In women with postmenopausal bleeding and a nonreassuring transvaginal ultrasound evaluation, a nondiagnostic office endometrial sample does not rule out endometrial cancer and further endometrial sampling is advisable.

The objective is to assess among women with postmenopausal bleeding the relationship of age and time since menopause on one hand and the presence of endometrial cancer and atypical hyperplasia on the other hand. In a multicenter prospective cohort study, 614 women presenting with postmenopausal bleeding were included. Women underwent transvaginal sonography and, in cases where the endometrial thickness was >4 mm, endometrial sampling. Splines were used to assess the association between each of the continuous variables and (pre)malignancy of the endometrium. Subsequently, univariate and multivariate analysis were performed. The average age for women without (pre)malignancy was 61.7 years (SD 9.8). As malignant and premalignant cases were found to have similar age, these subgroups were merged in the analyses. Age was an independent predictor of (pre)malignancy. In women younger than 55 years, the odds ratio was 1.9 (95% CI: 1.1-3.3) for each year under 55 years of age and 1.03 (95% CI: 1.00-1.06) for each year over 55 years of age. The risk of (pre)malignancy of the endometrium was 4.9% in women less than 3 years postmenopausal versus 19.7% in women more than 20 years postmenopausal. However, in a multivariate analysis only age contributed to the prediction of risk. This study demonstrates that, in postmenopausal women with vaginal bleeding, the risk of (pre)malignancy of the endometrium is low in women under 50 years of age, increases considerably until 55 years of age, and rises only modestly with further advancing age. Future studies should explore whether these findings can be incorporated in the diagnostic work-up of women with postmenopausal bleeding.

II.