Accuracy of outpatient endometrial biopsy in the diagnosis of endometrial hyperplasia.

Abstract

To determine the accuracy of outpatient endometrial biopsy in diagnosing endometrial hyperplasia in women with abnormal uterine bleeding. Systematic quantitative review of published medical literature. Relevant papers were identified through electronic scanning of MEDLINE (1980-1999) and EMBASE (1980-1999), manual searching of bibliography of known primary and review articles and contact with manufacturers. Studies were selected if accuracy of outpatient endometrial biopsy, in women with abnormal pre or postmenopausal uterine bleeding, was estimated compared to a reference standard, which was endometrial histology obtained by tissue sampling under anesthesia. Quality assessment and data extraction were performed in duplicate. Diagnostic accuracy was determined by pooled likelihood ratios (LR) for positive and negative test results for endometrial hyperplasia. There were 881 subjects in 8 diagnostic evaluations reported in 6 primary studies. Postmenopausal women represented 25% of the participants studied. There were 43 patients in whom outpatient sampling was inadequate. A positive test result on outpatient biopsy diagnosed endometrial hyperplasia with a pooled LR of 12.0 (95% CI 7.8-18.6) while a negative test result had a pooled LR of 0.2 (95% CI 0.1-0.3). With a positive test result, the posttest probability of endometrial hyperplasia was 57.7% (95% CI 41.1%-72.7%) while it was 2.2% (95% CI 0.9%-4.1%) with a negative test. Outpatient endometrial biopsy has modest accuracy in diagnosing endometrial hyperplasia. Therefore, additional endometrial assessment should be undertaken, especially if symptoms persist or intrauterine structural abnormalities are suspected.