Author + information
- Received December 19, 2014
- Revision received March 11, 2015
- Accepted April 9, 2015
- Published online August 1, 2015.
- Amar M. Salam, MBBS∗,
- Ebru Ertekin, BSc†,
- Iris M. van Hagen, MD†,
- Jassim Al Suwaidi, MB, ChB∗,
- Titia P.E. Ruys, MD, PhD†,
- Mark R. Johnson, MD, PhD‡,
- Lina Gumbiene, MD, PhD§,
- Alexandra A. Frogoudaki, MD, PhD‖,
- Khaled A. Sorour, MD¶,
- Laurence Iserin, MD#,
- Magalie Ladouceur, MD#,
- A. Carla C. van Oppen, MD, PhD∗∗,
- Roger Hall, MD†† and
- Jolien W. Roos-Hesselink, MD, PhD†∗ ()
- ∗Hamad Medical Corporation, Doha, Qatar
- †Erasmus Medical Center, Rotterdam, the Netherlands
- ‡Imperial College School of Medicine, Chelsea and Westminster Hospital, London, United Kingdom
- §Vilnius University Hospital Santariskiu Klinikos, Vilnius University, Vilnius, Lithuania
- ‖Attikon University Hospital, Second Cardiology Department, Athens University, Athens, Greece
- ¶Kasr AL-Aini Hospitals, Cairo University, Cairo, Egypt
- #Necker Hospital and Georges Pompidou European Hospital, Assistance publique-Hôpitaux de Paris, Paris, France
- ∗∗University Medical Center, Utrecht, the Netherlands
- ††Norwich Medical School, University of East Anglia, Norwich, United Kingdom
- ↵∗Reprint requests and correspondence:
Prof. Dr. Jolien W. Roos-Hesselink, Department of Cardiology, Office Ba 583a, Erasmus University Medical Center, PO Box 2040, 3000 CA Rotterdam, the Netherlands.
Objectives Atrial fibrillation (AF)/atrial flutter (AFL) during pregnancy in these women is associated with adverse outcome of pregnancy.
Background The incidence, timing, and consequences of AF or AFL during pregnancy in patients with heart disease are not well known.
Methods Between January 2008 to June 2011, 60 hospitals in 28 countries prospectively enrolled 1,321 pregnant women with congenital heart disease, valvular heart disease, ischemic heart disease, or cardiomyopathy in the ROPAC (Registry of Pregnancy and Cardiac Disease). We studied the incidence, onset, and predictors of AF/AFL during pregnancy and assessed the pregnancy outcome. An overview of the existing literature is provided.
Results Seventeen women (1.3%) developed AF/AFL during pregnancy, mainly in the second trimester (61.5%). Univariable analysis identified the following pre-pregnancy risk factors for AF/AFL in pregnancy: AF/AFL before pregnancy (odds ratio [OR]: 7.1, 95% confidence interval [CI]: 1.5 to 32.8); mitral valvular heart disease (OR: 6.9, 95% CI: 2.6 to 18.3); beta-blocker use (OR: 3.3, 95% CI: 1.2 to 9.0); and left-sided lesions (OR: 2.9, 95% CI: 1.0 to 8.3). Maternal mortality was higher in women with than in women without AF/AFL (11.8% vs. 0.9%; p = 0.01), although heart failure was not seen more often. Low birth weight (<2,500 g) occurred more often in women with than in women without AF/AFL (35% vs. 14%; p = 0.02).
Conclusions AF/AFL occurs in 1.3% of pregnant patients with structural heart disease with a peak at the end of the second trimester. AF/AFL during pregnancy in cardiac patients is associated with unfavorable maternal outcome and also has an impact on fetal birth weight.
- adult congenital heart disease
- atrial fibrillation
- atrial flutter
- pregnancy and heart disease
- valvular heart disease
Major cardiovascular adaptations occur in women during pregnancy (1). Cardiac output increases 30% to 50% above baseline during normal pregnancy. In early pregnancy, increased cardiac output is primarily related to the rise in stroke volume. In late pregnancy, heart rate is the major factor (2). Pregnancy is also a hypercoagulable state (3). These physiological alterations have more of an impact in patients with pre-existing heart disease (4,5).
Arrhythmia, especially supraventricular tachycardia (SVT), is a known complication during pregnancy in patients with heart disease (6,7). The incidence of SVT seems to increase during pregnancy due to hemodynamic and maybe also hormonal changes, but evidence is scarce (8). Atrial fibrillation (AF) is the most common clinically significant cardiac arrhythmia in the general population. Even though AF or atrial flutter (AFL) is rarely directly life-threatening, each increases the risk of thromboembolism and may worsen cardiac function (9–11). Starting medical treatment for AF/AFL presents a difficult choice in practice, because most antiarrhythmic drugs can be regarded as potentially harmful to the fetus (6,9).
Current literature on the occurrence of arrhythmia during pregnancy and its effect in women with cardiac disease is limited. Therefore, we studied the incidence, onset, and predictors of AF/AFL in pregnant patients with heart disease, and we report on the maternal and fetal outcomes. In addition, we present an overview of the current literature.
The European ROPAC (Registry on Pregnancy and Cardiac Disease) was established in 2008 by the European Society of Cardiology. Pregnant patients from 2007 were included retrospectively, assuming that the complete data of these patients were available and reliable. From January 2008 to June 2011, patients were included prospectively with follow-up of 6 months. In this period, 60 hospitals in 28 countries contributed to the registry, and 1,321 pregnant patients with structural heart disease were enrolled. Nonstructural heart diseases, such as arrhythmias occurring in the context of a normal heart, were excluded.
The study protocol and first results of this registry were published in 2013 (7).
Onset of AF/AFL was displayed as weeks of gestational age. Baseline characteristics before pregnancy were analyzed, including cardiac diagnosis, maternal age, nulliparity, clinical signs of heart failure, AF before pregnancy, hypertension, smoking status, beta-blocker use before pregnancy, and New York Heart Association functional class. Recorded cardiac diagnoses were congenital heart disease, valvular heart disease (VHD), cardiomyopathy, and ischemic heart disease. Countries were divided into developed or developing nations according to the International Monetary Fund Classification.
The type of cardiac lesion was classified in 3 categories for the univariable logistic regression analysis: right-sided lesions (e.g., Ebstein anomaly, tetralogy of Fallot, and pulmonary stenosis); left-sided lesions (e.g., aortic valve disease, mitral valve disease, and cardiomyopathies); and shunt lesions (e.g., atrial septal defects and ventricular septal defects). Fractional shortening <30% on echocardiography was also analyzed with the univariable logistic regression analysis.
Categorical data are presented as frequencies and percentages. Continuous data are presented as mean ± SD when considered normally distributed, as was checked using Kolmogorov-Smirnov tests. The chi-squared or Fisher exact tests were used to compare differences in categorical data between independent patient groups, whereas Student t test was used to compare differences in continuous data between independent patient groups. The intention was to adjust birth weight by a linear regression for multiple factors. However, due to the small number of cases in the AF/AFL group, adjustment was only applied for gestational age. Baseline patient characteristics associated with AF/AFL were identified with univariable logistic regression analysis. Multivariable analysis was not done because of the small number of cases. A p value of <0.05 (2-sided test) was considered statistically significant. All statistical analyses were performed using SPSS (version 21.0, SPSS Inc., Chicago, Illinois).
Of the 1,321 patients included in the registry, 17 (1.3%) developed AF/AFL during pregnancy. Of these 17, 2 had 1 episode of AF before pregnancy and experienced a new event during pregnancy, whereas 5 others had pre-existing paroxysmal AF/AFL. Baseline characteristics of patients with and without AF/AFL are shown in Table 1. Differences were found in underlying diagnoses. VHD was the most common underlying problem in patients with AF/AFL. No patients with ischemic heart disease developed AF/AFL. The incidence of AF/AFL within each cardiac diagnosis category is depicted in Figure 1.
Detailed information on the patients with AF/AFL is provided in Table 2.
Predictors and onset of AF/AFL
The results of the univariable logistic regression are shown in Table 3. It appeared that patients enrolled in centers with at least 1 AF/AFL event had different characteristics than the remaining patients did. Unfortunately, due to the low number of endpoint events, it was not possible to correct for the differences between the centers. Online Table 1 shows the differences between centers with AF/AFL events and centers without AF/AFL events. Of the univariable predictors, only pre-pregnancy AF and beta-blocker use were not different in the centers with or without AF/AFL.
In all patients, AF/AFL occurred during pregnancy and once after delivery (Figure 2). Onset of AF/AFL was mainly at the end of the second trimester.
Medication before and during pregnancy
During pregnancy, 14 patients (82%) used medication. Details concerning antiarrhythmic medication and anticoagulants are listed in Table 2.
Mode of delivery in patients with AF/AFL was vaginal in 9 (53%), by elective cesarean section in 7 (41%), and by emergency cesarean section in 1 patient because of cardiac reasons. Elective cesarean section was for cardiac reasons in 7 patients and for obstetric reasons in 1 patient.
Maternal and fetal outcome
Maternal mortality occurred in 2 of 17 patients (12%). The first patient was 29 years of age and had moderate mitral stenosis and mitral regurgitation before pregnancy. After delivery, anticoagulation was stopped because of severe bleeding. She suddenly died 1 week after delivery and was highly suspected based on clinical judgment of a thromboembolic complication. The second patient was 25 years of age and also had a combination of severe mitral stenosis and mitral regurgitation. Early spontaneous abortion occurred in this patient, and she died 6 weeks after the expulsion of the pregnancy due to sepsis, which was followed by the event of AF. Both patients did not suffer from heart failure prior to their demise.
There were no significant differences in cardiac and obstetric complications during pregnancy in the patients with versus without AF/AFL (Table 4). Hospitalization during pregnancy was needed for 12 patients (71%). All hospital admissions were for cardiac reasons.
Fetal outcome is summarized in Table 4.
This is the first study to examine in detail the incidence, predictors, and outcomes of AF/AFL in pregnant patients with heart disease. In this large prospective international registry of 1,321 patients with heart disease, 1.3% experienced documented AF/AFL, which occurred most frequently at the end of the second trimester. Pre-pregnancy predictors of AF/AFL were AF before pregnancy, mitral valve disease, beta-blocker use, and left-sided lesions. AF/AFL was associated with higher rates of maternal mortality and lower fetal birth weight.
Literature and incidence of AF/AFL
Existing data on the incidence of AF/AFL, during pregnancy in patients with heart disease is limited and based on small case series and individual case reports (Online Table 1) (12–22). Silversides et al. (17) did report on the recurrence rate of SVT during pregnancy in patients with a history of SVT. Li et al. (18) reported the prevalence of cardiac arrhythmia among 136,422 pregnant patients hospitalized in a single center. Of 226 patients who had an arrhythmia, 3 had AF, and all these patients had structural heart disease. Two of the 3 patients had a previous episode of AF and none of the patients died. Huisman et al. (23) assessed the incidence and risk factors for severe maternal morbidity and cardiovascular mortality during pregnancy, delivery, and puerperium in the general Dutch population between 2004 and 2006. Of 2,552 cases of severe maternal morbidity, 84 (3.3%) involved the cardiovascular system, among which 2 cases had AF, neither of whom died.
Congenital heart disease
We observed AF/AFL in 0.7% of CHD patients. There was 1 previous report (12) that assessed the progress and outcome of 482 pregnancies in 232 patients with CHD. AFL occurred in that study in 2 patients (0.4%).
We report on 1 patient with cardiomyopathy who developed AF in the second trimester. The patient delivered in the third trimester by cesarean section without fetal or maternal complications. There are no previous data in the literature on pregnancy in cardiomyopathy developing AF/AFL.
Valvular heart disease
AF/AFL occurred in 10 patients with mitral VHD (3%) in our study (7); 4 of them had valve surgery in the past. Available literature on AF/AFL in VHD and pregnancy is in the form of case series reports and incidence varies from 2% to 17.5% (13,14,16).
We also collected the literature regarding AF/AFL in pregnancy in women without structural cardiac disease. This is summarized in Online Table 2, which is available in the Online Appendix. In summary, these studies are mainly case reports of women experiencing AF during the third trimester, with good outcomes of both mother and fetus.
Predictors for AF/AFL
It was not surprising to find that AF before pregnancy was an independent predictor of AF/AFL during pregnancy. Silversides et al. (17) found a recurrence rate of arrhythmia during pregnancy of 50% in patients with SVT in general and 52% in patients with AF. Adverse fetal events occurred in 35% of patients with paroxysmal AF/AFL and 50% with permanent AF.
Other pre-pregnancy predictors of AF/AFL were mitral valve disease and left-sided lesions, although these associations might be explained by the fact that the centers reporting AF/AFL in these patients were predilection sites for these underlying diagnoses. The physiological changes that occur during pregnancy producing a state of high volume, high output, and low cardiovascular resistance are likely responsible mechanisms. These changes may amplify wall stress, which, combined with the previous cardiac abnormalities, causes electrophysiological imbalance and lowers the threshold for onset of arrhythmias. To our surprise, we found no correlation with heart failure.
Patients in our study who had been prescribed medication before pregnancy, in particular beta-blockers, probably had more severe cardiac disease, accounting for the increased risk of developing AF/AFL in this group. This might be related to the fact that some of the patients were administered beta-blockers before pregnancy because of previous arrhythmias.
Onset during pregnancy
Most instances of AF/AFL were found in the second trimester in our study, with a peak of occurrence between weeks 23 and 26. In contrast, previously reported cases of AF in pregnancy were commonly observed in the third trimester. Most comparable is the cohort of Silversides et al. (17), and our results do agree with theirs.
Maternal and fetal outcome
We report an increase in maternal deaths in cardiac patients with AF/AFL during or shortly after pregnancy (11.8%). This is probably attributable partly to the severity of the underlying cardiac condition and the thromboembolic risk might also play a role. As we report on 2 mortality cases only, a definitive conclusion on causative mechanisms cannot be drawn. Only 1 previous report (13) has described mortality in a patient due to heart failure.
We did not find any increase in fetal loss in women with AF/AFL. A low birth weight was more common in the offspring of patients with than those without AF/AFL. Adjusted for pregnancy duration, the absolute birth weight remained remarkably lower in patients suffering AF/AFL during pregnancy. However, it was not appropriate to statistically correct for factors such as the use of beta-blockers during pregnancy and the type of underlying cardiac disease, which might influence the birth weight as well.
Management of AF/AFL during pregnancy
In the management of patients with AF/AFL in pregnancy, several issues should be taken into consideration. First, the teratogenic potential of antiarrhythmic drugs limits the use of these agents during pregnancy. From many of these drugs, information on safety in pregnancy is lacking and most pharmaceutical companies advise not taking these drugs during pregnancy. Second, the hemodynamic alteration during pregnancy can alter the pharmacokinetics of the antiarrhythmic drugs. For instance, for digoxin, the dose needed for adequate serum levels during pregnancy is higher than for those outside pregnancy. In addition, anticoagulation needs consideration because pregnancy is a thrombogenic state (6). However, bleeding complications are also more often encountered, in particular during delivery. Anticoagulants and beta-blockers constitute the medications of first choice in the management of patients with AF/AFL in pregnancy and both were used in two-thirds of our patients. No increase in bleeding complication rates was observed in our patients on anticoagulants. Finally, electrical cardioversion is reported to be safe during pregnancy, but anesthetic management needs careful consideration in the pregnant woman. In our study, this was not reported.
To achieve a more definite management of the arrhythmia in case of poor tolerance and drug resistance, radiofrequency catheter ablation might be considered. Radiofrequency ablation for AF/AFL during pregnancy has not been reported to date. There is limited evidence from case reports and series (24,25) on radiofrequency ablation being safe in pregnant women suffering from atrial or atrioventricular re-entry tachycardia provided that an electroanatomical mapping system is used instead of fluoroscopy.
Implications for clinical practice and research
Although rare in pregnancy, the management of AF/AFL in women with heart disease poses a unique challenge to physicians. A multidisciplinary approach with close monitoring of mother and baby and timely therapy is required to optimize maternal and fetal outcomes. Despite our data, experience with the management of such arrhythmias remains limited and further studies in larger numbers of patients are still awaited. In addition, there is recent evidence that nonpregnant patients at risk for AF may have asymptomatic AF, which exposes them to the risk of ischemic stroke and other thromboembolic events (26,27). The observation that most of our patients developed the arrhythmia in the second trimester alerts us to the period of highest risk.
As with all registries, there was some missing information, which comprised 4% in our registry. Some of the previously identified predictors (e.g., hypothyroidism and stress) were neither confirmed nor contradicted in this study, because these data were not collected. Although AF/AFL is an important complication especially in women with underlying heart disease, its incidence during pregnancy is low. It is difficult to draw firm conclusions for the different cardiac diseases as the numbers of patients with AF/AFL per cardiac disease category were small. Also, the hierarchical structure (60 hospitals within 28 countries) should ideally be accounted for in the analysis. However, unfortunately, due to the low number of AF/AFL events, adequate estimates of hierarchical model parameters could not be obtained. Hence, we were not able to adjust for differences in clinical characteristics between patient populations. Interpretation of our data should be done against this background. Our findings should be seen as first results needing confirmation in larger populations.
In this large prospective international registry of women with heart disease, the incidence of AF/AFL during pregnancy was 1.3% and occurred mainly at the end of the second trimester. AF/AFL before pregnancy, mitral valve disease, beta-blocker use, and left-sided lesions were predictors of AF/AFL.
Our data suggest an increase in maternal mortality and low birth weight in patients with AF/AFL during pregnancy.
COMPETENCY IN MEDICAL KNOWLEDGE: Although AF/AFL during pregnancy has a low incidence rate of 1.3% in patients with structural heart disease, it is associated with an increased maternal mortality and low birth weight. With a peak in the second trimester, these patients should be routinely screened for this arrhythmia during that period and closely followed throughout pregnancy and delivery with a special focus on patients with history of AF/AFL before pregnancy, mitral valve disease, beta-blocker use, and left-sided lesions.
TRANSLATIONAL OUTLOOK: Further studies are warranted to examine the incidence and effects of asymptomatic AF/AFL occurring during pregnancy in patients with structural heart disease (using 24-h Holter or transtelephonic monitoring), particularly in the second trimester.
The authors thank the EORP (EURObservational Research Programme) team for their excellent support: Elin Folkesson Lefrancq; Cecile Laroche; Charles Taylor; Gerard Gracia; Viviane Missiamenou; Marème Konte; Maryna Andarala; Emanuela Fiorucci; Patti-Ann McNeill; Myriam Glémot; and Malika Manini. The investigators participating in the ROPAC on June 1, 2011, are listed in the Online Appendix.
ROPAC is included in the EORP of the European Society of Cardiology. During the course of this registry, the EORP was sponsored by the following companies: Abbott Vascular International; Amgen; Bayer Pharma; Bristol Myers Squibb; Boehringer Ingelheim; Boston Scientific International; Daiichi Sankyo; Menarini International; Merck & Co. (MSD); Novartis, Pfizer; and Servier International. If you are interested in joining this registry, please visit: http://www.escardio.org/guidelines-surveys/eorp/surveys/pregnancy/Pages/welcome.aspx.
For a list of the ROPAC investigators as well as supplemental tables, please see the online version of this paper.
ROPAC is a registry within the EORP (EURObservational Research Programme) of the European Society of Cardiology. The companies that support EORP (see the acknowledgments) were not involved in any part of the study or this report. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- atrial flutter
- supraventricular tachycardia
- valvular heart disease
- Received December 19, 2014.
- Revision received March 11, 2015.
- Accepted April 9, 2015.
- American College of Cardiology Foundation
- Hunter S.,
- Robson S.C.
- Robson S.C.,
- Hunter S.,
- Boys R.J.,
- Dunlop W.
- Siu S.C.,
- Sermer M.,
- Colman J.M.,
- et al.,
- for the CARPREG Investigators
- Siu S.C.,
- Sermer M.,
- Harrison D.A.,
- et al.
- European Society of Gynecology,
- Association for European Paediatric Cardiology,
- German Society for Gender Medicine,
- et al.
- Roos-Hesselink J.W.,
- Ruys T.P.,
- Stein J.I.,
- et al.,
- for the ROPAC Investigators
- Tan H.L.,
- Lie K.I.
- European Heart Rhythm Association,
- European Association for Cardio-Thoracic Surgery,
- Camm A.J.,
- et al.
- Autore C.,
- Conte M.R.,
- Piccininno M.,
- et al.
- Charitos E.I.,
- Stierle U.,
- Ziegler P.D.,
- et al.