Detection of occult endometrial carcinoma.

Endometrial carcinoma is commonly diagnosed as a consequence of abnormal uterine bleeding. In a study published in 1962, it was documented that endometrial cancer may be detected by vaginal pool smears in asymptomatic postmenopausal women. As a consequence of these observation, a systematic search for occult endometrial carcinoma was initiated in 1979, supported by a contract from the National Cancer Institute. The techniques used in this study and the problems encountered in the diagnosis of occult endometrial carcinoma will be discussed. Within 3-1/2 years of this study encompassing 2,586 peri- or postmenopausal women 16 occult endometrial carcinomas were discovered by direct endometrial sampling. Two carcinomas, missed on initial screening, were subsequently documented in this cohort, for a prevalence rate of 6.96 per 1,000. The incidence data based on follow-up examination of 1,754 women was 1.71 per 1,000 woman-years. An elaborate epidemiologic questionnaire was evaluated. Contrary to some prevailing views, obesity, hypertension, and diabetes failed to reach statistical significance as risk factors. The only risk factor of statistical value was delay in the onset of menopause past age 49, observed in about 50% of the cohort. It was noted that the administration of estrogens to women in Quetelet Index groups below means was more likely to be associated with carcinoma than in women in higher Quetelet groups but, again, the difference was not statistically significant. It was noted that in spite of an active search for endometrial hyperplasias, the rate of these lesions was nearly identical to the prevalence and incidence rates for carcinoma. It was postulated that some or perhaps most, endometrial carcinomas in postmenopausal women are not preceded by hyperplasia but originate ab initio in the endometrium.