Methods

Women (≥16 years of age) with TGA who had undergone ASO at 2 large pediatric surgical centers (Hospital for Sick Children, Toronto, Ontario, Canada, and Boston Children’s Hospital, Boston, Massachusetts) were identified (n = 74). Medical records were reviewed to determine women who had a pregnancy. Baseline demographics, underlying cardiac anatomy, surgical details pertaining to the ASO and other surgeries, and cardiac status before or at the time of the first antenatal visit were retrospectively obtained by chart review. Echocardiographic data were obtained from the last available report before pregnancy, from echocardiograms done during pregnancy, and within 12 months after pregnancy if possible. Left ventricular ejection fraction (EF), severity of aortic valve regurgitation, and diameter of the aortic root at the level of the sinus of Valsalva were recorded. A left ventricular EF <55% was considered abnormal according to American Society of Echocardiography recommendations. Severity of aortic regurgitation was defined according previously published guidelines and was categorized as normal, mild, moderate, or severe.

Adverse maternal cardiac events during pregnancy, age at gestation, method of delivery, birth weight, and fetal and neonatal adverse events were obtained from cardiac and obstetric charts. Adverse maternal cardiac events of interest included tachyarrhythmias and bradyarrhythmias requiring treatment, pulmonary edema (diagnosed by chest x-ray), myocardial infarction, or other thromboembolic events. When possible, fetal and neonatal adverse events were also obtained from chart review and included fetal (≥20 weeks of gestation) or neonatal death (within 28 days after birth), premature birth (delivery <37 weeks of gestation), small-for-gestational-age birth weight (<10th percentile of gestational age), intraventricular hemorrhage, respiratory distress, or congenital heart disease in the newborn.

Results

Of 74 women of child-bearing age in this cohort, 12% (9 of 74) had 17 pregnancies from August 2000 to August 2009. Prepregnancy characteristics of the 9 women are presented in Table 1 . Two women had a right ventricular to pulmonary artery conduit and 1 woman had a pulmonary homograft. In the remaining 6 women the Lecompte maneuver had been performed at the time of surgery. Six women (67%) had cardiac surgery since the initial ASO. The most common indication for surgery was for relief of right ventricular outflow tract obstruction and pulmonary artery stenosis (n = 5). Six women (67%) had clinically important ventricular or valve lesion (including a mechanical mitral valve) before the index pregnancy.

There were 4 first-trimester miscarriages (1 molar pregnancy, 1 incomplete abortion, 1 ruptured ectopic pregnancy, and 1 spontaneous abortion). The remaining 13 pregnancies were delivered at term. Mean maternal age at time of delivery was 22 ± 4 years (range 17 to 32). Fifteen percent (2 of 13) of pregnancies were complicated by an adverse maternal cardiac event. One woman underwent a subaortic stenosis resection and mitral valve repair 1 year after her ASO. She was left with mitral regurgitation and required a mechanical mitral valve replacement in childhood. This woman developed postpartum valve thrombosis after anticoagulants were held at the time of delivery and required a mitral valve replacement 3 weeks postpartum. During surgery, she had a cardiac arrest and was successfully resuscitated but was left with mild to moderately impaired left ventricular systolic function. One woman who had mild left ventricular systolic dysfunction before pregnancy had recurrent nonsustained ventricular tachycardia at 16 weeks of gestation, which was treated with amiodarone. There were no maternal deaths during pregnancy. None of the women developed pulmonary edema and there were no ischemic or thrombotic events during pregnancy. One woman had progression of supravalvar pulmonary stenosis over the course of her 3 pregnancies and required subsequent balloon angioplasty 12 months after her third pregnancy.

Serial echocardiographic data were available in 69% (9 of 13) of the pregnancies. In 2 women, there was a decrease in left ventricular systolic function during or after pregnancy. In 1 women with mild left ventricular systolic dysfunction before pregnancy, there was an increase in left ventricular end-diastolic diameter (46 mm before pregnancy vs 55 mm at week 35 of gestation) and a decrease in left ventricular EF (48% prepregnancy vs 35% at 35 weeks of gestation). In the woman with postpartum mitral valve thrombosis described earlier, left ventricular EF decreased after her mitral valve operation (EF 60% before pregnancy vs 48% postoperatively). None of the women had moderate or severe aortic regurgitation before or after pregnancy. In 2 women, the aortic root was dilated before pregnancy (sinus of Valsalva diameters 37 and 44 mm) with no significant increase during or after pregnancy.

There were no preterm deliveries. There were no fetal or neonatal deaths. Birth weight was not available for 6 of the births. In those with birth weights available (n = 7), mean birth weight was 3.3 ± 0.5 kg (range 2.3 to 4.0). One child had a birth weight that was small for its gestational age (2,330 g, delivered at 38 weeks of gestation). The mother of this child had several risk factors associated with low birth weight including impaired left ventricular function, arrhythmia during pregnancy, treatment with β-blocker therapy during pregnancy, and smoking throughout pregnancy. One child had a restrictive ventricular septal defect. One child required surgery for pyloric stenosis.