The imaginative study of Fixler and Threlkeld ('98) evaluated maternal exposures within a Down syndrome infant population, comparing those with cardiac malformations (''cases'') to those without (''controls''). They suggest that their failure to detect any environmental risk factors could indicate that cardiac malformation in Down syndrome is a direct result of chromosomal duplication.
This interpretation is highly consistent with and complementary to the results of the Baltimore-Washington Infant Study, which were published just as the above article was being submitted (Ferencz et al., '97). In separate statistical models, we compared data on infants who had atrioventricular septal defects with and without Down syndrome to randomly selected population controls representative of the birth cohort. We found only a single major risk factor in the Down syndrome group (n ϭ 190): advanced maternal age. In contrast, many risk factors (familial, medical, and environmental) were found for the nonchromosomal cases (n ϭ 76), most notably maternal diabetes (odds ratio, 9.3; 95% confidence interval, 3.1-28.2) (Ferencz et al., '97). The study also indicated that infants of mothers with diabetes and infants with trisomies have a similar diagnostic spectrum of major cardiac defects, possibly indicating common disturbances of early cardiogenesis (Ferencz et al., '97).
Thus our data support the cautious interpretation by Fixler and Threlkeld ('98) of their findings, and we join them in emphatically urging multicenter, multidisciplinary investigations which will identify the specific morphogenetic alterations which account for the consistent findings of our two epidemiologic studies.
This is an opportune moment to study the origins of Down syndrome and abnormal cardiogenesis.