A randomized double-blind controlled trial of the value of stilboestrol therapy in pregnancy: long-term follow-up of mothers and their offspring

Abstract

In the early 1950s, a randomized, double-blind, controlled trial of the value of prophylactic stilboestrol therapy given antenatally to reduce the incidence of late pregnancy toxaemia and to improve perinatal mortality was conducted at University College Hospital, London. Women expecting their first baby were allocated to one or other of two groups. Those in the stilboestrol group started treatment at the 12th week of pregnancy on average and received a mean dose of about 11.5 g of the drug while those in the control group received placebo tablets. In spite of the fact that the original trial documentation was lost, it was possible to be fairly certain which was the treated group and follow-up data from 650 mothers and 660 offspring were obtained from death certificates, cancer registrations and questionnaires sent to general practitioners. We found no indication of any harmful long-term effect of stilboestrol exposure during pregnancy on the mothers--in particular 10 out of 331 women in the untreated group and 9 out of 319 women in the treated group were found to have developed breast cancer. Amongst the daughters, those in the treated group suffered an excess of minor benign lesions of the cervix uteri and an excess (not statistically significant) of unfavourable pregnancy outcomes. None of the daughters had developed clear cell adenocarcinoma of the vagina or cervix uteri. Amongst the sons, we discovered no evidence of any significant excess of genital tract disorders or of impaired reproductive performance in the treated group but one son developed a (fatal) teratoma of the testis. Unexpectedly, psychiatric disease (especially depression and anxiety) was reported by general practitioners about twice as often in the treated group offspring (sons and daughters) as in the untreated group. This result cannot be due to bias, and is unlikely to be due to confounding or chance, and may thus represent an adverse effect of exposure to stilboestrol in utero.