Recurrent postmenopausal bleeding: Pathological findings and predictive factors. A multicenter, prospective, observational study.

ABSTRACT

INTRODUCTION: Recurrent postmenopausal bleeding (PMB) occurs in 6%-25% of postmenopausal women who have experienced a previous episode of PMB. The question of whether recurrent PMB leads to a higher risk of endometrial cancer (EC) in comparison to a single episode of PMB is, however, controversial. Furthermore, little is known about predictive factors for recurrent PMB. MATERIAL AND METHODS: A multicenter prospective cohort study was conducted over a 5-year period in four hospitals in the Netherlands. Women with PMB undergoing endometrial sampling and aged 40 years and older were included. Occurrence of recurrent PMB was retrospectively determined. Primary outcomes included (1) the incidence of recurrent PMB and (2) differences in pathological findings between patients with a single episode vs recurrent PMB. Secondary outcomes included (1) the association between diagnosis of benign polyps at first PMB and pathological findings at recurrent PMB and (2) factors predictive for recurrent PMB. RESULTS: A total of 437 women with PMB were included, of whom 360 were at risk of recurrent PMB. With a median follow-up of 61 months (IQR (Interquartile range) 44-73), 26.4% experienced recurrent PMB. Patients with recurrent PMB were more often diagnosed with benign polyps (34.7% vs. 25.1%, p-value 0.015) and less frequently with a malignancy (5.3% vs. 17.8%, p-value 0.015), compared to patients with a single episode of PMB. Benign polyps at initial PMB were not associated with a (pre)malignancy at recurrence (OR 4.16, 95% CI 0.75-23.03). Predictive factors for recurrent PMB included use of hormone replacement therapy (HRT) (OR 3.32, 95% CI 1.64-6.72), and benign polyps at initial PMB (OR 1.80, 95% CI 1.07-3.04). CONCLUSIONS: Recurrent PMB is common in women with a previous episode of PMB. Compared to patients with a single episode of PMB, patients with recurrent PMB and benign histological outcomes at accurate workup during their first episode were less often diagnosed with malignancies and more frequently with benign polyps. Benign polyps at first PMB are predictive for recurrent PMB, but not for a higher risk of (pre)malignancy.