A population-based case-control study of histologically confirmed endometrial neoplasia was conducted in Toronto, Canada, between 1977 and 1978. Pathologic slide review resulted in classification of patients to diagnostic categories of endometrial carcinoma, endometrial carcinoma in situ, and endometrial hyperplasia. This report concerns the subset of patients classified as having adenomatous endometrial hyperplasia. Two age-matched neighborhood controls were selected for each case, and analyses included only those case-control sets in which the case and at least one control were concordant on menopausal status. Among postmenopausal women (87 cases and 151 controls), noncontraceptive estrogen use was a statistically significant risk factor (adjusted odds ratio (OR) = 6.7), as was heavy body weight relative to height (Quetelet index) (adjusted OR = 3.2). Long-term (5+ years) estrogen use conferred an estimated relative risk of more than 20, but use of oral contraceptives did not alter risk significantly. Among premenopausal women (62 cases and 97 controls), a high Quetelet index appeared to be a significant protective factor, even when adjusted for oral contraceptive use (adjusted OR = 0.2). While oral contraceptive use was not associated with a change in risk, long duration of use was highly, but statistically not significantly, protective against hyperplasia. The data for pre- and postmenopausal women indicate that both endogenous and exogenous estrogen exposure are important factors in the etiology of adenomatous endometrial hyperplasia.
Read full abstract