The Swedish Child Cardiology Registry (CCR) and the Swedish Medical Birth Registry (MBR) were used to investigate a possible association between maternal smoking during pregnancy and congenital heart defects. Among 1,413,811 infants born in 1983-1996 with known smoking exposure in early pregnancy, 3384 infants with congenital heart defects were selected (458 term infants with persistent ductus arteriosus (PDA) identified from MBR or CCR, and 2926 infants with other heart defects, identified from CCR). After controlling for year of birth, maternal age, parity, and educational level, a weak, statistically significant association between all heart defects and maternal smoking was found (odds ratio (OR): 1.09; 95% confidence interval (CI): 1.01-1.19). When infants with isolated PDA were removed from the case group the significance disappeared (OR: 1.07; 95% CI: 0.98-1.17). For truncus abnormalities (OR: 1.23; 95% CI: 1.02-1.49), atrial septal defects (OR: 1.63; 95% CI: 1.04-2.57), and PDA (OR: 1.30; 95% CI: 1.05-1.62), a rather strong effect of maternal smoking was indicated. The increased OR for PDA remained when gestational duration and intrauterine growth was also controlled for. Further research based on independent data sets is needed to conclude whether the risk estimates for maternal smoking are true and truly differ between the groups. The classification system used (with 24 classes) is described in enough detail to permit a repetition of the study.