Background: Recent advances in clinical, pathological, and genetic aspects of atrioventricular septal defects (AVSD) have set the stage for epidemiologic investigations into possible risk factors. Previous analyses of the total case group of AVSD included complete and partial subtypes without analysis of the subsets.

Methods: To address the question of possible morphogenetic heterogeneity of AVSD, the Baltimore-Washington Infant Study data on live-born cases and controls (1981-1989) was reanalyzed for potential environmental and genetic risk-factor associations in complete AVSD (n = 213), with separate comparisons to the atrial (n = 75) and the ventricular (n = 32) forms of partial AVSD.

Results: Complete and ventricular forms of AVSD had a similar proportion of isolated cases (12.2% and 15.6%, respectively, without associated extracardiac anomalies) and high rates of Down syndrome, whereas the atrial form of partial AVSD included 55% isolated cases. Trisomy 18 occurred in 22% of infants with the ventricular form, compared with <2% in the other AVSD groups. Analysis of potential risk factors revealed further distinctions. Complete AVSD as an isolated cardiac defect was strongly associated with maternal diabetes (odds ratio [OR] = 20.6; 95% confidence interval [CI] =5.6-76.4) and also with antitussive use (OR = 8.8; CI = 1.2-48.2); there were no strong associations other than maternal age among Down syndrome infants with this type of heart defect. Isolated cases with the atrial type of partial AVSD were associated with a family history of heart defects (OR = 6.2; CI = 1.4-24.4) and with paternal occupational exposures to ionizing radiation (OR = 5.1; CI = 1.4-27.4), but no risk factors were associated with Down syndrome. There were no significant associations of any risk factors in the numerically small subsets of isolated and Down syndrome cases with the ventricular form of partial AVSD.

Conclusions: These results indicate a similar risk profile of complete AVSD and the ventricular type of partial AVSD, with a possible subset of the latter due to trisomy 18. Maternal diabetes constituted a potentially preventable risk factor for the most severe, complete form of AVSD.