant from the Dutch Heart Foundation(2007B75). Dr van Veldhuisen is an established investigator of the Dutch Heart Foundation (D97-017). The Dutch Heart Foundation had no role in the design, data collection, analysis, interpretation, writing of the manuscript, or the decision to submit this manuscript for publication.

93 92 Chapter 5 Conflict of Interest Disclosures None.

94 Uteroplacental blood flow in women with CHD. 93 References 1. Siu SC, Sermer M, Colman JM, Alvarez AN, Mercier LA, Morton BC, Kells CM, Bergin ML, Kiess MC, Marcotte F, Taylor DA, Gordon EP, Spears JC, Tam JW, Amankwah KS, Smallhorn JF, Farine D, Sorensen S. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001; 104: Khairy P, Ouyang DW, Fernandes SM, Lee-Parritz A, Economy KE, Landzberg MJ. Pregnancy outcomes in women with congenital heart disease. Circulation 2006; 113: Stangl V, Schad J, Gossing G, Borges A, Baumann G, Stangl K. Maternal heart disease and pregnancy outcome: a single-centre experience. Eur J Heart Fail 2008; 10: Drenthen W, Pieper PG, Roos-Hesselink JW, van Lottum WA, Voors AA, Mulder BJ, Van Dijk AP, Vliegen HW, Yap SC, Moons P, Ebels T, van Veldhuisen DJ. Outcome of pregnancy in women with congenital heart disease: a literature review. J Am Coll Cardiol 2007; 49: Drenthen W, Boersma E, Balci A, Moons P, Roos-Hesselink JW, Mulder BJ, Vliegen HW, Van Dijk AP, Voors AA, Yap SC, van Veldhuisen DJ, Pieper PG. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010; 31: Aardema MW, Lander M, Oosterhof H, De Wolf BT, Aarnoudse JG. Doppler ultrasound screening predicts recurrence of poor pregnancy outcome in subsequent pregnancies, but not the recurrence of PIH or preeclampsia. Hypertens Pregnancy 2000; 19: Bamfo JE, Kametas NA, Chambers JB, Nicolaides KH. Maternal cardiac function in normotensive and pre-eclamptic intrauterine growth restriction. Ultrasound Obstet Gynecol 2008; 32: Vasapollo B, Valensise H, Novelli GP, Altomare F, Galante A, Arduini D. Abnormal maternal cardiac function precedes the clinical manifestation of fetal growth restriction. Ultrasound Obstet Gynecol 2004; 24: Yinon Y, Kingdom JC, Odutayo A, Moineddin R, Drewlo S, Lai V, Cherney DZ, Hladunewich MA. Vascular dysfunction in women with a history of preeclampsia and intrauterine growth restriction: insights into future vascular risk. Circulation 2010; 122: Balci A, Sollie KM, Mulder BJ, de Laat MW, Roos-Hesselink JW, Van Dijk AP, Wajon EM, Vliegen HW, Drenthen W, Hillege HL, Aarnoudse JG, van Veldhuisen DJ, Pieper PG. Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study. Am Heart J 2011; 161: Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, Chandrasekaran K, Solomon SD, Louie EK, Schiller NB. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010; 23: Vahanian A, Baumgartner H, Bax J, Butchart E, Dion R, Filippatos G, Flachskampf F, Hall R, Iung B, Kasprzak J, Nataf P, Tornos P, Torracca L, Wenink A. Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology. Eur Heart J 2007; 28: Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise J, Solomon S, Spencer KT, St John SM, Stewart W. Recommendations for chamber quantification. Eur J Echocardiogr 2006; 7:

95 94 Chapter Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A, Griffin BP, Iung B, Otto CM, Pellikka PA, Quinones M. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. J Am Soc Echocardiogr 2009; 22: Warnes CA, Liberthson R, Danielson GK, Dore A, Harris L, Hoffman JI, Somerville J, Williams RG, Webb GD. Task force 1: the changing profile of congenital heart disease in adult life. J Am Coll Cardiol 2001; 37: Regitz-Zagrosek V, Blomstrom LC, Borghi C, Cifkova R, Ferreira R, Foidart JM, Gibbs JS, Gohlke- Baerwolf C, Gorenek B, Iung B, Kirby M, Maas AH, Morais J, Nihoyannopoulos P, Pieper PG, Presbitero P, Roos-Hesselink JW, Schaufelberger M, Seeland U, Torracca L, Bax J, Auricchio A, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Knuuti J, Kolh P, McDonagh T, Moulin C, Poldermans D, Popescu BA, Reiner Z, Sechtem U, Sirnes PA, Torbicki A, Vahanian A, Windecker S, Baumgartner H, Deaton C, Aguiar C, Al-Attar N, Garcia AA, Antoniou A, Coman I, Elkayam U, Gomez-Sanchez MA, Gotcheva N, Hilfiker-Kleiner D, Kiss RG, Kitsiou A, Konings KT, Lip GY, Manolis A, Mebaaza A, Mintale I, Morice MC, Mulder BJ, Pasquet A, Price S, Priori SG, Salvador MJ, Shotan A, Silversides CK, Skouby SO, Stein JI, Tornos P, Vejlstrup N, Walker F, Warnes C. ESC Guidelines on the management of cardiovascular diseases during pregnancy: the Task Force on the Management of Cardiovascular Diseases during Pregnancy of the European Society of Cardiology (ESC). Eur Heart J 2011; 32: Prefumo F, Sharma R, Brecker SJ, Gaze DC, Collinson PO, Thilaganathan B. Maternal cardiac function in early pregnancies with high uterine artery resistance. Ultrasound Obstet Gynecol 2007; 29: Osol G, Mandala M. Maternal uterine vascular remodeling during pregnancy. Physiology (Bethesda) 2009; 24: : Madazli R, Somunkiran A, Calay Z, Ilvan S, Aksu MF. Histomorphology of the placenta and the placental bed of growth restricted foetuses and correlation with the Doppler velocimetries of the uterine and umbilical arteries. Placenta 2003; 24: Vriend JW, Drenthen W, Pieper PG, Roos-Hesselink JW, Zwinderman AH, van Veldhuisen DJ, Mulder BJ. Outcome of pregnancy in patients after repair of aortic coarctation. Eur Heart J 2005; 26: Drenthen W, Pieper PG, Roos-Hesselink JW, Schmidt AC, Mulder BJ, Van Dijk AP, Vliegen HW, Sollie KM, Voors AA, Ebels T, van Veldhuisen DJ. Non-cardiac complications during pregnancy in women with isolated congenital pulmonary valvar stenosis. Heart 2006; 92: Yap SC, Drenthen W, Meijboom FJ, Moons P, Mulder BJ, Vliegen HW, Van Dijk AP, Jaddoe VW, Steegers EA, Roos-Hesselink JW, Pieper PG. Comparison of pregnancy outcomes in women with repaired versus unrepaired atrial septal defect. BJOG 2009; 116: Koestenberger M, Nagel B, Avian A, Ravekes W, Sorantin E, Cvirn G, Beran E, Halb V, Gamillscheg A. Systolic right ventricular function in children and young adults with pulmonary artery hypertension secondary to congenital heart disease and tetralogy of Fallot: tricuspid annular plane systolic excursion (TAPSE) and magnetic resonance imaging data. Congenit Heart Dis 2012; 7: Koestenberger M, Nagel B, Ravekes W, Everett AD, Stueger HP, Heinzl B, Sorantin E, Cvirn G, Fritsch P, Gamillscheg A. Systolic right ventricular function in pediatric and adolescent patients with tetralogy of Fallot: echocardiography versus magnetic resonance imaging. J Am Soc Echocardiogr 2011; 24: Franz MB, Andreas M, Schiessl B, Zeisler H, Neubauer A, Kastl SP, Hess G, Rhomberg F, Zdunek D, Maurer G, Schlembach D, Heinze G, Szekeres T, Gottsauner-Wolf M. NT-proBNP is increased

96 Uteroplacental blood flow in women with CHD. 95 in healthy pregnancies compared to non-pregnant controls. Acta Obstet Gynecol Scand 2009; 88: Tanous D, Siu SC, Mason J, Greutmann M, Wald RM, Parker JD, Sermer M, Colman JM, Silversides CK. B-type natriuretic peptide in pregnant women with heart disease. J Am Coll Cardiol 2010; 56: Borges VT, Matsubara BB, Magalhaes CG, Peracoli JC, Rudge MV. Effect of physiological overload on pregnancy in women with mitral regurgitation. Clinics (Sao Paulo) 2011; 66: Chen CH, Huang MC, Liu HC, Huang CJ, Lin HC, Kou YR. Increased risk of preterm birth among women with mitral valve prolapse: a nationwide, population-based study. Ann Epidemiol 2011; 21: Bellamy L, Casas JP, Hingorani AD, Williams DJ. Pre-eclampsia and risk of cardiovascular disease and cancer in later life: systematic review and meta-analysis. BMJ 2007; 335: Bonamy AK, Parikh NI, Cnattingius S, Ludvigsson JF, Ingelsson E. Birth characteristics and subsequent risks of maternal cardiovascular disease: effects of gestational age and fetal growth. Circulation 2011; 124: Mannisto T, Mendola P, Vaarasmaki M, Jarvelin MR, Hartikainen AL, Pouta A, Suvanto E. Elevated blood pressure in pregnancy and subsequent chronic disease risk. Circulation 2013; 127: Hale SA, Schonberg A, Badger GJ, Bernstein IM. Relationship between prepregnancy and early pregnancy uterine blood flow and resistance index. Reprod Sci 2009; 16:

97

98 Chapter 6 Uteroplacental Doppler flow and pregnancy outcome in women with Tetralogy of Fallot Marlies A.M. Kampman Catharina M. Bilardo Dirk J. van Veldhuisen Ali Balci Martijn A. Oudijk Henk Groen Barbara J.M. Mulder Jolien W. Roos-Hesselink Gertjan Tj. Sieswerda Monique W.M. de Laat Krystyna M. Sollie-Szarynska Petronella G. Pieper on behalf of the ZAHARA investigators A Submitted.

99 98 Chapter 6 Abstract Background Pregnancy in women with Tetralogy of Fallot (ToF) is associated with cardiac, obstetric and offspring complications. We compared pregnancy outcome and uteroplacental Doppler flow (UDF) measurements in women with ToF and healthy women and aimed to assess whether in women with ToF a relationship exists between cardiac function and UDF. Methods We prospectively evaluated women with ToF and healthy women from the ZAHARA-studies. Clinical evaluation, standardized echocardiogram and UDF measurements were performed at 20 and 32 weeks gestation. Results We included 55 women with ToF and 69 healthy controls. Cardiac complications, mostly arrhythmias, occurred in 9.1% of ToF women. Offspring of ToF women was more growth restricted (20.0% vs. 4.4%,p=0.006), had low birth weight (16.4% vs. 4.9%, p=0.009) and congenital heart defects (5.6% vs. 0.0%, p=0.049) than the healthy controls. Abnormal UDF measurements at 32 weeks were more frequent in ToF women (more early diastolic notching of uterine artery flow (7.6% vs. 0%, p=0.020) and higher pulsatility index of umbilical artery flow (1.02 ± 0.20 vs ± 0.17, p=0.015)). Pre-pregnancy right and left ventricular function parameters and at 20 weeks gestation were significantly related with abnormal UDF. UDF parameters were associated with adverse offspring outcome. Conclusion The majority of women with surgically corrected ToF tolerate pregnancy well. However, UDF indices are more frequently abnormal in these women, suggesting an impaired placentation. Impaired right and left ventricular function parameters were associated with abnormal UDF parameters. UDF parameters were related to adverse obstetric and offspring outcome. Key words: congenital heart disease, pregnancy outcome, uteroplacental Doppler flow.

100 Uteroplacental blood flow in women with ToF. 99 Introduction Tetralogy of Fallot (ToF) is the most common cyanotic heart defect, accounting for approximately 10% of all congenital heart defects 1. After corrective surgery, the prognosis is excellent and the majority reaches reproductive age 2. Pregnancy in women with surgically corrected Tetralogy of Fallot is associated with an increased rate of maternal and offspring complications 3-5. Cardiac complication rates differ largely between studies (ranging between 0% and 17.5%), while obstetric and offspring complications are inconsistently described. Impaired uteroplacental circulation plays a central role in the pathogenesis of both obstetric and offspring complications in the general population 6. Previous studies on pregnancy in women with heart disease showed that maternal cardiac complications and offspring complications are related and they share similar predictors 7, 8. We therefore hypothesized that maternal cardiac dysfunction may be associated with placental dysfunction and thus with obstetric and offspring complications in women with congenital heart disease. Indeed, in the ZAHARA II study (Zwangerschap bij Aangeboren HARtAfwijkingen, pregnancy in congenital heart disease) population we reported abnormal uteroplacental Doppler flow parameters in women with congenital heart disease (CHD) compared to healthy pregnant women 9, 10. In women with CHD, an association was found between pre-pregnancy right ventricular function and high umbilical artery resistance indices (which are indicative of abnormal placentation), while the abnormal uteroplacental flow was associated with adverse offspring outcome 10. In this previous report we described a mixed population with various types of congenital heart disease. The aim of this prospective study was to investigate the prevalence of cardiac, obstetric and offspring complications in pregnant women with Tetralogy of Fallot and in healthy women. Subsequently, we compared uteroplacental Doppler flow patterns in pregnant women with ToF with healthy pregnant women and in order to further explore the association between uteroplacental circulation and cardiac function, we related cardiac function in women with ToF to uteroplacental Doppler flow patterns. 6 Methods The ZAHARA II (Zwangerschap bij Aangeboren HARtAfwijkingen, pregnancy in congenital heart disease) study is a prospective multicenter observational cohort study conducted between March 2008 and August For the current study, we prospectively extended the cohort of pregnant women with Tetralogy of Fallot included in the ZAHARA II study with consecutive pregnant women with Tetralogy of Fallot between October 2011 and October 2014 (ZAHARA III). Pregnant women with (surgically corrected) Tetralogy of Fallot, aged 18 years and a gestational age < 20 weeks gestation, presenting in one of the participat-

101 100 Chapter 6 ing centers were eligible for enrollment. The healthy controls included in ZAHARA II were recruited from low risk midwife practices in Groningen and Rotterdam in the Netherlands. The study design and primary results of ZAHARA II have been reported before 9, 10. Data of 40 women with ToF from ZAHARA II are incorporated in previously published reports The study protocol was approved by the Research Ethics Committee of the participating centers and all participating women gave written informed consent. The ZAHARA II study was supported by a grant from the Netherlands Heart foundation (2007B75), the ZAHARA III study was supported by a grant from ZonMW ( ). Baseline data and follow-up Pre-pregnancy baseline data were collected during the first ante-partum visit using medical records. Baseline data included maternal age, additional congenital heart defects, prior cardiovascular interventions, previous cardiac events, cardiac medication use, New York Heart Association (NYHA) functional class, modified WHO risk class before pregnancy 14, 12-lead ECG, laboratory results, echocardiographic recordings (including parameters concerning ventricular function (left ventricular ejection fraction, TAPSE, systolic tissue velocities (S )) and valve regurgitation or valve stenosis) smoking prior to pregnancy and obstetric history. All women visited the cardiac outpatient clinic at 20 and 32 weeks gestation for clinical evaluation (including NYHA class assessment) and standardized transthoracic echocardiogram. Uteroplacental Doppler flow measurements (pulsatility and resistance indices of the umbilical and left and right uterine artery and the presence of early diastolic notching) were performed at the prenatal care outpatient clinic at 20 and 32 weeks gestation. All transthoracic echocardiographic recordings were evaluated off line by four experienced cardiologists, blinded to the endpoints. Assessment of systolic and diastolic ventricular function, chamber quantification and valvular function were performed according to the current guidelines, as described before 9. Cardiac, obstetric and neonatal outcome Primary maternal cardiovascular events were recorded during pregnancy and up to 6 months post-partum and included any of the following: need for an urgent invasive cardiovascular procedure, heart failure (according to the guidelines of the European Society of Cardiology and documented by the attending physician 15 ), new onset or symptomatic tachy- or brady-arrhythmia requiring new or extended treatment, thrombo-embolic events, myocardial infarction, cardiac arrest, cardiac death, endocarditis and aortic dissection 9. Primary obstetric events were defined as: instrumental vaginal delivery (vacuum or forceps), cesarean section (planned or emergency), pregnancy induced hypertension (systolic blood pressure 140 mmhg or diastolic blood pressure 90 mmhg, in the absence of proteinuria, measured at two different moments with an interval of minimal 4 hours, occurring after 20 weeks gestation), pre-eclampsia (pregnancy induced hypertension with 0.3 gram/24

102 Uteroplacental blood flow in women with ToF. 101 hours proteinuria) 16, eclampsia (pre-eclampsia with grand mal seizures) 16, HELLP syndrome (haemolysis, elevated liver enzymes, low platelet syndrome), gestational Diabetes Mellitus (fasting blood glucose 6.1 mmol/l and after 75 gram oral glucose tolerance testing 7.8 mmol/l), non-cardiac death, placental abruption, post-partum hemorrhage, premature labor (spontaneous onset of labor < 37 weeks of gestation), and preterm premature rupture of membranes (spontaneous rupture of membranes before the onset of uterine contractions and < 37 weeks gestation) 9. Neonatal events were fetal death (still birth 20 weeks gestation), perinatal death (number of stillbirths from 20 weeks gestation plus deaths in the first 28 days of life), premature birth (spontaneous premature birth before 37 weeks gestation, occurring after premature labor or PPROM or indicated premature birth (induced for fetal or maternal reasons, such as pregnancy induced hypertension, pre-eclampsia or intra-uterine growth restriction), intraventricular hemorrhage (any grade) 17, neonatal respiratory distress syndrome (any grade), occurrence of congenital heart disease, small for gestational age (< 10 th birth weight percentile, adjusted for gestation age and based on population values in the Netherlands) and low birth weight (< 2500 grams) 9. The pulsatility index of the Uterine or Umbilical artery was considered abnormal if the value exceeded the 95 th percentile reference values according to gestational age in a low risk patient population 18, 19. A review of the literature was performed using the Pubmed Database, using the search terms congenital heart disease AND pregnancy. No limits or time restrictions for publication date were used. Relevant reports were selected after reading titles and abstracts. Eligible reports were retrieved and carefully reviewed. Maternal and offspring outcome were registered as defined above and tabulated using Microsoft Excel for Windows. Data from the literature were presented only for the sake of comparison with our current data and were not subjected to statistical testing. 6 Statistical analyses For continuous data, means and standard deviations (SD) or medians with interquartile range (IQR) were calculated, depending on their distribution. Absolute numbers and percentages were calculated for categorical data. The students T-test, Mann-Whitney U test, Chi-Square or Fishers exact test were used for intergroup comparison, as appropriate. Univariate linear regression analysis was used to assess associations between cardiac function and uteroplacental Doppler flow parameters. Multivariate linear regression was omitted since missing data from pre-pregnancy echocardiograms resulted in a patient cohort too small to perform a meaningful analysis. The following predefined pre-pregnancy parameters were assessed: maternal age, NYHA functional class, WHO pregnancy risk class, use of cardiac medication, right ventricular diastolic diameter, right ventricular function parameters (TAPSE and right ventricular systolic tis-

103 102 Chapter 6 sue velocity (S )), left ventricular ejection fraction, mean left ventricular systolic tissue velocity (S ) (septal-lateral), mean left ventricular early diastolic tissue velocity (E ) (septal-lateral), the presence of valve regurgitation/stenosis. The following variables were assessed at 20 weeks gestation: high Nt-proBNP (>128 pg/ml), right ventricular function parameters (TAPSE and right ventricular systolic tissue velocity (S )), left ventricular ejection fraction. Statistical analysis was performed using STATA software package (version 11, college station, Texas, USA). A two-tailed p-value < 0.05 was considered significant. Results A total of 55 pregnant women with Tetralogy of Fallot (ToF) were eligible for inclusion (72.0% from the primary ZAHARA II cohort). A total of 71 healthy pregnant women were selected in the participating midwives practices. Two women were excluded (one with a previously undetected atrial septal defect type II, one was lost to follow up), resulting in 69 healthy controls. Baseline and pregnancy outcome Table 1 provides the baseline characteristics of the women with ToF and healthy pregnant women. The two groups did not differ for mean age and parity. The majority was nulliparous. All ToF women had undergone complete repair and moderate or severe pulmonary valve regurgitation was present in 61.5%, while 20.0% had moderate or severe pulmonary valve stenosis. Right ventricular dysfunction was present in 22.5% and left ventricular dysfunction in 9.3%. Pregnancy complications in women with ToF and in healthy controls are shown in table 2 and figure 1. No maternal deaths occurred. Cardiac complications were seen in 9.1% of the ToF pregnancies, mostly supraventricular arrhythmias. No statistical significant differences in obstetric complications occurred, except for more deliveries per planned caesarean section in women with ToF. Gestational age at delivery in women with ToF and in healthy controls was 38.4 ± 3.0 and 39.8 ± 1.5 weeks (p=0.002), respectively. Results were comparable after excluding women with a planned caesarean section. There was a not significant trend for higher perinatal mortality rates in offspring of women with ToF. In one case pregnancy was terminated because of severe CHD in the fetus. One child died due to prematurity-related complications after birth at 27 weeks following premature rupture of the membranes at 20 weeks gestation. Offspring of women with ToF had significantly more low birth weight and were more often small for gestational age (table 2). This difference persisted after exclusion of women treated with beta-blockers. In addition, congenital heart disease occurred significantly more often in offspring of women with ToF (1 ventricular septal defect, 2 Tetralogy of Fallot).

104 Uteroplacental blood flow in women with ToF. 103 Table 1. Baseline characteristics for women with Tetralogy of Fallot (N=55) and healthy women (N=69). Women with ToF (N (%)) Healthy women (N (%)) P-value Demographic data Maternal age (years ± SD) 31.3 ± ± Parity status (60.0) 44 (63.8) 1 16 (29.1) 17 (24.6) 2 6 (10.9) 8 (11.6) Smoking prior to pregnancy 13 (23.6) 23 (33.3) 0.23 NYHA class I 38 (69.1) II 17 (30.9) Modified WHO class* II 53 (96.4) III 2 (3.6) Complete repair Infundibular resection ± valvulotomy 17 (30.9) RV outflowtract patch ± valvulotomy 17 (30.9) Transannular patch ± valvulotomy 19 (34.5) Pulmonary valve replacement 17 (30.9) Biological valve prosthesis 16 (29.1) Mechanical valve prosthesis 1 (1.8) Past medical history 6 History of congestive heart failure 1 (1.8) History of arrhythmia (requiring treatment) 1 (1.8) Pacemaker 3 (5.5) Medication prior to pregnancy 6 (10.9) Betablocker 5 (9.1) Echocardiographic parameters Tricuspid valve regurgitation 2 (3.9) Pulmonary valve stenosis 10 (20.0) Pulmonary valve regurgitation 32 (61.5) Left ventricular systolic dysfunction 4 (9.30) Right ventricular systolic dysfunction 11 (22.5) *Modified World Health Organization class according to ESC guidelines 14. Missing data excluded from analysis; Moderate or severe regurgitation; peak gradient 36 mmhg; Ejection fraction < 45%; TAPSE < 16 mm.

105 104 Chapter 6 Table 2. Pregnancy outcome in women with Tetralogy of Fallot and healthy women. Literature (%) Women with ToF (N=55 (%)) Healthy women (N=69(%)) P-value* Cardiac events 40/562 (7.1) pregnancies 5 (9.1) 0 (0.0) Heart failure 14 (2.5) 1 (1.8) 0 (0.0) 0.26 Arrhythmia 19 (3.4) 3 (5.5) 0 (0.0) Trombo-embolic event 2 (0.36) 1 (1.8) 0 (0.0) 0.26 Obstetric events 82/462 (18.8) pregnancies 33 (60.0) 30 (43.5) Instrumental vaginal delivery not reported 22 (40.0) 24 (34.8) 0.55 (vacuum or forceps) Cesarean section 68/323 (21.1%) 9 (16.4) 9 (13.1) Planned cesarean section Inconsistently 5 (9.1) 1 (1.5) reported Emergency cesarean section Inconsistently 4 (7.3) 8 (11.6) 0.42 reported Gestational diabetes Mellitus 1 (0.2) 2 (3.6) 0 (0.0) 0.11 HELLP syndrome not reported 1 (1.8) 0 (0.0) 0.26 Post-partum haemorrhage 29 (6.3) 2 (3.6) 4 (5.8) 0.58 Pre-eclampsia 8 (1.7) 4 (7.3) 1 (1.5) 0.10 Pregnancy induced hypertension 17 (3.7) 5 (9.1) 7 (10.1) 0.84 Premature labour 17 (3.7) 4 (7.3) 3 (4.4) 0.48 Preterm premature rupture of 9 (1.9) 4 (7.3) 2 (2.9) 0.26 membranes Offspring events 109/515 (21.1) 18 (32.7) 8 (11.6) pregnancies Fetal death not reported 1 (1.8) 0 (0.0) 0.26 Intra-ventricular hemorrhage not reported 0 (0.0) 1 (1.5) 0.37 Low birth weight (<2500 grams) not reported 9 (16.4) 2 (2.9) Perinatal death 9 (1.7) 2 (3.6) 0 (0.0) 0.11 Neonatal respiratory distress not reported 4 (7.3) 1 (1.5) 0.10 syndrome Occurence of CHD 18 (3.5) 3 (5.6) 0 (0.0) Premature birth 38 (7.4) 6 (10.9) 4 (5.8) 0.49 Spontaneous Inconsistently 4 (7.3) 2 (2.9) 0.25 reported Indicated Inconsistently 2 (3.6) 2 (2.9) 0.80 reported Small for gestational age (< 10th percentile) 45 (8.7) 11 (20.0) 3 (4.4) * Women with Tetralogy of Fallot in this study compared to healthy controls. Total number incorporates only the reported complications

106 Uteroplacental blood flow in women with ToF. 105 Figure 1: Number of cardiovascular events (A), obstetric events (B) and offspring events (C) in patients with Tetralogy of Fallot after literature review (black), current study cohort of women with ToF (light grey) and healthy controls in ZAHARA II (dark grey). * p < 0.05 compared to healthy controls in ZAHARA II. Total number of pregnancies described in the literature were 562 (A), 462 (B) and 515 (C) respectively. For comparison purposes, total represents the total percentage of complications mentioned in the figure only. abruptio, abruptio placentae; CHD, occurrence of congenital heart disease; GDM, Gestational Diabetes Mellitus; PB, premature birth; PE, preeclampsia; PIH, pregnancy induced hypertension; PL, premature labour; PPH, post-partum hemorrhage; PPROM, preterm premature rupture of membranes; SGA, small for gestational age. 6 Uteroplacental Doppler flow measurements At 32 weeks gestation, the pulsatility index (PI) of the umbilical artery was significantly higher in fetuses of women with ToF than in the healthy controls (1.02 ± 0.20 vs ± 0.17,p=0.015, table 3). The incidence of abnormal PI of the umbilical artery at 32 weeks did not significantly differ between both groups (6.4% vs. 3.1%,p=0.40) though, and it was not associated with offspring outcome. No differences were found between women with ToF and healthy women for pulsatility index (PI) of the uterine artery at 20 and 32 weeks gestation or in the resistance index (RI) of the uterine artery at 20 weeks and 32 weeks gestation (table 3). There was no difference in the percentage of abnormal uterine artery PI at 20 and 32 weeks of gestation between women with ToF and healthy controls(4.6% vs. 3.3%(p=0.74) and 12.8% vs. 6.2%(p=0.24), respectively). Results were comparable after

107 106 Chapter 6 Table 3. Uteroplacental Doppler flow measurements in women with Tetralogy of Fallot and healthy women. Women with ToF Healthy women P-value Measurements at 20 weeks gestation Uterine artery pulsatility index 0.93 ± ± Uterine artery resistance index 0.55 ± ± Presence of early diastolic notch (N(%)) 4 (7.4) 1 (1.5) Umbilical artery pulsatility index 1.17 ± ± Measurements at 32 weeks gestation Uterine artery pulsatility index 0.75 ± ± Uterine artery resistance index 0.49 ± ± Presence of early diastolic notch (N(%)) 4 (7.6) 0 (0.0) Umbilical artery pulsatility index 1.02 ± ± exclusion of pregnancies with a CHD in the fetus. The only difference was that women with ToF had more often a persisting early diastolic notch (EDN) in the uterine artery at 32 weeks (7.6% vs. 0%,p=0.020). In women with ToF, EDN in the uterine artery waveform at 20 weeks was associated with low birth weight(2.34 ( ),p=0.050), whereas RI was associated with Umbilical artery PI at 32 weeks of gestation TAPSE pre-pregnancy Umbilical artery PI at 32 weeks of gestation Umbilical artery PI at 32 weeks of gestation A Right ventricular systolic tissue velocity pre-pregnancy (cm/s) B Umbilical artery PI at 32 weeks of gestation Right ventricular systolic tissue velocity at 20 weeks of gestation (cm/s) C Left ventricular ejection fraction at 20 weeks of gestation (%) D Figure 2: Scatterplot of pulsatility index (PI) of the umbilical artery at 32 weeks gestation and TAPSE prepregnancy (A), right ventricular systolic tissue velocity pre-pregnancy (B), right ventricular systolic tissue velocity at 20 weeks gestation (C) and left ventricular ejection fraction at 20 weeks gestation (D), respectively.

108 Uteroplacental blood flow in women with ToF. 107 obstetric complications( 7.73 ( ),p=0.032). No associations were found between uterine artery PI and RI and offspring outcome. Similarly to our previous report on uteroplacental flow and cardiac function we chose for the regression analysis to report on uterine artery RI at 20 weeks gestation 10. Results were comparable for uterine artery RI and PI. For the umbilical artery we used PI, since this was available in 85% of all patients at 32 weeks gestation. Results of univariable regression are displayed in table 4. Right ventricular function parameters pre-pregnancy and at 20 weeks gestation, as well as left ventricular ejection fraction at 20 weeks gestation, showed a correlation with uteroplacental Doppler flow parameters (figure 2). Table 4. Univariable analysis of uteroplacental doppler flow parameters N Beta (95% CI) P-value Resistance index Uterine artery 20 weeks Pre-pregnancy variables Modified WHO risk class (reference: WHO II) ( ) NYHA functional class (reference: NYHA I) ( ) 0.18 Right ventricular systolic tissue velocity (S ) ( ) 0.81 Right ventricular function (TAPSE) ( ) Left ventricular ejection fraction (%) ( ) 0.29 Pulmonary valve regurgitation ( ) Pulmonary valve stenosis ( ) Tricuspid valve regurgitation ( ) Variables at 20 weeks gestation Right ventricular systolic tissue velocity (S ) ( ) Right ventricular function (TAPSE) ( ) 0.73 Left ventricular ejection fraction (%) ( ) 0.94 NTproBNP > 128 pg/ml ( ) 0.68 Pulsatility index Umbilical artery 32 weeks Pre-pregnancy variables Modified WHO risk class ( ) 0.79 NYHA functional class ( ) 0.95 Right ventricular systolic tissue velocity (S ) ( ) Right ventricular function (TAPSE) ( ) Left ventricular ejection fraction (%) ( ) 0.55 Pulmonary valve regurgitation ( ) 0.45 Pulmonary valve stenosis ( ) 0.10 Tricuspid valve regurgitation ( ) 0.75 Variables at 20 weeks gestation Right ventricular systolic tissue velocity (S ) ( ) Right ventricular function (TAPSE) ( ) 0.50 Left ventricular ejection fraction ( ) NTproBNP > 128 pg/ml ( )

109 108 Chapter 6 Discussion This is the first study comparing pregnancy outcome and uteroplacental Doppler flow parameters of women with ToF and healthy controls. We found that women with ToF experience more cardiovascular complications (in particular arrhythmias) and more offspring complications, in particular low birth weight, small for gestational age and congenital heart disease in the fetus. Women with ToF had more often abnormal UDF indices, indicative of defective placentation. Right and left ventricular function parameters were found to be associated with UDF parameters. UDF parameters were associated with obstetric and offspring outcomes. As can be seen in figure 1 and table 2, the cardiac complication rate in our study was slightly higher than the overall complication rate found in the literature (7.1% cardiac complications during 562 pregnancies in women with ToF) 3-5, 8, 20-30, with more arrhythmia and less heart failure, but rates do differ largely between studies. Differences in study design and patient cohort probably account for these differences. The obstetric complication rate was not significantly different in women with ToF than in healthy women, although a trend towards a higher prevalence of hypertensive disorders of pregnancy (PIH/PE/HELLP) and premature birth was observed in women with ToF. The obstetric and offspring complication rate in this study is higher than the incidence found in the literature, but numbers differ largely between studies owing to differences in definitions and, more importantly, underreporting. Hypertensive disorders of pregnancy, growth restriction and premature birth, the most frequently observed pregnancy complications in this study, are associated with abnormal placentation 6, reflected by elevated uteroplacental resistance 31, 32. This means that pregnancy complications in women with ToF are associated with abnormal placentation. In this study we found an association between right ventricular parameters (lower TAPSE pre-pregnancy and lower S both pre-pregnancy and at 20 weeks gestation) and impaired uteroplacental Doppler flow parameters. These findings are in line with our previous study in a larger population with mixed types of CHD, where lower TAPSE pre-pregnancy was associated with increased umbilical artery resistance index 10. Interestingly, in this population with ToF not only right ventricular function, but also impaired left ventricular ejection fraction at 20 weeks was associated with increased umbilical artery PI at 32 weeks gestation. There is increasing evidence of a relationship between cardiac function and poor placentation and a role for right ventricular dysfunction in the patho-physiological mechanism has been suggested 33. A recent systematic review of our group suggests a link between preexistent subclinical cardiac dysfunction and poor placentation, reflected by high resistance in the uteroplacental circulation 33. There is also evidence that previously healthy women with high uteroplacental resistance and poor pregnancy outcome have an increased prevalence of systolic and diastolic left ventricular dysfunction. Additionally, right ventricular systolic and diastolic dysfunction has been described in healthy women with early pre-eclampsia 34.

110 Uteroplacental blood flow in women with ToF. 109 Melchiorre and colleagues described a higher prevalence of previously unknown functionally significant cardiac defects in women with increased uterine artery Doppler indices 35. Interestingly, most of these cardiac defects had right-sided sequelae. The underlying pathophysiological mechanism responsible for these observations might be analogous to what is known in patients with congestive heart failure, where dysfunction of other organs, including kidneys and liver, is related to left ventricular systolic dysfunction but also to elevated central venous pressure 36, 37. The placenta may be regarded as a temporary extra organ and may be affected by cardiac dysfunction in a similar way. The normal placental bed arteries are low resistance vessels due to remodeling of the spiral artery walls by trophoblast invasion, leaving the auto-regulatory capacity of the uteroplacental circulation limited 38, 39. Therefore, the uteroplacental circulation is directly dependent on maternal cardiac performance. Left ventricular dysfunction affects cardiac output and underperfusion may influence placental development directly, without acting through an abnormal throphoblastic invasion. Right ventricular dysfunction influences left ventricular function through ventriculo-ventricular interdependence, while right ventricular dysfunction can also lead to increased central venous pressure and subsequent venous congestion. Gyselaers et al. postulated that venous hemodynamic dysfunction (i.e. increased venular pressure) may lead to increased intervillous space pressure, causing damage to the trophoblast and resulting in defective trophoblastic remodeling of spiral arteries 40. Especially the second wave of trophoblastic invasion of the spiral arteries, occurring when intervillous space circulation is established, may be hampered. This may explain the finding of increased resistance in the uterine artery waveforms of these patients at 32 weeks gestation. This hypothesis might be an essential part of the explanation why ventricular function parameters are increasingly found to be associated with altered uteroplacental Doppler flow indices. Further experimental and clinical research with histological examination of the placenta to explore this association is warranted. 6 Strengths & Limitations This is the first study that investigates uteroplacental circulation during pregnancy in women with ToF and compares pregnancy outcome in these women with healthy controls. In addition, the prospective nature of the study makes the results more valuable. Because of the design of the study, pre-pregnancy data were collected retrospectively and missing data were inevitable. In addition, the study population is relatively small. As a consequence, multivariate linear regression analysis of the uteroplacental flow indices was not feasible and some of the observed differences in obstetric and offspring outcomes between women with ToF and controls might have not reached statistical significance. A formal power analysis was not performed since the current study concerns a secondary analysis. These limitations should be kept in mind and the results interpreted with caution.

111 110 Chapter 6 Conclusion Most women with surgically corrected ToF tolerate pregnancy well. However, cardiovascular complications are common, mainly arrhythmias, and offspring complications are more frequently observed in women with ToF compared to healthy controls. Early diastolic notching of the uterine artery waveform occurs more often in women with ToF, which is indicative of disturbed placentation. In addition, umbilical artery pulsatility index is higher at 32 weeks gestation. Right and left ventricular function parameters pre-pregnancy and at 20 weeks gestation are negatively associated with resistance in the uteroplacental circulation. The findings of this study add more evidence that cardiac dysfunction might play an important role in the placentation process and the occurrence of placenta-related complications in women with congenital heart disease. More fundamental research is required for better understanding of the underlying pathophysiological mechanism.

112 Uteroplacental blood flow in women with ToF. 111 References 1. Bashore TM. Adult congenital heart disease: Right ventricular outflow tract lesions. Circulation Apr 10; 115(14): Cuypers JA, Menting ME, Konings EE, Opic P, Utens EM, Helbing WA, et al. Unnatural history of tetralogy of fallot: Prospective follow-up of 40 years after surgical correction. Circulation Nov 25; 130(22): Balci A, Drenthen W, Mulder BJ, Roos-Hesselink JW, Voors AA, Vliegen HW, et al. Pregnancy in women with corrected tetralogy of fallot: Occurrence and predictors of adverse events. Am Heart J Feb; 161(2): Khairy P, Ouyang DW, Fernandes SM, Lee-Parritz A, Economy KE, Landzberg MJ. Pregnancy outcomes in women with congenital heart disease. Circulation Jan 31; 113(4): Veldtman GR, Connolly HM, Grogan M, Ammash NM, Warnes CA. Outcomes of pregnancy in women with tetralogy of fallot. J Am Coll Cardiol Jul 7; 44(1): Brosens I, Pijnenborg R, Vercruysse L, Romero R. The great obstetrical syndromes are associated with disorders of deep placentation. Am J Obstet Gynecol Mar; 204(3): Drenthen W, Boersma E, Balci A, Moons P, Roos-Hesselink JW, Mulder BJ, et al. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J Sep; 31(17): Siu SC, Sermer M, Colman JM, Alvarez AN, Mercier LA, Morton BC, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation Jul 31; 104(5): Balci A, Sollie KM, Mulder BJ, de Laat MW, Roos-Hesselink JW, van Dijk AP, et al. Associations between cardiovascular parameters and uteroplacental doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the zwangerschap bij aangeboren hartafwijking (ZAHARA) II study. Am Heart J Feb; 161(2): 269,275.e Pieper PG, Balci A, Aarnoudse JG, Kampman MA, Sollie KM, Groen H, et al. Uteroplacental blood flow, cardiac function, and pregnancy outcome in women with congenital heart disease. Circulation Dec 3; 128(23): Kampman MA, Balci A, van Veldhuisen DJ, van Dijk AP, Roos-Hesselink JW, Sollie-Szarynska KM, et al. N-terminal pro-b-type natriuretic peptide predicts cardiovascular complications in pregnant women with congenital heart disease. Eur Heart J Mar; 35(11): Kampman MA, Balci A, Groen H, van Dijk AP, Roos-Hesselink JW, van Melle JP, et al. Cardiac function and cardiac events 1-year postpartum in women with congenital heart disease. Am Heart J Feb; 169(2): Balci A, Sollie-Szarynska KM, van der Bijl AG, Ruys TP, Mulder BJ, Roos-Hesselink JW, et al. Prospective validation and assessment of cardiovascular and offspring risk models for pregnant women with congenital heart disease. Heart Sep; 100(17): European Society of Gynecology, Association for European Paediatric Cardiology, German Society for Gender Medicine, Authors/Task Force Members, Regitz-Zagrosek V, Blomstrom Lundqvist C, et al. ESC guidelines on the management of cardiovascular diseases during pregnancy: The task force on the management of cardiovascular diseases during pregnancy of the european society of cardiology (ESC). Eur Heart J Dec; 32(24): McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The task force for the 6

113 112 Chapter 6 diagnosis and treatment of acute and chronic heart failure 2012 of the european society of cardiology. developed in collaboration with the heart failure association (HFA) of the ESC. Eur Heart J Jul; 33(14): ACOG Committee on Practice Bulletins--Obstetrics. ACOG practice bulletin. diagnosis and management of preeclampsia and eclampsia. number 33, january Obstet Gynecol Jan; 99(1): Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: A study of infants with birth weights less than 1,500 gm. J Pediatr Apr; 92(4): Gomez O, Figueras F, Fernandez S, Bennasar M, Martinez JM, Puerto B, et al. Reference ranges for uterine artery mean pulsatility index at weeks of gestation. Ultrasound Obstet Gynecol Aug; 32(2): Acharya G, Wilsgaard T, Berntsen GK, Maltau JM, Kiserud T. Reference ranges for serial measurements of umbilical artery doppler indices in the second half of pregnancy. Am J Obstet Gynecol Mar; 192(3): Aggarwal N, Suri V, Kaur H, Chopra S, Rohila M, Vijayvergiya R. Retrospective analysis of outcome of pregnancy in women with congenital heart disease: Single-centre experience from north india. Aust N Z J Obstet Gynaecol Aug; 49(4): Egidy Assenza G, Cassater D, Landzberg M, Geva T, Schreier J, Graham D, et al. The effects of pregnancy on right ventricular remodeling in women with repaired tetralogy of fallot. Int J Cardiol Oct 3; 168(3): Daliento L, Dal Bianco L, Bagato F, Secco E, Sarubbi B, Mazzotti E, et al. Gender differences and role of pregnancy in the history of post-surgical women affected by tetralogy of fallot. PLoS One. 2012; 7(12): e Gelson E, Gatzoulis M, Steer PJ, Lupton M, Johnson M. Tetralogy of fallot: Maternal and neonatal outcomes. BJOG Feb; 115(3): Hidaka Y, Akagi T, Himeno W, Ishii M, Matsuishi T. Left ventricular performance during pregnancy in patients with repaired tetralogy of fallot: Prospective evaluation using the tei index. Circ J Aug; 67(8): Kamiya CA, Iwamiya T, Neki R, Katsuragi S, Kawasaki K, Miyoshi T, et al. Outcome of pregnancy and effects on the right heart in women with repaired tetralogy of fallot. Circ J. 2012; 76(4): Niwa K, Tateno S, Akagi T, Himeno W, Kawasoe Y, Tatebe S, et al. Arrhythmia and reduced heart rate variability during pregnancy in women with congenital heart disease and previous reparative surgery. Int J Cardiol Nov 15; 122(2): Pedersen LM, Pedersen TA, Ravn HB, Hjortdal VE. Outcomes of pregnancy in women with tetralogy of fallot. Cardiol Young Aug; 18(4): Singh H, Bolton PJ, Oakley CM. Pregnancy after surgical correction of tetralogy of fallot. Br Med J (Clin Res Ed) Jul 17; 285(6336): Uebing A, Arvanitis P, Li W, Diller GP, Babu-Narayan SV, Okonko D, et al. Effect of pregnancy on clinical status and ventricular function in women with heart disease. Int J Cardiol Feb 18; 139(1): Zuber M, Gautschi N, Oechslin E, Widmer V, Kiowski W, Jenni R. Outcome of pregnancy in women with congenital shunt lesions. Heart Mar; 81(3):

114 Uteroplacental blood flow in women with ToF Harrington K, Cooper D, Lees C, Hecher K, Campbell S. Doppler ultrasound of the uterine arteries: The importance of bilateral notching in the prediction of pre-eclampsia, placental abruption or delivery of a small-for-gestational-age baby. Ultrasound Obstet Gynecol Mar; 7(3): Cnossen JS, Morris RK, ter Riet G, Mol BW, van der Post JA, Coomarasamy A, et al. Use of uterine artery doppler ultrasonography to predict pre-eclampsia and intrauterine growth restriction: A systematic review and bivariable meta-analysis. CMAJ Mar 11; 178(6): Kampman MA, Bilardo CM, Mulder BJ, Aarnoudse JG, Ris-Stalpers C, van Veldhuisen DJ, et al. Maternal cardiac function, uteroplacental doppler flow parameters and pregnancy outcome: A systematic review. Ultrasound Obstet Gynecol Oct Melchiorre K, Sutherland GR, Watt-Coote I, Liberati M, Thilaganathan B. Severe myocardial impairment and chamber dysfunction in preterm preeclampsia. Hypertens Pregnancy. 2012; 31(4): Melchiorre K, Sutherland GR, Liberati M, Bhide A, Thilaganathan B. Prevalence of maternal cardiac defects in women with high-resistance uterine artery doppler indices. Ultrasound Obstet Gynecol Mar; 37(3): van Deursen VM, Damman K, Hillege HL, van Beek AP, van Veldhuisen DJ, Voors AA. Abnormal liver function in relation to hemodynamic profile in heart failure patients. J Card Fail Jan; 16(1): Damman K, van Deursen VM, Navis G, Voors AA, van Veldhuisen DJ, Hillege HL. Increased central venous pressure is associated with impaired renal function and mortality in a broad spectrum of patients with cardiovascular disease. J Am Coll Cardiol Feb 17; 53(7): Aardema MW, Oosterhof H, Timmer A, van Rooy I, Aarnoudse JG. Uterine artery doppler flow and uteroplacental vascular pathology in normal pregnancies and pregnancies complicated by pre-eclampsia and small for gestational age fetuses. Placenta May; 22(5): Pijnenborg R, Anthony J, Davey DA, Rees A, Tiltman A, Vercruysse L, et al. Placental bed spiral arteries in the hypertensive disorders of pregnancy. Br J Obstet Gynaecol Jul; 98(7): Gyselaers W, Peeters L. Physiological implications of arteriovenous anastomoses and venous hemodynamic dysfunction in early gestational uterine circulation: A review. J Matern Fetal Neonatal Med Jun; 26(9):

115

116 Chapter 7 Maternal cardiac function, uteroplacental Doppler flow parameters and pregnancy outcome: a systematic review Marlies A.M. Kampman Catharina M. Bilardo Barbara J. M. Mulder Jan G. Aarnoudse Carrie Ris-Stalpers Dirk J. van Veldhuisen Petronella G. Pieper 8 A Ultrasound in Obstetrics and Gynecology 2015; 46:

117 116 Chapter 7 Abstract Objective To investigate the existing evidence for a link between maternal cardiac function, abnormal utero-placental flow and poor perinatal outcome in women with and without known cardiac disease. Methods Pubmed and Embase databases were searched systematically for studies relating cardiac functional parameters and uteroplacental Doppler flow with pregnancy outcome in women with pre-existent congenital cardiac disease and women without known cardiac disease. Only studies based on echocardiography were included. Results From 1732 citations, 10 articles were included. In women with known congenital heart disease, a relationship was found between abnormal utero-placental Doppler flow patterns and cardiac function before and during pregnancy. Conversely, women without a history of congenital heart disease, but with abnormal uterine artery resistance and pregnancy complications, more often show global left ventricular diastolic dysfunction (33%; p=0.0001), impaired myocardial relaxation (72%; p<0.0001) and left ventricular systolic dysfunction (17%; p=0.006), even up to 1 year post-partum. Conclusion There is increasing evidence for an association between pre-existent subclinical cardiac dysfunction, poor placentation (reflected by uteroplacental Doppler flow abnormalities) and poor pregnancy outcome. It may be postulated that pre-existent suboptimal cardiac performance, either as a result of either congenital heart disease or a subclinical latent condition, is one of the common denominators of poor placentation, leading to poor pregnancy outcome. Key words: Maternal cardiac function; Congenital heart disease; Pregnancy; Uteroplacental Doppler flow.

118 Maternal cardiac function and uteroplacental Doppler flow: systematic review 117 Introduction In pregnant women without underlying heart disease, inadequate adaptation of the uteroplacental circulation to pregnancy is responsible for maternal and fetal complications. The pathophysiological mechanism is poor trophoblast invasion of the spiral arteries during the placentation process 1, causing failure of the placental-bed arteries to transform from high- to low resistance vessels. Uteroplacental Doppler flow investigations (measurement of resistance index, pulsatility index and bilateral diastolic notching) provide insight into the placentation process and are commonly used as screening tools to predict the future development of hypertensive disease of pregnancy, (early) pre-eclampsia (PE), fetal growth restriction (FGR), stillbirth and placental abruption 2,3. Women with heart disease experience more pregnancy-related complications than do healthy pregnant women, specifically small for gestational age (SGA) fetuses and hypertensive disorders of pregnancy 4-7. The ZAHARA and CARPREG studies demonstrated that in women with congenital heart disease, these pregnancy complications are related to maternal cardiac function, but the underlying pathophysiology remained unclear 4,8. In patients with chronic congestive heart failure, worsening cardiac function is associated with dysfunction of other organs A similar situation may be present during pregnancy, during which cardiac dysfunction may influence placental development and function. Recently, the ZAHARA II study indicated that in women with congenital heart defects abnormal placentation, as suggested by high uterine artery resistance indexes, may be caused by abnormal pre-pregnancy cardiovascular function 13,14. The aim of this systematic review was to investigate the existing evidence for a link between maternal cardiac function, abnormal utero-placental flow and poor perinatal outcome, in women with and without known cardiac disease. Methods 7 Search strategy A systematic literature search was performed, using PubMed and Embase databases on 11 November 2013, using the following search terms: Mother (MeSH OR all fields) OR maternal (all fields) AND echocardiography (MeSH OR all fields) AND pregnancy (MeSH OR all fields). The search was limited to original papers of human studies and English or German language, in order to reduce misinterpretation. No time restrictions for publication date were used. The search was performed by one author (M.K.). Titles and abstracts of the identified articles were carefully read by one author (M.K.), and reports that were obviously irrelevant were excluded. Eligible full text reports were retrieved, carefully reviewed and included if all of the following predefined inclusion criteria were met:

119 118 Chapter 7 (1) study population consisted of pregnant women with known heart disease or pregnant women with no history of known medical conditions; (2) singleton pregnancy; (3) gestational age at inclusion specified; (4) echocardiography performed according to the recommendations of the American Society of Echocardiography or European Association of Echocardiography; (5) timing of echocardiography reported; (6) definitions of SGA, FGR or PE were reported when appropriate; and (7) patient inclusion according to uteroplacental Doppler flow criteria (only applicable in populations of healthy women without medical history of heart disease) or when the relation of uteroplacental Doppler flow to cardiac function was investigated (only applicable for women with known cardiac disease). When uncertainty about eligibility existed, a second author (P.G.P.) was asked to interpret the data. References of the selected articles were cross-checked and full-text articles included when they met the inclusion criteria. Only studies based on echocardiography were included to provide a more reliable comparison and interpretation of the data. Efforts were made to filter out duplicate articles, and publications from the same institution were checked for overlapping data collection periods. The format of the review was structured according to the PRISMA statement for reporting systematic reviews and meta-analyses 15. We considered anatomical parameters such as the presence of structural cardiac defects, cardiac chamber dimensions and wall thickness (assessed with echocardiography), hemodynamic parameters, such as stroke volume, heart rate, cardiac output, systolic and diastolic blood pressure, total vascular resistance (based on systolic and diastolic blood pressure and cardiac output), and echocardiographic systolic function parameters (left ventricular ejection fraction, fractional shortening, systolic tissue velocities (S ) of left ventricular septal and lateral wall) and diastolic function parameters (E and A wave velocities of mitral inflow, E/A ratio, deceleration time of E wave, isovolumetric relaxation time (IVRT), diastolic tissue velocities of left ventricular septal and lateral wall (E, A ), E/E ratio as a measure of left ventricular filling pressure). Results are only described in terms of dysfunction when the data were classified according to the current clinical algorithms and guidelines. Study characteristics and study outcomes were tabulated in Microsoft Excel for Windows (Microsoft Corp., Redmond, WA, USA). Results The literature search yielded a total of 1732 articles retrieved from either PubMed (n=798) or Embase (n=934). After exclusion and cross referencing, 10 articles were included in this systematic review 14, Articles were excluded mainly because they did not relate uteroplacental Doppler measurements to maternal cardiovascular function. An overview of the study characteristics of the included studies is provided in table 1. Details concerning the excluded

120 Maternal cardiac function and uteroplacental Doppler flow: systematic review 119 Table 1. Summary of the included studies evaluating cardiac function in pregnancies with abnormal uterine artery Doppler measurements. Evaluation Source Method Inclusion Pieper et al. Prospective cohort study Melchiorre et al. Prospective case control study Melchiorre et al. Prospective casecohort study Number of patients Gestational week Post partum (months) March 2008-August January 2008-january * March 2008-October * Novelli et al. Prospective cohort study * 24 6 Prefumo et al Prospective cohort study Prefumo et al. Prospective case control study Valensise et al. Prospective cohort study Valensise et al. Prospective cohort study February - August * April 2003-September * August - September * April May * Valensise et al Prospective cohort study * Vasapollo et al Prospective cohort study * Only first author of each study is given. * Total amount of patients with high resistance in uterine arteries and/ or bilateral notching. articles, including reason of exclusion, can be found in table S1. Figure 1 provides an overview of the entire search and inclusion process. The selected articles could be subdivided in two groups: studies relating heart disease in women and uteroplacental flow patterns (n=2) and studies in healthy women relating uteroplacental flow patterns to cardiac function (n=8). Of the first group of studies, Melchiorre et al. investigated the association between abnormal uterine artery resistance indices and maternal cardiac anatomy and function. In 491 pregnancies without known risk factors, the incidence of undiagnosed cardiac defects was significantly higher in women with increased uterine artery resistance than in women with normal uteroplacental flow patterns (relative risk of 12.6 (95% confidence interval )) 17. All newly diagnosed cardiac defects were considered functionally significant. This study suggests a relationship between pre-pregnancy cardiac function and uteroplacental flow. The ZAHARA II study was the only study describing the relationship between maternal cardiac function and uteroplacental Doppler flow patterns in women with known heart disease 14. In these women, 7

121 120 Chapter 7 Figure 1: Flowchart summarizing study selection of papers on maternal cardiac function, uteroplacental Doppler flow and pregnancy outcome. resting heart rate at 20 weeks gestation and cardiac medication use at 20 weeks were associated with uterine artery resistance index. Pre-pregnancy pulmonary atrioventricular valve regurgitation, pre-pregnancy tricuspid annular plane systolic excursion (TAPSE) (as a measure of subpulmonary function) and systemic atrioventricular valve regurgitation at 20 weeks gestation were all associated with altered umbilical artery resistance index at 32 weeks gestation 14. In the second group of studies in healthy women, hemodynamic parameters were studied in women with normal and abnormal uterine artery flow patterns. Women with high uterine artery resistance indices and subsequent poor pregnancy outcome showed lower heart rates, significantly higher blood pressure and higher total vascular resistances in the second trimester than did women with uneventful pregnancies (Table 2). Comparable results were also found in the third trimester of pregnancy (data not shown) 19. Additionally, left ventricular end diastolic diameter and volume were smaller and the relative wall thickness and left ventricular mass index were significantly higher in women with bilateral notching and pregnancy complications than in women with normal pregnancies with no uteroplacental Doppler abnormalities and in women with bilateral notching, but an uncomplicated pregnancy outcome (Table 3).

122 Maternal cardiac function and uteroplacental Doppler flow: systematic review 121 Table 2. Hemodynamic parameters during and after delivery in normal pregnancies with a normal outcome and in pregnancies with bilateral notching of the uterine artery waveform, with normal or complicated outcome. Normal uteroplacental Doppler and outcome Bilateral notching of the uterine artery waveform (controls) Normal outcome Complicated outcome Hemodynamic parameter/study Heart rate (bpm) 24 weeks Post partum 24 weeks Post partum 24 weeks Post partum 78 (73-82) 90 (81-101)* Melchiorre et al. 81(75-87) 79 (71-85) 81 (74-92) Novelli et al ± ± 8 83 ± ± 8 75 ± 15* 71 ± 11 Valensise et al ± ± 9 75 ± 14* 75 ± Vasapollo et al ± ± 9 75 ± 13* 75 ± 11 Diastolic blood pressure (mmhg) Novelli et al ± 9 66 ± ± 9 64 ± ± 11* 71 ± 13* Valensise et al ± ± ± 10* 71 ± Vasapollo et al ± ± ± 12* 70 ± 13* Valensise et al ± ± 9* -- Mean blood pressure (mmhg) Melchiorre et al. 80 (73-83) Novelli et al ± 7 84 ± 9 80 ± 7 82 ± ± 10* 88 ± 10* Valensise et al ± 8 85 ± 9 87 ± 8* 89 ± Vasapollo et al ± 9 85 ± 9 84 ± 10* 88 ± 10* Valensise et al ± ± 7* -- Total vascular resistance (dyne/s/cm 5 ) Melchiorre et al ( ) 1248 ( ) 1570 ( ) Novelli et al ± ± ± ± ± 363* 1449 ± 299* Valensise et al ± ± ± 248* 1458 ± 285* Vasapollo et al ± ± ± 317* 1447 ± 312* Valensise et al ± ± 346* -- Only first author of each study is given. Data are given as mean±sd or median (interquartile range) Complicated outcome included only patients with preeclampsia developing before 34 weeks gestation; * p < 0.05 uncomplicated outcome vs. complicated outcome. p < 0.05 complicated outcome vs. controls. p < 0.05 at 24 weeks vs. post-partum period. 7

123 122 Chapter 7 Table 3. Echocardiographic left ventricular dimensions during pregnancy and after delivery in normal pregnancies and in pregnancies with bilateral notching of the uterine artery waveform, with normal or complicated outcome. Normal uteroplacental Doppler and outcome Bilateral notching of the uterine artery waveform (controls) Normal outcome Complicated outcome Dimension/study 24 weeks Post partum Left ventricular end diastolic diameter (cm) Novelli et al ± weeks Post partum 24 weeks Post partum 4.56 ± ± ± ± 0.31* 4.59 ± ± ± 4.58 ± 0.32* 4.62 ± 0.28 Valensise et al ± 0.29* 4.64 ± Vasapollo et al ± ± 0.28 Valensise et al ± ± 0.3* -- Relative wall thickness Melchiorre et al ( ) Novelli et al ± ( ) 0.30 ± ± ± 0.05 Valensise et al ± ± ( )* ± 0.39 ± 0.05* 0.06* 0.39 ± 0.06* 0.33 ± 0.04* 0.40 ± 0.05* 0.33 ± 0.04* (56-71) (57-71) -- Vasapollo et al ± ± 0.04 Valensise et al ± ± 0.07* -- Left ventricular mass (g) Valensise et al ± ± ± 28* 115 ± 23* Vasapollo et al ± ± ± 32* 112 ± 23 Valensise et al ± Left ventricular mass index (g/m 2 ) Melchiorre et al. 62 (53-71) Novelli et al ± 8 26 ± 6 33 ± 8 26 ± 6 37 ± 9* 30 ± 7* Valensise et al ± 6 26 ± 5 35 ± 7* 30 ± 7* Vasapollo et al ± 6 25 ± 5 38 ± 8* 30 ± 7* Valensise et al ± ± Left ventricular end diastolic volume (ml) Valensise et al ± ± 13* -- Only first author of each study is given. Data are given as mean±sd or median (interquartile range) Complicated outcome included only patients with preeclampsia developing before 34 weeks gestation; * p < 0.05 uncomplicated outcome vs. complicated outcome. p < 0.05 complicated outcome vs. controls. p < 0.05 at 24 weeks vs. post-partum period.

124 Maternal cardiac function and uteroplacental Doppler flow: systematic review 123 Table 4. Echocardiographic parameters of left ventricular diastolic function during and after delivery in normal pregnancies with normal outcome and in pregnancies with bilateral notching of the uterine artery waveform, with normal or complicated outcome. Normal uteroplacental Doppler and outcome Bilateral notching of the uterine artery waveform (controls) Normal outcome Complicated outcome Parameter/study 24 weeks Left atrial maximal area (cm 2 ) Post partum 24 weeks Post partum 24 weeks Post partum ± 7 44 ± ± 12* 43 ± ( ± ± Valensise et al ± ± ± 2.2* 14.2 ± Vasapollo et al ± ± ± 2.7* 14.3 ± 2.7 Valensise et al ± Left atrial fractional area change (%) Valensise et al Vasapollo et al ± 6 43 ± ± 10* 42 ± 9 Valensise et al ± ± 4.7* -- Valensise et al ± Mitral inflow E/A ratio Melchiorre et al ( ) ( ) ) Valensise et al ± 0.50* 1.26 ± 0.26* Valensise et al ± ± Valensise et al ± Septal E (cm/s) Melchiorre et al. 12 (11-15) (11-16) (6-14)* Lateral E (cm/s) Melchiorre et al. 18 (16-19) (16-20) (11-17)* Septal Em/Am (cm/s) Melchiorre et al. 2.7 ( ) Lateral Em/Am (cm/s) Melchiorre et al. 2.9 ( ) Isovolumetric relaxation time (ms) ( ) ( )* ( ) ( ) -- Melchiorre et al. 74 (62-77) (74-80) (71-89) Valensise et al ± 9 84 ± 9 93 ± 8* 94 ± 11* 7

125 124 Chapter 7 Table 4. Echocardiographic parameters of left ventricular diastolic function during and after delivery in normal pregnancies with normal outcome and in pregnancies with bilateral notching of the uterine artery waveform, with normal or complicated outcome. (continued) Normal uteroplacental Doppler and outcome Bilateral notching of the uterine artery waveform (controls) Normal outcome Complicated outcome Parameter/study 24 weeks Post partum 24 weeks Post partum 24 weeks Valensise et al ± ± 7* -- Valensise et al ± Left chamber filling pressure (E/E 1 ) ( ) (5.3-7) -- Melchiorre et al ( ) Post partum Only first author of each study is given. Data are given as mean±sd or median (interquartile range). E, peak early diastolic velocity at mitral valve annulus; Em, peak early diastolic myocardial velocity at the level of the basal myocardium; Am, peak late diastolic myocardial velocity at the level of the basal myocardium. Complicated outcome included only patients with preeclampsia developing before 34 weeks gestation; * p < 0.05 uncomplicated outcome vs. complicated outcome. p < 0.05 complicated outcome vs. controls. p < 0.05 at 24 weeks vs. post-partum period. Table 4 provides an overview of the left ventricular diastolic parameters in women with bilateral notching and poor or normal pregnancy outcome. Women with bilateral notching of the uterine artery and poor pregnancy outcome had smaller left atrial maximal area and compromised atrial function compared with women with an uneventful outcome. Left ventricular diastolic function seemed to be reduced, as indicated by longer IVRT and lower diastolic myocardial velocities, in women with a poor pregnancy outcome. Most of the studies presenting diastolic function parameters interpreted these indices without taking into account load and age-dependency. Given the current knowledge on diastolic function, this is a limitation, hampering the interpretation of the data. Of the included studies, only that of Melchiorre et al. reported and interpreted diastolic function parameters according to current recommendations, and found a higher incidence of left ventricular diastolic dysfunction in women with abnormal uteroplacental flow patterns and early PE (33%), compared with women with an uneventful pregnancy and abnormal (3,1%) or normal (1,9%) uteroplacental flow (p=0.0001) 24. Of the women with early PE, 72% showed impaired myocardial relaxation (according to strain and strain-rate measurements) compared with 21% of women with an uneventful pregnancy and normal uteroplacental flow and 20% of women with uneventful pregnancy and abnormal uteroplacental flow (p< ). Table 5 shows the results of left ventricular systolic function in the same three groups of patients. Function was compromised especially in women with bilateral notching and poor pregnancy outcome, as indicated by the lower cardiac output and stroke volume, significantly lower ejection fraction and midwall fractional shortening 16,18-20,22. Additionally, the average systolic tissue velocity of the lateral and septal sites of the mitral annulus (S (m/s)) was

126 Maternal cardiac function and uteroplacental Doppler flow: systematic review 125 Table 5. Echocardiographic parameters of left ventricular systolic function during pregnancy and after delivery in normal pregnancies with normal outcome and in pregnancies with bilateral notching of the uterine artery waveform, with normal or complicated outcome. Normal uteroplacental Doppler and outcome Bilateral notching of the uterine artery waveform (controls) Normal outcome Complicated outcome Parameter/study Stroke volume (ml) 24 weeks Post partum 24 weeks Post partum 24 weeks Post partum Melchiorre et al. 72 (60-86) (56-80) (57-70) Novelli et al ± ± 9 82 ± ± 8 66 ± 11* 71 ± 13 Valensise et al ± ± ± 13* 68 ± Vasapollo et al ± ± ± 14* 68 ± 13 Valensise et al ± ± 6* -- Cardiac output (L/min) Melchiorre et al ( ) ( ) ( ) -- Novelli et al ± ± ± ± ± 1.49* 5.07 ± 1.32 Valensise et al ± ± ± 1.09* 5.03 ± Vasapollo et al ± ± ± 1.33* 5.05 ± 1.21 Valensise et al ± ± 0.8* -- Ejection fraction (%) Melchiorre et al. 62 (55-67) (62-70)* (48-67) Novelli et al ± 8 65 ± 8 68 ± 9 64 ± 8 62 ± 4* 65 ± 7 Valensise et al ± ± 6* -- Midwall fractional shortening (%) Novelli et al ± ± ± ± ± 5.0* 21.6 ± 4.6 Sc Midwall fractional shortening (%) Novelli et al ± ± ± ± ± 20.7* 87.8 ± 19.1* Only first author of each study is given. Data are given as mean±sd or median (interquartile range). Stresscorrected midwall fractional shortening. Complicated outcome included only patients with preeclampsia developing before 34 weeks gestation; * p < 0.05 uncomplicated outcome vs. complicated outcome. p < 0.05 complicated outcome vs. controls. p < 0.05 at 24 weeks vs. post-partum period. 7

127 126 Chapter 7 significantly lower in women with high uterine artery resistance during the first trimester, compared to that in pregnancies with normal uterine artery resistance (0.11 ± 0.02 vs ± 0.02, respectively) 21. Left ventricular longitudinal systolic dysfunction, defined according to American Society of Echocardiography or European Association of Echardiography guidelines, was seen only in cases of early PE and abnormal uteroplacental flow patterns (17%; p=0.006) 24. Myocardial contractility, as indicated by strain and strain rate measurements, was preserved in all groups. Tables 2-5 show the differences in cardiac function, dimensions and hemodynamic parameters in the postpartum period in the same three groups of patients 16,18,20,22. Women with bilateral notching and poor pregnancy outcome showed higher total vascular resistance and significantly higher blood pressure postpartum than did women with high uterine artery resistance and normal outcome. In addition, relative left ventricular wall thickness was greater and the prolonged IVRT persisted up to 1 year postpartum. Stress-corrected mid-wall fractional shortening (adjusted for LV afterload (table 5) was lower six months post-partum in women with bilateral notching of the uterine artery and poor pregnancy outcome than in women in the two other groups. Discussion To our knowledge, this is the first review exploring the relationship between maternal cardiac function and uteroplacental flow in pregnancy. Although scarce, the existing literature on the subject suggests that maternal cardiovascular status influences the process of placentation and pregnancy outcome. Women with known cardiac disease are known to have a higher incidence of abnormal uteroplacental flow and poor pregnancy outcome and, conversely, women with high uterine artery resistance and pregnancy complications are more likely to be diagnosed with cardiac defects or cardiac dysfunction 14,17,24. Cardiac dysfunction was still present up to 1 year after pregnancy 16,18,20,22. During the placentation process, the penetration of invasive trophoblast into the spiral arteries results in remodeling and subsequent dilatation of the arterial wall, converting the placental bed arteries from high- to low resistance vessels 25,26. Because of limited auto-regulatory capacity, uteroplacental flow depends directly on maternal cardiac performance. During normal pregnancy, cardiac output increases and several adaptations in diastolic function are required, probably as a response to the physiological cardiac hypertrophy developing during pregnancy, which is reflected by the increased left ventricular mass and mass index. During normal uncomplicated pregnancies, diastolic function is not impaired and left ventricular filling pressure remains normal 27. The majority of the studies included in this review investigated cardiac function in healthy women with normal uterine artery resistence vs. high uterine artery resistance. Women

128 Maternal cardiac function and uteroplacental Doppler flow: systematic review 127 with increased uterine artery resistance and subsequent poor pregnancy outcome more often showed impaired systolic and diastolic left ventricular function in comparison with the other groups. Signs of impairment in diastolic function persisted for six months to one year post-partum. Moreover, blood pressure and total vascular resistance were significantly higher and more cardiac hypertrophy was seen in women with high uterine artery resistance and adverse pregnancy outcome than in women with an uncomplicated outcome. Whether these abnormalities are pre-existing or are acquired as a result of pregnancy is open to question. The results of ZAHARA II indicate that not only cardiac dysfunction during pregnancy, but also cardiac dysfunction prior to pregnancy is associated with abnormal uteroplacental Doppler flow and subsequent complicated pregnancy outcome. This finding, combined with the results of Melchiorre at al., that women had cardiac dysfunction prior to pregnancy, supports the hypothesis that, at least in a subset of women, high uterine artery resistance during pregnancy is primarily a sign of pre-existent suboptimal cardiac performance and a limited ability to adapt properly to pregnancy 14,17. The fact that abnormalities in cardiac adaption are seen before clinical onset of PE also strengthens this hypothesis 16. Roberts et al. postulated that the initiating phase of PE is characterized by placental hypoperfusion 28. Failure to increase cardiac output owing to suboptimal cardiac performance that existed before pregnancy may contribute to placental hypoperfusion and therefore contribute subsequently to the abnormal placental development and high uterine artery resistance. Almost all studies described here focused on left ventricular function, although right ventricular dysfunction might also play a key role in placental development and uteroplacental flow regulation, since right ventricular dysfunction can also limit cardiac output. Melchiorre et al. found recently that women with pre-term PE showed global systolic dysfunction and impaired myocardial contractility, as well as global diastolic dysfunction and impaired myocardial relaxation of the right ventricle during pregnancy, persisting up to one year after pregnancy 29,30. From this point of view, our finding of an association between prepregnancy right ventricular function (TAPSE) and uteroplacental flow patterns in pregnancy is very interesting 14, and suggests that placental development and uteroplacental flow regulation might be influenced by pre-existing abnormal right ventricular function. This aspect needs further research. The results of Melchiorre et al. also highlight an important role for right ventricular dysfunction, since most of the cardiac defects described in their paper are defects with right-sided sequelae 17. Although immune mechanisms have been advocated as a cause of defective trophoblast invasion of the spiral arteries 31, this review suggest that mechanisms involving perturbed hemodynamics during the process of placentation may result in the same common pathway, i.e. altered placental perfusion as indicated by increased resistance to flow in the uterine arteries. A few limitations should be borne in mind while interpreting these results. Small sample size in some of the studies included in the review and variable data presentation hampers 7

129 128 Chapter 7 comparison of the results. Parameters of diastolic function did not always include tissue Doppler measurements, preventing proper interpretation of diastolic dysfunction. Moreover, echocardiographic measurements and calculations are subject to measurement errors. Based on the findings of this review, we suggest that there is a role for routine assessment of uteroplacental Doppler investigations in women with known cardiac dysfunction, since this could be associated with abnormal uteroplacental flow patterns. Early recognition of risk factors can better tailor pregnancy care in these women. Moreover, in a subset of healthy women, increased uterine artery resistance may be a sign of suboptimal subclinical cardiac performance. It is known from the general population that the presence of asymptomatic diastolic or systolic dysfunction is associated with the later development of clinical overt heart failure 32,33. Thus, it may be appropriate to subject women with hypertension, abnormal uterine artery Doppler measurements and subsequent complicated pregnancy outcome to a detailed cardiac assessment in order to diagnose cardiac dysfunction and unrecognized cardiac defects and to tailor treatment to prevent progression to overt clinical heart failure. Moreover, when assessing women with cardiovascular symptoms, pregnancy outcome should be taken into account for accurate risk stratification. In conclusion, there is increasing evidence for a link between pre-existent subclinical cardiac dysfunction and poor placentation. Further study is warranted in order to fully clarify the exact pathophysiology of abnormal placentation in these circumstances. Conflicts of interest None.

130 Maternal cardiac function and uteroplacental Doppler flow: systematic review 129 References 1. Prefumo F, Sebire NJ, Thilaganathan B. Decreased endovascular trophoblast invasion in first trimester pregnancies with high-resistance uterine artery Doppler indices. Hum Reprod 2004; 19 : Cnossen JS, Morris RK, ter Riet G, Mol BW, van der Post JA, Coomarasamy A, Zwinderman AH, Robson SC, Bindels PJ, Kleijnen J, Khan KS. Use of uterine artery Doppler ultrasonography to predict pre-eclampsia and intrauterine growth restriction: a systematic review and bivariable meta-analysis. CMAJ 2008; 178 : Harrington K, Cooper D, Lees C, Hecher K, Campbell S. Doppler ultrasound of the uterine arteries: the importance of bilateral notching in the prediction of pre-eclampsia, placental abruption or delivery of a small-for-gestational-age baby. Ultrasound Obstet Gynecol 1996; 7 : Drenthen W, Boersma E, Balci A, Moons P, Roos-Hesselink JW, Mulder BJ, Vliegen HW, van Dijk AP, Voors AA, Yap SC, van Veldhuisen DJ, Pieper PG, ZAHARA Investigators. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010; 31 : Drenthen W, Pieper PG, Roos-Hesselink JW, van Lottum WA, Voors AA, Mulder BJ, van Dijk AP, Vliegen HW, Yap SC, Moons P, Ebels T, van Veldhuisen DJ, ZAHARA Investigators. Outcome of pregnancy in women with congenital heart disease: a literature review. J Am Coll Cardiol 2007; 49 : Khairy P, Ouyang DW, Fernandes SM, Lee-Parritz A, Economy KE, Landzberg MJ. Pregnancy outcomes in women with congenital heart disease. Circulation 2006; 113 : Siu SC, Colman JM, Sorensen S, Smallhorn JF, Farine D, Amankwah KS, Spears JC, Sermer M. Adverse neonatal and cardiac outcomes are more common in pregnant women with cardiac disease. Circulation 2002; 105 : Siu SC, Sermer M, Colman JM, Alvarez AN, Mercier LA, Morton BC, Kells CM, Bergin ML, Kiess MC, Marcotte F, Taylor DA, Gordon EP, Spears JC, Tam JW, Amankwah KS, Smallhorn JF, Farine D, Sorensen S, Cardiac Disease in Pregnancy (CARPREG) Investigators. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001; 104 : Damman K, van Deursen VM, Navis G, Voors AA, van Veldhuisen DJ, Hillege HL. Increased central venous pressure is associated with impaired renal function and mortality in a broad spectrum of patients with cardiovascular disease. J Am Coll Cardiol 2009; 53 : Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Charlesworth A, Hampton JR, van Veldhuisen DJ. Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation 2000; 102 : van Deursen VM, Damman K, Hillege HL, van Beek AP, van Veldhuisen DJ, Voors AA. Abnormal liver function in relation to hemodynamic profile in heart failure patients. J Card Fail 2010; 16 : Westenbrink BD, Voors AA, de Boer RA, Schuringa JJ, Klinkenberg T, van der Harst P, Vellenga E, van Veldhuisen DJ, van Gilst WH. Bone marrow dysfunction in chronic heart failure patients. Eur J Heart Fail 2010; 12 : Balci A, Sollie KM, Mulder BJ, de Laat MW, Roos-Hesselink JW, van Dijk AP, Wajon EM, Vliegen HW, Drenthen W, Hillege HL, Aarnoudse JG, van Veldhuisen DJ, Pieper PG. Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study. Am Heart J 2011; 161 : e1. 7

131 130 Chapter Pieper PG, Balci A, Aarnoudse JG, Kampman MA, Sollie KM, Groen H, Mulder BJ, Oudijk MA, Roos-Hesselink JW, Cornette J, van Dijk AP, Spaanderman ME, Drenthen W, van Veldhuisen DJ, ZAHARA II investigators. Uteroplacental blood flow, cardiac function, and pregnancy outcome in women with congenital heart disease. Circulation 2013; 128 : Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009; 6 : e Novelli GP, Vasapollo B, Gagliardi G, Tiralongo GM, Pisani I, Manfellotto D, Giannini L, Valensise H. Left ventricular midwall mechanics at 24 weeks gestation in high-risk normotensive pregnant women: relationship to placenta-related complications of pregnancy. Ultrasound Obstet Gynecol 2012; 39 : Melchiorre K, Sutherland GR, Liberati M, Bhide A, Thilaganathan B. Prevalence of maternal cardiac defects in women with high-resistance uterine artery Doppler indices. Ultrasound Obstet Gynecol 2011; 37 : Valensise H, Vasapollo B, Gagliardi G, Novelli GP. Early and late preeclampsia: two different maternal hemodynamic states in the latent phase of the disease. Hypertension 2008; 52 : Prefumo F, Muiesan ML, Perini R, Paini A, Bonzi B, Lojacono A, Agabiti-Rosei E, Frusca T. Maternal cardiovascular function in pregnancies complicated by intrauterine growth restriction. Ultrasound Obstet Gynecol 2008; 31 : Vasapollo B, Novelli GP, Valensise H. Total vascular resistance and left ventricular morphology as screening tools for complications in pregnancy. Hypertension 2008; 51 : Prefumo F, Sharma R, Brecker SJ, Gaze DC, Collinson PO, Thilaganathan B. Maternal cardiac function in early pregnancies with high uterine artery resistance. Ultrasound Obstet Gynecol 2007; 29 : Valensise H, Vasapollo B, Novelli GP, Larciprete G, Romanini ME, Arduini D, Galante A, Romanini C. Maternal diastolic function in asymptomatic pregnant women with bilateral notching of the uterine artery waveform at 24 weeks gestation: a pilot study. Ultrasound Obstet Gynecol 2001; 18 : Valensise H, Novelli GP, Vasapollo B, Borzi M, Arduini D, Galante A, Romanini C. Maternal cardiac systolic and diastolic function: relationship with uteroplacental resistances. A Doppler and echocardiographic longitudinal study. Ultrasound Obstet Gynecol 2000; 15 : Melchiorre K, Sutherland G, Sharma R, Nanni M, Thilaganathan B. Mid-gestational maternal cardiovascular profile in preterm and term pre-eclampsia: a prospective study. BJOG 2013; 120 : Aardema MW, Oosterhof H, Timmer A, van Rooy I, Aarnoudse JG. Uterine artery Doppler flow and uteroplacental vascular pathology in normal pregnancies and pregnancies complicated by pre-eclampsia and small for gestational age fetuses. Placenta 2001; 22 : Pijnenborg R, Anthony J, Davey DA, Rees A, Tiltman A, Vercruysse L, van Assche A. Placental bed spiral arteries in the hypertensive disorders of pregnancy. Br J Obstet Gynaecol 1991; 98 : Fok WY, Chan LY, Wong JT, Yu CM, Lau TK. Left ventricular diastolic function during normal pregnancy: assessment by spectral tissue Doppler imaging. Ultrasound Obstet Gynecol 2006; 28 : Roberts JM, Gammill HS. Preeclampsia: recent insights. Hypertension 2005; 46 :

132 Maternal cardiac function and uteroplacental Doppler flow: systematic review Melchiorre K, Sutherland GR, Watt-Coote I, Liberati M, Thilaganathan B. Severe myocardial impairment and chamber dysfunction in preterm preeclampsia. Hypertens Pregnancy 2012; 31 : Melchiorre K, Sutherland GR, Liberati M, Thilaganathan B. Preeclampsia is associated with persistent postpartum cardiovascular impairment. Hypertension 2011; 58 : Laresgoiti-Servitje E, Gomez-Lopez N, Olson DM. An immunological insight into the origins of pre-eclampsia. Hum Reprod Update 2010; 16 : Redfield MM, Jacobsen SJ, Burnett JC,Jr, Mahoney DW, Bailey KR, Rodeheffer RJ. Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic. JAMA 2003; 289 : Kane GC, Karon BL, Mahoney DW, Redfield MM, Roger VL, Burnett JC,Jr, Jacobsen SJ, Rodeheffer RJ. Progression of left ventricular diastolic dysfunction and risk of heart failure. JAMA 2011; 306 :

133 132 Chapter 7 Supplementary Table Table S1. Characteristics and reason for exclusion of excluded studies. Source Method Reason for exclusion Melchiorre et al Review No original research paper. Melchiorre et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Kuleva et al Prospective cohort study Study describing twin pregnancies. Ghi et al Prospective cohort study Study describing twin pregnancies. Melchiorre et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Strobl et al Case-control study No inclusion according to uteroplacental flow criteria. Comas et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Vettori et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Ogueh et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Zentner et al Longitudinal study No inclusion according to uteroplacental flow criteria. Bamfo et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Turan et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Khaw et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Bamfo et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Bamfo et al Longitudinal study No inclusion according to uteroplacental flow criteria. Bamfo et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Fok et al Prospective observational study No inclusion according to uteroplacental flow criteria. Valensise et al Prospective observational study No inclusion according to uteroplacental flow criteria. Bamfo et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Vasapollo et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Valensise et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Vasapollo et al Prospective case-control study No inclusion according to uteroplacental flow criteria. Kametas et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Kametas et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Tyldum et al Prospective observational study No inclusion according to uteroplacental flow criteria. Kametas et al Cross-sectional study No inclusion according to uteroplacental flow criteria. Valensise et al Prospective case-control study No inclusion according to uteroplacental flow criteria.

134 Maternal cardiac function and uteroplacental Doppler flow: systematic review 133 References 1. Melchiorre K, Thilaganathan B. Maternal cardiac function in preeclampsia. Curr Opin Obstet Gynecol 2011; 23 : Melchiorre K, Sutherland GR, Liberati M, Thilaganathan B. Preeclampsia is associated with persistent postpartum cardiovascular impairment. Hypertension 2011; 58 : Kuleva M, Youssef A, Maroni E, Contro E, Pilu G, Rizzo N, Pelusi G, Ghi T. Maternal cardiac function in normal twin pregnancy: a longitudinal study. Ultrasound Obstet Gynecol 2011; 38 : Ghi T, Kuleva M, Youssef A, Maroni E, Nanni M, Pilu G, Rizzo N, Pelusi G. Maternal cardiac function in complicated twin pregnancy: a longitudinal study. Ultrasound Obstet Gynecol 2011; 38 : Melchiorre K, Sutherland GR, Baltabaeva A, Liberati M, Thilaganathan B. Maternal cardiac dysfunction and remodeling in women with preeclampsia at term. Hypertension 2011; 57 : Strobl I, Windbichler G, Strasak A, Weiskopf-Schwendinger V, Schweigmann U, Ramoni A, Scheier M. Left ventricular function many years after recovery from pre-eclampsia. BJOG 2011; 118 : Comas M, Crispi F, Cruz-Martinez R, Martinez JM, Figueras F, Gratacos E. Usefulness of myocardial tissue Doppler vs conventional echocardiography in the evaluation of cardiac dysfunction in early-onset intrauterine growth restriction. Am J Obstet Gynecol 2010; 203 : 45.e1-45.e7. 8. Vettori DV, Rohde LE, Clausell N. Asymptomatic left ventricular dysfunction in puerperal women: an echocardiographic-based study. Int J Cardiol 2011; 149 : Ogueh O, Brookes C, Johnson MR. A longitudinal study of the maternal cardiovascular adaptation to spontaneous and assisted conception pregnancies. Hypertens Pregnancy 2009; 28 : Zentner D, du Plessis M, Brennecke S, Wong J, Grigg L, Harrap SB. Deterioration in cardiac systolic and diastolic function late in normal human pregnancy. Clin Sci (Lond) 2009; 116 : Bamfo JE, Kametas NA, Chambers JB, Nicolaides KH. Maternal cardiac function in normotensive and pre-eclamptic intrauterine growth restriction. Ultrasound Obstet Gynecol 2008; 32 : Turan OM, De Paco C, Kametas N, Khaw A, Nicolaides KH. Effect of parity on maternal cardiac function during the first trimester of pregnancy. Ultrasound Obstet Gynecol 2008; 32 : Khaw A, Kametas NA, Turan OM, Bamfo JE, Nicolaides KH. Maternal cardiac function and uterine artery Doppler at weeks in the prediction of pre-eclampsia in nulliparous women. BJOG 2008; 115 : Bamfo JE, Kametas NA, Nicolaides KH, Chambers JB. Reference ranges for tissue Doppler measures of maternal systolic and diastolic left ventricular function. Ultrasound Obstet Gynecol 2007; 29 : Bamfo JE, Kametas NA, Nicolaides KH, Chambers JB. Maternal left ventricular diastolic and systolic long-axis function during normal pregnancy. Eur J Echocardiogr 2007; 8 : Bamfo JE, Kametas NA, Chambers JB, Nicolaides KH. Maternal cardiac function in fetal growthrestricted and non-growth-restricted small-for-gestational age pregnancies. Ultrasound Obstet Gynecol 2007; 29 : Fok WY, Chan LY, Wong JT, Yu CM, Lau TK. Left ventricular diastolic function during normal pregnancy: assessment by spectral tissue Doppler imaging. Ultrasound Obstet Gynecol 2006; 28 :

135 134 Chapter Valensise H, Vasapollo B, Novelli GP, Pasqualetti P, Galante A, Arduini D. Maternal total vascular resistance and concentric geometry: a key to identify uncomplicated gestational hypertension. BJOG 2006; 113 : Bamfo JE, Kametas NA, Turan O, Khaw A, Nicolaides KH. Maternal cardiac function in fetal growth restriction. BJOG 2006; 113 : Vasapollo B, Valensise H, Novelli GP, Altomare F, Galante A, Arduini D. Abnormal maternal cardiac function precedes the clinical manifestation of fetal growth restriction. Ultrasound Obstet Gynecol 2004; 24 : Valensise H, Vasapollo B, Novelli GP, Larciprete G, Andreoli A, Altomare F, Di Pierro G, Galante A, Arduini D, De Lorenzo A, Caserta D. Maternal cardiac systolic function and total body water estimation in normal and gestational hypertensive women. Acta Diabetol 2003; 40 Suppl 1 : S Vasapollo B, Valensise H, Novelli GP, Larciprete G, Di Pierro G, Altomare F, Casalino B, Galante A, Arduini D. Abnormal maternal cardiac function and morphology in pregnancies complicated by intrauterine fetal growth restriction. Ultrasound Obstet Gynecol 2002; 20 : Kametas NA, McAuliffe F, Cook B, Nicolaides KH, Chambers J. Maternal left ventricular transverse and long-axis systolic function during pregnancy. Ultrasound Obstet Gynecol 2001; 18 : Kametas NA, McAuliffe F, Hancock J, Chambers J, Nicolaides KH. Maternal left ventricular mass and diastolic function during pregnancy. Ultrasound Obstet Gynecol 2001; 18 : Tyldum EV, Backe B, Stoylen A, Slordahl SA. Maternal left ventricular and endothelial functions in preeclampsia. Acta Obstet Gynecol Scand 2012; 91 : Kametas NA, McAuliffe F, Krampl E, Chambers J, Nicolaides KH. Maternal cardiac function during pregnancy at high altitude. BJOG 2004; 111 : Valensise H, Novelli GP, Vasapollo B, Di Ruzza G, Romanini ME, Marchei M, Larciprete G, Manfellotto D, Romanini C, Galante A. Maternal diastolic dysfunction and left ventricular geometry in gestational hypertension. Hypertension 2001; 37 :

136

137

138 Chapter 8 Summary and future perspectives A

139

140 Summary & future perspectives 139 Summary In part I of this thesis we described the incidence of cardiovascular complications, new predictors of cardiovascular complications during and after pregnancy, and we assessed the impact of pregnancy on cardiac function and remodeling. Pregnancy in women with CHD is known to be associated with increased incidence of cardiovascular complications, but also obstetric and neonatal complications are common and several risk prediction models have been developed 1,2. In chapter 2 we showed that cardiovascular complications occur in 10.3% of the pregnancies in the ZAHARA II study. Women with congenital heart disease had higher Nt-proBNP levels at 20 weeks gestation compared to healthy pregnant women. Women with cardiovascular complications had higher Nt-proBNP levels at 20 weeks gestation compared to women without cardiovascular complications during pregnancy and a Nt-proBNP level > 128 pg/ml at 20 weeks gestation was independently associated with the occurrence of cardiovascular events later in pregnancy. The negative predictive value of Nt-proBNP < 128 pg/ml was 96.9% and Nt-proBNP levels > 128 pg/ml had additional value in predicting cardiovascular complications during pregnancy on top of the other identified predictors. The risk prediction models that are commonly used to predict pregnancy risk, all use preconception parameters. Nt-proBNP levels < 128 pg/ml appear helpful in identifying patients at low risk of cardiovascular complications while they are already pregnant, which provides a helpful additional tool for doctors caring for these patients. In chapter 3 we demonstrated that no statistically significant changes occur in right and left ventricular function parameters and ventricular dimensions during and after pregnancy in women with CHD. The absolute level of ventricular function parameters and dimensions differs clearly between women with CHD and healthy pregnant women. Fitted longitudinal profiles over time show that the changes seen in women with CHD are comparable to healthy women. This is valuable information during preconception counseling of these women. In this study, visual patterns and model fit criteria suggest non-linear variation in parameters over time; however, there was no statistically significant effect of time for any of these parameters. This might be caused by insufficient power. Follow-up with echocardiography is still valuable, since changes in the individual patient do occur and can be of clinical relevance. In chapter 4 we described that cardiovascular complications during the first year post-partum are relatively rare. However, women with a cardiovascular event during pregnancy are prone to develop cardiovascular events during the first year post-partum and they have increased subpulmonary ventricular diameters compared with preconception values. These findings underline the importance of post-partum follow-up. 8 In part II of the thesis we investigated if maternal cardiac dysfunction plays a role in the pathogenesis of obstetric and neonatal complications. Placenta related complications (i.e.

141 140 Chapter 8 hypertensive disorders of pregnancy and fetal growth restriction) are common in women with congenital heart disease and the ZAHARA II study was designed in order to investigate the effect of impaired maternal cardiac function on the uteroplacental circulation and is relationship with the occurrence of adverse obstetric and offspring outcome 3. In Chapter 5 we described the main findings of the ZAHARA II study. We found differences in uteroplacental Doppler flow (UDF) parameters between women with CHD and healthy pregnant women. Cardiac dysfunction was associated with an abnormal UDF pattern, while abnormal UDF was associated with neonatal complications. In women with tetralogy of Fallot, we found that abnormal UDF patterns were more prevalent compared to healthy pregnant women. Right and left ventricular dysfunction appeared to be associated with impaired UDF (chapter 6). We confirmed in a smaller, but more homogeneous group of patients the results of the ZAHARA II study. These are important findings, which suggest that maternal cardiac dysfunction is one of the contributors to the occurrence of defective placentation and impaired placenta function, with subsequent poor pregnancy outcome. More fundamental research is required to provide a better understanding of the mechanisms involved. Finally, several studies have reported impaired cardiac function in healthy women with poor uteroplacental flow and poor pregnancy outcome, which is also present one year postpartum. In chapter 7 we reviewed the literature systematically and we found increasing evidence for an association between pre-existing cardiac dysfunction, poor placentation and poor pregnancy outcome. We postulated that pre-existing cardiac dysfunction, as a result of either known heart disease or a subclinical latent condition is one of the common denominators of poor placentation leading to poor pregnancy outcome. Discussion In order to reduce the burden of cardiovascular complications during pregnancy in women with congenital heart disease, it is of utmost importance to counsel patients and their partners about the risks during pregnancy, and it is essential to have an adequate follow-up plan during pregnancy. Current European Society of Cardiology (ESC) guidelines on management of cardiovascular disease during pregnancy state that regular follow-up of women with CHD during pregnancy is indicated. The frequency of follow-up depends on modified WHO risk classification (see further) 4. However, since publication of the guidelines in 2011, important new insights were reported. Integrating the new insights gathered in this thesis into the current knowledge on pregnancy in women with CHD, I propose a new follow-up algorithm, as displayed in figure 1. It should be noticed that several very high risk patients, i.e. those with a mechanical valve or with severe aortic dilatation are excluded from this algorithm, since monthly follow-up is indicated irrespective of the findings during follow-up.

142 Summary & future perspectives 141 Preconception risk assessment 1 Modified WHO risk class WHO II WHO III WHO IV WHO I Discuss termination of pregnancy. Follow-up first trimester Follow-up first trimester Follow-up second trimester Continuing pregnancy? Follow up as WHO III. Refer all patients to obstetrician: UDF at 20 & 32 weeks gestation. Clinical assessment Echocardiogram Refer all patients to obstetrician: UDF at 20 & 32 weeks gestation. Clinical assessment Echocardiogram Refer all patients to obstetrician: UDF at 20 & 32 weeks gestation. Clinical assessment Echocardiogram Nt-proBNP measurement Yes: Monthly follow-up with clinical assessment and echocardiogram. Deteriorating cardiac / valvular function and/or ventricular dilatation? Yes: Bimonthly follow-up with clinical assessment and echocardiogram. Deteriorating cardiac / valvular function and/or ventricular dilatation? Yes: Follow-up during third trimester with clinical assessment, Nt-proBNP and echocardiogram.. Deteriorating cardiac / valvular function and/or ventricular dilatation? Clinical assessment Echocardiogram Nt-proBNP measurement No: Next follow-up at 20 weeks gestation. No: Follow-up next trimester Nt-proBNP < 128 pg/ml: Follow-up postpartum. No: Nt-proBNP value? Yes: Monthly follow-up with clinical assessment and echocardiogram. Deteriorating cardiac / valvular function and/or ventricular dilatation on echocardiogram? Nt-proBNP > 128 pg/ml: Follow-up during third trimester with clinical assessment and echocardiogram. Nt-proBNP > 128 pg/ml No: Nt-proBNP value? Nt-proBNP < 128 pg/ml: Bimonthly follow-up with clinical assessment and echocardiogram. Figure 1: Proposed follow-up algorithm for pregnant women with congenital heart disease. 1 Ideally performed before conception. When patient presents for the first time when already pregnant, perform risk assessment as soon as possible. 8

143 142 Chapter 8 Adequate follow-up starts with the assessment of maternal cardiac risk during pregnancy, which should be expressed according to the modified WHO risk classification 4. This classification integrates all known risk predictors, co-morbidities, and current cardiac condition (figure 2), and it has recently proven to be superior to other prediction models in predicting the pregnancy risk of women with CHD 5-7. NYHA functional class Residual lesions Exercise capacity Previous cardiovascular events Type of congenital heart disease Current myocardial and valvular function Medication use Risk assessment Comorbid conditions Modified WHO risk classification Figure 2: Information required for cardiovascular risk assessment during pregnancy in women with congenital heart disease. According to the current guidelines, the follow-up frequency depends on the modified WHO risk class, but should at least be done once every trimester with performance of an echocardiogram 4. A recent report from the European registry on pregnancy and cardiac disease (ROPAC) shows that most of the cardiovascular events were encountered by women in modified WHO risk class III and IV, with heart failure being the most frequent observed complication 8. Heart failure tends to develop more frequently during the late second or early third trimester and post-partum period, with modified WHO risk class III being an important risk factor 9. A recent report from the ZAHARA II study also shows that cardiovascular events are more prevalent in women in modified WHO risk classes III and IV, with no cardiovascular complications at all in women with modified WHO risk class I 5. Based on these reports, it should be considered for women in modified WHO risk class I to monitor them only once during the second and the third trimester (figure 1).

144 Summary & future perspectives 143 As can be seen in figure 1, an echocardiogram should be performed during every visit to the outpatient clinic and follow-up should be intensified when deterioration is present. When echocardiography is not sufficient for adequate follow-up (i.e. determination of aortic dimensions), other imaging modalities (MRI/CT without contrast) should be used. Cornette et al. showed that cardiac output, stroke volume and E/E ratio increased during pregnancy and that left ventricular ejection fraction decreased. In chapter 3 of this thesis, visual patterns and model-fit criteria did also suggest non-linear variation over time, but it did not reach statistical significance in our study. This might be caused by insufficient power. However, since changes in cardiac function can occur and the changes seen might be of clinical relevance for individual patients, follow-up with echocardiography is indicated during pregnancy in women with CHD. It would be helpful for doctors caring for pregnant women with CHD, to have an additional tool to determine the risk of future cardiovascular complications while the patient is already pregnant. Tanous et al. were the first to report on B-type natriuretic peptide (BNP) during pregnancy in women with heart disease. They found that women with heart disease had higher values of BNP during pregnancy and that high levels were associated with cardiovascular complications during pregnancy. BNP levels > 100 pg/ml had a sensitivity of 100% and a specificity of 70% of identifying women with cardiovascular events during pregnancy 10. In chapter 2 of this thesis, we described that Nt-proBNP > 128 pg/ml at 20 weeks gestation was independently associated with the occurrence of cardiovascular events later in pregnancy and had additional value in predicting the occurrence of cardiovascular events on top of other identified predictors 11. Nt-proBNP levels > 128 pg/ml at 20 weeks gestation had a sensitivity of 81.3% and a specificity of 61.8% for cardiovascular complications later in pregnancy. The findings of both studies underline that natriuretic peptide levels can be helpful in identifying patients at risk of cardiovascular complications during pregnancy. The negative predictive value of Nt-proBNP < 128 pg/ml is very high. Therefore, it seems justifiable to perform follow-up only once during pregnancy (at 20 weeks gestation) for women in modified WHO risk class I, with Nt-proBNP < 128 pg/ml, and no deterioration of cardiac or valvular function at 20 weeks gestation and no dilatation of ventricular dimensions. It should be kept in mind that the predictive value of a test is strongly dependent on the prevalence of cardiovascular complications. Therefore, it seems reasonable to use Nt-proBNP levels to increase follow-up frequency in women with modified WHO III-IV risk class, since they are at highest risk of cardiovascular complications. 8 As discussed in chapter 1, women with CHD are prone to develop obstetric complications and have higher risk of adverse neonatal outcome during pregnancy. In particular more placentarelated complications (i.e. hypertensive disorders of pregnancy, fetal growth restriction) are observed. In chapter 5 we described the main results of the ZAHARA II study. We found dif-

145 144 Chapter 8 ferences in UDF patterns between pregnant women with CHD compared to healthy women and cardiac function was associated with abnormal UDF patterns. TAPSE preconception (as a measure of systolic subpulmonary ventricular function), as well as pulmonary AV valve regurgitation preconception, and systemic AV valve regurgitation at 20 weeks gestation, were associated with abnormal UDF patterns in a multivariable analysis. In a cohort of 55 patients with tetralogy of Fallot, right and left ventricular function parameters were associated with abnormal UDF patterns (chapter 6). These findings suggest that pre-existent cardiac dysfunction, as well as cardiac dysfunction during pregnancy might influence placental development and placental function. However, the underlying mechanism remains unclear. After systematically reviewing the literature (chapter 7), we found increasing evidence for an association between pre-existing cardiac dysfunction, poor placentation, and poor pregnancy outcome. At this moment, uteroplacental Doppler flow measurements are not routinely performed during follow-up of women with CHD. I recommend referring all pregnant women with heart disease to an obstetrician, in order to routinely investigate uteroplacental flow at 20 and 32 weeks gestation. Early recognition of abnormal UDF patterns in these women might reduce the burden of obstetric and fetal complications. With early detection of placental insufficiency, adequate fetal monitoring can be started. When indicated, premature labor can be induced or therapy can be initiated. Not specified in figure 1 is the role of holter monitoring during follow-up. The role of performing holter monitoring routinely in every patient has never been investigated. However, several specific defects have a high a-priori risk of arrhythmia or conduction disorders (i.e. women with atrial repair of transposition of the great arteries). An increased risk of arrhythmia is also present in women with a history of arrhythmia. Holter monitoring, at least once during the third trimester, in these type of patients seems justifiable, since most of the hemodynamic changes have reached a steady state by then. It provides also useful information for management during delivery. In patients reporting complaints of palpitations or dizziness holter monitoring should be performed at least once during pregnancy. When symptomatic arrhythmia is present, therapy should be induced. For the post-partum period, I recommend follow-up with clinical assessment and the performance of an echocardiogram one year post-partum, in particular for women with cardiovascular complications during pregnancy. Chapter 4 of this thesis, as well as other studies, showed that pregnancy is associated with worsening cardiac function afterwards, and women with cardiovascular complications during pregnancy are prone to develop cardiovascular complications after pregnancy Close follow-up post-partum is therefore warranted in these patients. Women with a high mortality risk after delivery should be kept in hospital during the first two to six weeks. Thereafter, weekly follow-up is indicated. For women classified in modi-

146 Summary & future perspectives 145 fied WHO risk class IV follow-up should be performed one month post-partum. For women in modified WHO risk class III during pregnancy, scheduled follow-up three months after delivery is recommended. Follow-up as early as six weeks post-partum should be considered for women with considerable deterioration in cardiac function during pregnancy. In conclusion, this thesis added valuable new insights to the knowledge of cardiovascular risk in women with congenital heart disease during and after pregnancy, and some elucidation was brought about the pathogenesis of the increased obstetric and neonatal complication rate. Recommendations for integration of these findings into daily clinical practice have been discussed, but its final place should be addressed in future revisions of the ESC guideline management of cardiovascular disease during pregnancy. Future research An important problem that remains in the management of pregnancy in women with CHD is the management of women with mechanical valve prosthesis. In chapter 2 we identified mechanical valve prosthesis again as an important predictor of cardiovascular complications, with valve thrombosis being the most important, and potentially life-threatening, complication. There is no ideal anticoagulation regimen, as every approach has inherent risks and benefits for both mother and fetus. A recent report of the ROPAC registry showed again that there is no clear evidence in support of one approach over another 19. It is highly recommended to perform a large, prospective observational study, in which the outcomes of different anticoagulation regimens can be compared. Adding strict protocols for monitoring the level of anticoagulation in the different regimens should be considered (INR / peak-/ through anti Xa levels or both). This will be the only way to see which anticoagulation regimen is superior and what kind of monitoring should be recommended to that regimen. It would also be valuable to perform a study which investigates whether or not it is useful to measure Nt-proBNP levels in women with modified WHO risk class II. With the current evidence it does not seem appropriate to remove third trimester follow-up from the schedule, but this specific research may be helpful to make an evidence based decision. The predictive role of Nt-proBNP before pregnancy in predicting cardiovascular complications during pregnancy could also be a valuable addition to preconception counseling, in particular for the risk of the development of heart failure and arrhythmia. Another intriguing issue that should be addressed in future research is the role of the right ventricle during pregnancy in women with CHD. We identified right ventricular dysfunction as an independent predictor of complications during pregnancy (chapter 2) and the right ventricle tends to dilate in women with complications during pregnancy (chapter 4). Interest- 8

147 146 Chapter 8 ingly, right ventricular dysfunction (TAPSE) was also associated with impaired uteroplacental flow in women with congenital heart disease. Future research should focus on potential mechanisms of how right ventricular dysfunction influences the placentation process. A part of the explanation might be sought in increased venous pressure, which was described previously as a potential mechanism for disturbed placentation 20. The association between systemic and pulmonary atrioventricular valve regurgitation and impaired uteroplacental circulation might also point in this direction. Histological examination of placental tissue combined with extensive uteroplacental Doppler investigations will be necessary to elucidate the patho-physiological mechanism further. Before this, it would be of interest to investigate whether uteroplacental Doppler flow is already disturbed in the first trimester of pregnancy. This would provide more fundamental evidence that cardiac dysfunction influences placental development. These are essential first steps before we can even speculate about influencing pregnancy outcome in women with CHD. Until then, watchful waiting with regular followup is warranted.

148 Summary & future perspectives 147 References 1. Drenthen W, Boersma E, Balci A, Moons P, Roos-Hesselink JW, Mulder BJ, et al. Predictors of pregnancy complications in women with congenital heart disease. Eur Heart J 2010 Sep; 31(17): Siu SC, Sermer M, Colman JM, Alvarez AN, Mercier LA, Morton BC, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001 Jul 31; 104(5): Balci A, Sollie KM, Mulder BJ, de Laat MW, Roos-Hesselink JW, van Dijk AP, et al. Associations between cardiovascular parameters and uteroplacental Doppler (blood) flow patterns during pregnancy in women with congenital heart disease: Rationale and design of the Zwangerschap bij Aangeboren Hartafwijking (ZAHARA) II study. Am Heart J 2011 Feb; 161(2): e1. 4. European Society of Gynecology, Association for European Paediatric Cardiology, German Society for Gender Medicine, Authors/Task Force Members, Regitz-Zagrosek V, Blomstrom Lundqvist C, et al. ESC Guidelines on the management of cardiovascular diseases during pregnancy: the Task Force on the Management of Cardiovascular Diseases during Pregnancy of the European Society of Cardiology (ESC). Eur Heart J 2011 Dec; 32(24): Balci A, Sollie-Szarynska KM, van der Bijl AG, Ruys TP, Mulder BJ, Roos-Hesselink JW, et al. Prospective validation and assessment of cardiovascular and offspring risk models for pregnant women with congenital heart disease. Heart 2014 Sep; 100(17): Pijuan-Domenech A, Galian L, Goya M, Casellas M, Merced C, Ferreira-Gonzalez I, et al. Cardiac complications during pregnancy are better predicted with the modified WHO risk score. Int J Cardiol 2015 Sep 15; 195: Lu CW, Shih JC, Chen SY, Chiu HH, Wang JK, Chen CA, et al. Comparison of 3 Risk Estimation Methods for Predicting Cardiac Outcomes in Pregnant Women With Congenital Heart Disease. Circ J 2015 Jun 25; 79(7): Roos-Hesselink JW, Ruys TP, Stein JI, Thilen U, Webb GD, Niwa K, et al. Outcome of pregnancy in patients with structural or ischaemic heart disease: results of a registry of the European Society of Cardiology. Eur Heart J 2013 Mar; 34(9): Ruys TP, Roos-Hesselink JW, Hall R, Subirana-Domenech MT, Grando-Ting J, Estensen M, et al. Heart failure in pregnant women with cardiac disease: data from the ROPAC. Heart 2014 Feb; 100(3): Tanous D, Siu SC, Mason J, Greutmann M, Wald RM, Parker JD, et al. B-type natriuretic peptide in pregnant women with heart disease. J Am Coll Cardiol 2010 Oct 5; 56(15): Kampman MA, Balci A, van Veldhuisen DJ, van Dijk AP, Roos-Hesselink JW, Sollie-Szarynska KM, et al. N-terminal pro-b-type natriuretic peptide predicts cardiovascular complications in pregnant women with congenital heart disease. Eur Heart J 2014 Mar; 35(11): Bowater SE, Selman TJ, Hudsmith LE, Clift PF, Thompson PJ, Thorne SA. Long-term outcome following pregnancy in women with a systemic right ventricle: is the deterioration due to pregnancy or a consequence of time? Congenit Heart Dis 2013 Jul-Aug; 8(4): Balint OH, Siu SC, Mason J, Grewal J, Wald R, Oechslin EN, et al. Cardiac outcomes after pregnancy in women with congenital heart disease. Heart 2010 Oct; 96(20): Egidy Assenza G, Cassater D, Landzberg M, Geva T, Schreier J, Graham D, et al. The effects of pregnancy on right ventricular remodeling in women with repaired tetralogy of Fallot. Int J Cardiol 2013 Oct 3; 168(3):

149 148 Chapter Guedes A, Mercier LA, Leduc L, Berube L, Marcotte F, Dore A. Impact of pregnancy on the systemic right ventricle after a Mustard operation for transposition of the great arteries. J Am Coll Cardiol 2004 Jul 21; 44(2): Kampman MA, Balci A, Groen H, van Dijk AP, Roos-Hesselink JW, van Melle JP, et al. Cardiac function and cardiac events 1-year postpartum in women with congenital heart disease. Am Heart J 2015 Feb; 169(2): Uebing A, Arvanitis P, Li W, Diller GP, Babu-Narayan SV, Okonko D, et al. Effect of pregnancy on clinical status and ventricular function in women with heart disease. Int J Cardiol 2010 Feb 18; 139(1): Cornette J, Ruys TP, Rossi A, Rizopoulos D, Takkenberg JJ, Karamermer Y, et al. Hemodynamic adaptation to pregnancy in women with structural heart disease. Int J Cardiol 2013 Sep 30; 168(2): van Hagen IM, Roos-Hesselink JW, Ruys TP, Merz WM, Goland S, Gabriel H, et al. Pregnancy in Women With a Mechanical Heart Valve: Data of the European Society of Cardiology Registry of Pregnancy and Cardiac Disease (ROPAC). Circulation 2015 Jul 14; 132(2): Gyselaers W, Peeters L. Physiological implications of arteriovenous anastomoses and venous hemodynamic dysfunction in early gestational uterine circulation: a review. J Matern Fetal Neonatal Med 2013 Jun; 26(9):

150

151

152 Appendices A