Cardiovascular birth defects and antenatal exposure to female sex hormones: a reevaluation of some base data.

A re-evaluation of the base data as reported by Heinonen et al. from the Drug Epidemiology Unit of the Boston Collaborative Perinatal Project (CPP) was undertaken in order to examine particularly three matters which were not fully considered in the publication. These were, first, the timing of administration of sex hormones during the index pregnancy, which is relevant to determining whether any statistical association reported between sex hormone exposure and malformations could be causal; second, the incidence of serious maternal vaginal bleeding in early pregnancy, which could be an indication of threatened abortion, which in turn is associated with an increased malformation rate, and in addition is an indication for sex hormone administration; and third, the incidence of malformations or other adverse outcome in previous pregnancies, which, if present, might play a material role in the risk of malformation in the index pregnancy. Examination of the records of the 19 cases described by Heinonen et al. as hormone-exposed/cardiac-malformed revealed that no preparation containing hormone was administered in two patients, that five cases were given hormones too late in the index pregnancy to have any effect on cardiac organogenesis (which by general consensus begins on day 19 and ends at the latest on day 50 of gestation), that two cases were given hormones too early and two cases had Down's Syndrome. Thus eight children were exposed to hormones during the critical period of cardiac organogenesis, out of 17 actual hormone takers (47%). The description of vaginal bleeding did not allow any conclusions regarding differential rates of threatened abortion. It was found that the incidence of major malformations was 17% in the index group and 4% in the non-malformed group. This suggests that the hormone-exposed patients who gave rise to children with cardiac malformations were a highly selected group. The re-evaluation therefore reveals that the incidence of exposure to sex hormones during the critical period of cardiac organogenesis was not significantly different statistically in those women whose children had cardiac lesions as compared to those without such lesions. It is suggested that the result of this re-evaluation should direct the attention of epidemiologists to the quality of their base data. Re-examination of the base data of the Boston CPP does not support their reported association between the exposure to female sex hormones during pregnancy and the occurrence of cardiac malformations.