RESUMO
OBJECTIVES: To assess associations between maternal smoking and congenital heart defects (CHDs) in offspring. STUDY DESIGN: We performed a retrospective case-control study using data for cases of CHD (n = 8339) and nonmalformed controls (n = 11 020) from all years (1997-2011) of the National Birth Defects Prevention Study. Maternal self-reported smoking 1 month before through 3 months after conception was evaluated as a binary (none, any) and categorical (light, medium, heavy) exposure. Multivariable logistic regression was used to estimate aOR and 95% CIs. Stratified analyses were performed for septal defects according to maternal age, prepregnancy body mass index, and maternal race/ethnicity. RESULTS: Multiple CHDs displayed modest associations with any level of maternal periconceptional smoking independent of potential confounders; the strongest associations were for aggregated septal defects (OR, 1.5; 95% CI, 1.3-1.7), tricuspid atresia (OR, 1.7; 95% CI, 1.0-2.7), and double outlet right ventricle (DORV) (OR, 1.5; 95% CI, 1.1-2.1). Tricuspid atresia and DORV also displayed dose-response relationships. Among heavy smokers, the highest odds were again observed for tricuspid atresia (aOR 3.0; 95% CI, 1.5-6.1) and DORV (aOR 1.5; 95% CI, 1.1-2.2). Heavy smokers ≥35 years old more frequently had a child with a septal defect when compared with similarly aged nonsmokers (aOR 2.3; 95% CI, 1.4-3.9). CONCLUSIONS: Maternal periconceptional smoking is most strongly associated with septal defects, tricuspid atresia, and DORV; the risk for septal defects is modified by maternal age.