Prenatal diagnosis and cesarean section in a large, population-based birth defects registry.

Objective. To describe the patterns of cesarean section (CS) and vaginal delivery by type of birth defect and determine whether prenatal diagnosis predicts a higher or lower likelihood of CS for selected defect categories. Methods. Data from a large population-based registry were analyzed to determine percentages of vaginal versus CS delivery for each of 49 categories of birth defects. Odds ratios and statistical significance were computed to determine if a record of prenatal diagnosis (PND) predicted delivery mode. Cases were liveborn children with any of these defects born in Texas between 1997 and 2005. Results. Forty-three percent of infants in the study were delivered by CS, with a range of 25.3% (aniridia) to 62.4% (spina bifida). A record of prenatal diagnosis of the primary assigned birth defect was found in 43.0% of all records but varied substantially by defect category. PND significantly predicted higher CS percentages for spina bifida without anencephaly, encephalocele, hydrocephaly, transposition of the great vessels, ventricular septal defect, pulmonary valve atresia/stenosis, craniosynostosis, diaphragmatic hernia, gastroschisis, and trisomy 21. Vaginal delivery was predicted by PND of anencephaly, agenesis, aplasia, or hypoplasia of the lung, renal agenesis or dysgenesis, and trisomy 18. Conclusion. Texas children with birth defects are more likely to have been delivered by CS than the population in general. For several types of defects, prenatal diagnosis is predictive of higher odds of CS; for others, especially fatal defects, PND predicts lower CS likelihood.