Summary
Background
Although previous studies showed that pregnancy with heart disease is associated with significant complications, few focused on patients with valvular heart disease in sub-Saharan Africa.
Methods
We report maternal and foetal outcomes in 50 pregnant women with heart disease admitted to the Department of Cardiology of the University of Dakar, during an 8-year period.
Results
Rheumatic heart disease was observed in 46 women, seven of whom had previously been operated on. Among the remaining 39, 32 had mitral stenosis (isolated or associated with other valvular lesions). At admission, 36 women presented with pulmonary oedema, two with pulmonary embolism and 18 with arrhythmia. There were 17 maternal deaths (34%). Maternal death was associated with: mitral stenosis ( P = 0.03); severe tricuspid regurgitation ( P = 0.001); New York Heart Association functional class III or IV ( P = 0.001); symptoms of heart failure ( P < 0.001). A favourable maternal outcome was associated with: prior cardiac events ( P < 0.001); prior surgical valve replacement ( P = 0.03); cardiac prosthetic valve ( P = 0.03). There were 30 live births, six foetal deaths and five therapeutic abortions; nine women were lost to follow-up. Delivery was vaginal in 19 out of 30 cases and by caesarean section in 11 cases. Median gestational age at delivery was 28 weeks (range, 8–38 weeks). Five births occurred preterm. There were four stillbirths (neonatal mortality, 7.6%).
Conclusions
Heart disease severely impacts maternal and foetal outcome in our study. Pregnant women who underwent appropriate valve replacement before pregnancy had a better prognosis.
Résumé
Contexte
Les études antérieures ont montré que la grossesse est associée chez les femmes atteintes de cardiopathie à un risque de complications cardiaques maternelles, fœtales et néonatales. Peu d’entre elles ont cependant porté sur les complications gravidiques des cardiopathies rhumatismales en Afrique subsaharienne.
Méthodes
Nous rapportons l’évolution maternelle et fœtale des grossesses de 50 femmes atteintes de cardiopathies (âge moyen 28,4 ans ; 18–43 ans) admises dans le service de cardiologie de l’université de Dakar, au cours d’une période de huit ans.
Résultats
Une cardiopathie rhumatismale est observée chez 46 femmes, dont sept ont été opérées auparavant. Une sténose mitrale, isolée ou associée à une autre atteinte valvulaire, est présente chez 32 des 39 autres femmes. À l’admission, 36 femmes ont un œdème pulmonaire, deux une embolie pulmonaire, 18 une arythmie, Onze sont asymptomatiques. Un décès maternel survient dans 17 cas (34 %), lié à une insuffisance cardiaque ( n = 8), à une embolie pulmonaire ( n = 3), un choc cardiogénique ( n = 2) ou septique ( n = 1), une hémorragie ( n = 2). Les facteurs associés à la mortalité maternelle sont la sténose mitrale ( p = 0,03), l’insuffisance tricuspide sévère ( p = 0,001), une classe fonctionnelle NYHA III ou IV ( p = 0,001), des symptômes d’insuffisance cardiaque ( p < 0,001). À l’inverse, les facteurs associés au pronostic maternel favorable sont les antécédents d’événements cardiaques préalables ( p < 0,001), de remplacement valvulaire ( p = 0,03) et la présence d’une prothèse valvulaire ( p = 0,03). Les 50 grossesses se terminent par la naissance de 30 nouveau-nés vivants, six morts fœtales et cinq avortements thérapeutiques. Neuf femmes sont perdues de vue. L’accouchement a lieu par voie naturelle dans 19 cas sur 30 (avec forceps dans trois cas) et par césarienne dans 11 cas. La médiane d’âge gestationnel à l’accouchement est de 28 semaines (8–38 semaines), avec cinq accouchements prématurés et quatre mort-nés. La mortalité néonatale est de 7,6 %. La médiane de poids à la naissance des 26 nouveau-nés survivants est de 2366 g (1350–3100 g).
Conclusion
La présence d’une cardiopathie impacte sévèrement le pronostic maternel et fœtal de la grossesse en Afrique subsaharienne, lorsqu’une prise en charge appropriée comprenant traitement médical et correction chirurgicale préalable, prévention, planning familial, disponibilité d’une salle de cathétérisme permettant la valvuloplastie percutanée, n’est pas accessible. À l’inverse, les femmes qui ont bénéficié d’un remplacement valvulaire dans leurs antécédents ont un meilleur pronostic gravidique.
Abbreviations
NYHA
New York Heart Association
VHD
valvular heart disease
Background
Circulatory changes during pregnancy may result in adverse maternal and foetal outcomes in women with heart disease. Circulating blood volume, cardiac output and heart rate increase progressively after the first trimester, and worsen during the uterine contractions of labour and subsequent delivery .
Previous studies performed in Western countries have shown that pregnancy in women with heart disease is associated with significant cardiac and neonatal complications, despite state-of-the-art obstetric and cardiac care . Risk factors associated with an unfavourable outcome have been recognized, and include older age, tobacco smoking, multiple gestations, anticoagulant therapy, poor functional class, cyanosis and left heart obstruction . However, only a few reports have focused specifically on pregnancy outcome in women with valvular heart disease (VHD) , especially in lower-income countries .
Sub-Saharan Africa is characterized by a high prevalence of rheumatic fever and rheumatic VHD. Prevalence of rheumatic heart disease has been estimated at two to three cases per 1000 among school-age children in Africa who underwent clinical screening and was recently reported to be about 10-fold higher if they also had an ultrasound examination . The spectrum of the disease in sub-Saharan Africa also includes severe valvular disease in patients who have previously been operated on, in some of those receiving anticoagulant therapy, and in many others who have declined surgery, mainly because of financial constraints. In many instances, lack of financial support, lack of information about the risks of pregnancy in women with heart disease, and social and cultural drawbacks preclude any appropriate prevention strategy. As a result, management of pregnancy in such patients remains a medical challenge in most low-income African countries.
In this study, we report maternal and foetal outcomes in 50 pregnant women with heart disease admitted to the Department of Cardiology of the University of Dakar, Senegal.
Methods
The study was a retrospective evaluation of pregnancy outcome in women with heart disease. All cases were followed during pregnancy, labour and delivery at the University Hospital of Dakar, Senegal, between February 1996 and February 2004. Exclusion criteria included: therapeutic abortion for non-cardiac reasons; miscarriage (foetal loss before 20 weeks’ gestation); hypertensive heart disease; and peripartum cardiomyopathy.
Clinical data were recorded at the first prenatal visit, including: age; occupational, educational and marital status; gestational age; parity status; cardiac conditions, prior cardiac events and therapy; NYHA functional class; cyanosis; comorbid conditions; and anaemia.
Treatment and outcome data were obtained during hospitalization. Prepartum, peripartum and postpartum complications were grouped into cardiac, neonatal or obstetric events. The mode of delivery and obstetric complications were documented.
Data were analysed to evaluate maternal and foetal outcomes. Criteria to evaluate maternal outcome included: dyspnoea; NYHA functional class; congestive heart failure; and new onset or exacerbated arrhythmia. Neonatal complications were defined as: premature birth (< 37 weeks’ gestation); small-for-gestational-age birth weight (< 10th percentile for gestational age); neonatal and respiratory distress syndrome using the Apgar score; foetal death (> 20 weeks’ gestation before death); and neonatal death (occurring between birth and age 28 days).
Statistical analysis
Continuous values are presented as means (standard deviations) and were compared using a non-paired t test. Comparisons of baseline characteristics and maternal outcome were made using the Chi 2 test. Statistical significance was accepted at the 95% confidence level ( P < 0.05).
Results
During the 8-year study (February 1996 to February 2004), 50 pregnant women with heart disease were included, accounting for 2.7% of patients hospitalized in the Department of Cardiology during the same period. The mean age of the women was 28.4 years (range: 18–43 years). Forty-five women (90%) had low educational status (no school or primary school only) and low socioeconomic status (income less than 2.5 $ per day); only three women (6%) were engaged in any professional occupation.
Previous heart disease was known before pregnancy in 37 women (74%), with a mean follow-up of 6.5 years (range: 4 months to 30 years).
Rheumatic VHD was the predominant cardiac condition, observed in 46 women (92%), of whom seven (16%) had previously been surgically treated with prosthetic mechanical valve replacement. Among the remaining 39 women with non-surgically-corrected VHD, 32 had mitral stenosis (76%), including seven isolated forms of mitral stenosis, three associated with mitral regurgitation and 22 associated with any other valvular lesions ( Table 1 ).
