Diagnostic workup of patients with benign or inconclusive reports on office endometrial biopsy after first episode of postmenopausal blood loss
Postmenopausal bleeding carries a risk of endometrial cancer, and office endometrial sampling (OES) is often the first diagnostic approach. Whether further diagnostic procedures such as hysteroscopy or saline infusion sonography (SIS) are required to rule out polyps and malignancies is uncertain. The objective of this study is to evaluate whether clinicians perform further diagnostic tests when OES is benign or inconclusive. Secondary outcomes include the patient characteristics associated with the likelihood of performing further diagnostic tests and the incidence of endometrial intraepithelial neoplasia and endometrial malignancy. We performed a multicenter, prospective cohort study, evaluating patients with a first episode of postmenopausal bleeding, with an endometrial thickness of >4 mm on 2D vaginal sonography and a benign or inconclusive result of an OES. Patients who underwent further diagnostic or therapeutic procedures (hysteroscopy or SIS) were compared to patients who did not. Univariate and multivariate analyses were performed to identify predictive factors for additional diagnostic procedures and the diagnosis of endometrial malignancy. Of the 350 eligible patients, 197 (56 %) underwent further diagnostic procedures. These patients had a thicker endometrium (median 8.6 mm vs. 6.0 mm; p < 0.001), more frequent suspicion of intracavitary abnormalities (34.0 % vs. 14.4 %; p < 0.001), and more frequent insufficient aspiration samples (20.8 % vs. 11.8 %; p = 0.025) compared to patients who received expectant management. Multivariate regression analysis confirmed these findings. The underlying risk of malignancy was 2.7 % with benign samples and 6.8 % with insufficient samples. Overall, malignancy incidence was higher in those who underwent further diagnostic workup (p = 0.04). In this prospective study, just over half of patients underwent further diagnostic procedures. Endometrial thickness, suspicion of intracavitary abnormality or an insufficient endometrial sample were predictive for further diagnostic workup. The incidence of malignancy was higher in patients who underwent diagnostic workup, which argues for a risk-based strategy in clinical decision-making. Central study approval was obtained at the Erasmus MC (MEC 2015-740). The study was registered in the Dutch trial register (www.onderzoekmetmensen.nl, NL7608).