Not a fact anymore

Diethylstilbestrol safely prevents miscarriage and other pregnancy complications.

What we know now

DES does not prevent miscarriage. Prenatal exposure can cause rare clear-cell cancers and reproductive abnormalities in exposed daughters and other health effects in exposed children.

Why it changed

DES was prescribed from the 1940s despite weak evidence. A controlled trial found no benefit, and a 1971 case series linked an otherwise rare vaginal cancer in young women to prenatal exposure, prompting warnings and withdrawal from pregnancy use.

Status
Overturned
Category
Medicine
Accepted for
≈26 years
Accepted approximately
1940s–1971
Changed approximately
1953–1971

DES was a powerful synthetic estrogen, and from the 1940s physicians prescribed it to pregnant women in the hope that hormonal support would prevent miscarriage and other complications. The treatment spread before convincing controlled evidence showed that it worked.

That evidence became uncomfortable early. A randomized clinical trial published in 1953 found that DES did not prevent miscarriage or the other pregnancy problems for which it was being given. Yet exposure continued for years, in part because established practice changed more slowly than the evidence.

The decisive safety alarm came in 1971, when physicians reported an unusual cluster of clear-cell adenocarcinoma, a rare cancer of gland-like cells, in the vaginas of very young women. Prenatal DES exposure emerged as the striking common factor. Subsequent research documented additional reproductive and health effects, and regulators warned against its use in pregnancy.

The episode therefore involved two different failures: the drug was ineffective for its intended preventive purpose, and some harms appeared only years later in the children who had been exposed before birth.

DES also had uses outside pregnancy, including limited cancer treatment. Most prenatally exposed people do not develop clear-cell cancer; the correction is specifically to the former assumption that pregnancy use was both effective and acceptably safe.

Evidence

Sources and what they establish

Primary research

Current evidence

Related entries

Or go somewhere else: