Diagnostic value of micro-histology in endometrial brushing

The efficacy of direct endometrial sampling by brushing in the collection of adequate and representative material is evaluated.

From January 1 2008 to October 31 2009, 195 women (age 29-82, mean 56, 113 postmenopausal), underwent endometrial brushing with Endoflower. All samplings were performed in an outpatient setting. 137 patients had abnormal uterine bleeding (70 postmenopausal), 25 had asymptomatic endometrial thickening (> 4 mm), 6 had atypical endometrial cells on pap-smear, 9 patients needed preoperative controls for uterine prolapse, 11 were treated with tamoxifen and 7 had other problems. The samples were fixed in a solution containing alcohol, water, EDTA and KCO3, and centrifuged. The supernatant was filtered and the pellet embedded in paraffin.

All patients reported that the technique was painless. Three cases suffered from shock. In 29 cases (15%), the sampling procedure was difficult due to cervical stenosis. A cellular sample large enough to prepare a cell-block was obtained in all cases. In 27 cases (14%), the sample was non-diagnostic. Cases were categorized as non-pathologic (negative) or pathologic (atypical and carcinoma). The correlation between cyto-histology on samples obtained with brushing and histology on biopsy or surgical specimen was possible in 46 cases (24%), with a diagnostic concordance of 93%. The rate of inadequate biopsies was 27% (8/30). 13 of 15 malignant neoplasias (2 carcinosarcomas, 13 endometrioid adenocarcinomas) were correctly diagnosed in samples collected with Endoflower. The sensitivity was 87% and specificity was 96%, with a positive predictive value of 92% and a negative predictive value of 90%.

Endometrial direct sampling with the Endoflower device in an outpatient setting is well tolerated and well accepted by the gynaecologist. This sampling procedure allows preparation of cell-blocks. Endometrial cyto-histology is less expensive and invasive than other procedures and it could therefore be used in association with transvaginal sonography, even in institutions where liquid-based cytology is not in use.