Purpose: The aim of this study was to investigate the diagnostic accuracy of liquid-based endometrial cytology, in comparison with histology.
Methods: 1987 patients scheduled for hysteroscopy were enrolled in this study. All patients proceeded sequentially through endometrial cytology, hysteroscopy and then dilatation and curettage (D&C). Cytology sampling was performed by brushing the uterus cavity using SAP-1 and the sample was prepared to liquid-based smear using SurePath technology. The slides were stained by Papanicolaou method. All cytological diagnosis was correlated with the D&C histological diagnosis.
Results: Cyto-histological correlations were possible in 1672 (89.3%) patients: in 254 (12.8%) patients the D&C was inadequate, in 75 (3.8%) patients the cytology was inadequate, and in 14 (0.7%) patients both were inadequate. In postmenopausal women, 758 of 790 cytologies (96.0%) were adequate, while 586 of 790 histologies (74.2%) were adequate. SAP-1 provided more sufficient materials for cytology than D&C for histology (P < 0.001). Taking atypical hyperplasia or worse as a positive result, the diagnostic accuracy of liquid-based endometrial cytology was 86.1%, sensitivity was estimated at 70.3%, specificity at 88.5%, positive predictive value at 48.0% and negative predictive value at 95.2%. Taking endometrial carcinoma as a positive result, the diagnostic accuracy of liquid-based endometrial cytology was 94.4%; sensitivity was estimated at 53.2%, specificity at 98.6%, positive predictive value at 79.8% and negative predictive value at 95.3%.
Conclusions: Liquid-based endometrial cytology can be considered a useful method for detecting of endometrial pathology as a first-line approach.
Keywords: Endometrial carcinoma; Endometrial cytology; Endometrial precancers; Screening.