Estrogen and endometrial carcinoma. An independent pathology review supporting original risk estimate.

Initial evidence suggested that estrogen therapy increases the risk of endometrial carcinoma. It was then suggested that some studies may have exaggerated the hazard of estrogen therapy by including patients with atypical endometrial hyperplasia among those having endometrial carcinoma. Three internationally recognized pathologists reviewed the histology slides available from the Ziel and Finkle study, which originally reported a risk ratio of 7.6 for estrogen users. At least one of the pathologists concurred with the original diagnosis in all but one case. Furthermore, all pathologists aggreed that 74 per cent (66/89) were correctly diagnosed. In the 66 patients with unanimous diagnosis, 61 per cent (40/66) had used conjugated estrogens, versus 57 per cent (54/94) in the original study. On the basis of 66 patients and 132 matched controls, the revised risk-ratio estimate is 8.1 (with a one-sided 95 per cent lower confidence limit of 4.5), validating the original estimate.