Women whose mothers took diethylstilbestrol (DES) during pregnancy are more likely than women who were not exposed to the drug in utero to experience delays in conceiving and are less likely ever to become pregnant. When DES-exposed women conceive, their risk of having a first-trimester spontaneous abortion is elevated by about 30%, and their odds of having a preterm birth, second-trimester pregnancy loss or ectopic pregnancy are 3-4 times those of unexposed women. These findings are based on data from what the researchers describe as "the largest number of women with documented in utero exposure to DES to be observed systematically throughout much of their reproductive life span."1
The analyses included 3,373 women who were exposed to DES and 1,036 unexposed controls who were enrolled in one of two longitudinal projects. Most--3,140 women with DES exposure and 826 controls--were part of a multisite national study begun in the mid-1970s. Exposed women in this study either had been identified through a review of medical records, had been referred to the project by a physician or had come to the project on their own; controls were matched by age and by their mothers' age to women identified through record review. The remaining 233 exposed women and 210 controls were part of a study that has followed the daughters of women who participated in a clinical trial of DES conducted in Chicago in the 1950s.
All of the women completed a questionnaire that was mailed to them in 1994, asking about their health history, pregnancies and pregnancy outcomes. On the basis of similarities of results among subgroups, the researchers combined the 1,683 women identified through record review in the multisite study and those from the Chicago study into one group for analysis, and the 1,690 women from the physician-referred and walk-in subgroups of the multisite study into another group. Thus, in their analyses of pregnancy outcomes, they compared three groups of women: "record-review," "physician and self-referral," and controls. The investigators note that since referrals probably were the result of reproductive problems or gynecologic abnormalities, the main emphasis is on the women making up the record-review group.
All three groups were similar in age (on average, 40-43 years), were overwhelmingly white (97-98%) and were predominantly married (72-74%). Women in the record-review group were slightly older at first pregnancy than those in the other two groups (26 vs. 25 years), and they started smoking at a later age than controls (18.4 vs. 17.8 years).
DES-exposed women were significantly more likely than others to have had difficulty conceiving: Some 32% of those identified through record review (and 37-40% of those who were referred by a physician or who walked in) had experienced a delay of 12 months or more when trying to become pregnant, compared with 19% of controls. In addition, the proportion who had ever been pregnant was lower among exposed women identified through record review or referred by a physician (74-75%) than among those in the control group (81%).
For women's first pregnancies and for all pregnancies combined, the analysts calculated relative risks to compare the occurrence of selected outcomes among DES-exposed women and those whose mothers had not taken the drug, while adjusting for women's age and the study center where they were originally followed up. According to these calculations, exposed women in the record-review group had a significantly reduced risk of having their first pregnancy end in a full-term live birth (risk ratio, 0.8). Furthermore, compared with controls, these women were 2.0 times as likely to have a spontaneous abortion, 3.1 times as likely to have a preterm birth (i.e., before 37 weeks' gestation) and 5.3 times as likely to have an ectopic pregnancy.
In their first pregnancy, women in the physician and self-referral group had poorer outcomes than those in the record-review group. Their risk of having a full-term live birth, relative to that of controls, was lower (0.6), and their chances of having a preterm birth or ectopic pregnancy were higher (5.0 and 9.2, respectively). While their risk of spontaneous abortion was similar to that of women in the record-review group (1.9), they had an added risk of stillbirth (6.3).
The analysis of adverse outcomes in any pregnancy yielded similar results for both groups of women who were exposed to DES. These women had a slightly reduced risk of ever having a live birth (0.95-0.96), and those who had any live births had fewer than controls. Women whose mothers took the drug had elevated risks of preterm birth (2.9 in the record-review group and 3.7 in the physician and self-referral group), spontaneous abortion in the first 14 weeks of pregnancy (1.3 and 1.6, respectively), pregnancy loss during weeks 15-27 (4.3 and 5.8) and ectopic pregnancy (3.8 and 5.5). Exposed women in the physician and self-referral group were less likely than controls to have had an induced abortion (0.7).
The analysts note that "even if... DES was no longer used in pregnancy after 1971 in the United States (which is not actually the case), there are still many DES-exposed women of reproductive age." Therefore, they conclude, "it is important for obstetrician-gynecologists to be aware of the consequences of DES exposure in utero on pregnancy outcome."--D. Hollander
REFERENCE
1. Kaufman RH et al., Continued follow-up of pregnancy outcomes in diethystilbestrol-exposed offspring, Obstetrics & Gynecology, 2000, 96(4):483-489.