Abstract

The incidence of endometrial adenocarcinoma in asymptomatic women is low. Nevertheless, some of these women might require endometrial surveillance. In this study, we evaluated the accuracy of liquid-based endometrial cytology compared to biopsy in asymptomatic postmenopausal women. Three hundred twenty women scheduled for hysteroscopy were enrolled for this study. After hysteroscopy, patients were submitted to endometrial cytology and to biopsy. Two hundred ninety-three (92%) women had sonographically thickened endometrium (>5 mm), 53 (17%) were on tamoxifen, and 16 (5%) were on hormonal substitutive treatment. The evaluation of the biopsies determined that six (2%) women had adenocarcinoma, one (<1%) had adenomatous atypical hyperplasia, and eight (3%) had simple nonatypical hyperplasia. Endometrial cytology evidenced 5 (2%) neoplastic cases, 2 (<1%) hyperplastic with atypia cases, and 25 (8%) hyperplastic without atypia cases. Two hundred twenty-two biopsies (69%) and 17 (5%) cytologies were inadequate. One adenocarcinoma and one simple nonatypical hyperplasia were underrated by cytology resulting, respectively, as atypical hyperplasia and as negative. Four cases were false positive (simple nonatypical hyperplasias on cytology, negative on biopsy). The sensitivity and specificity were estimated, respectively, at 94% and 95%; the positive and negative predictive value were estimated, respectively, at 80% and 99%. Endometrial cytology provided sufficient material more often than biopsy (P < 0.01). We suggest to introduce liquid-based endometrial cytology in the management of some subpopulations of asymptomatic postmenopausal women. Particularly, the combination of liquid-based endometrial cytology and transvaginal sonography may improve their diagnostic accuracy and reduce unnecessary more invasive and expensive procedures.