Endometrial blood flow mapping using transvaginal power Doppler sonography in women with postmenopausal bleeding and thickened endometrium

Objective To evaluate the role of transvaginal power Doppler sonography to discriminate between benign and malignant endometrial conditions in women presenting with postmenopausal bleeding and thickened endometrium at baseline sonography. Methods Ninety-one postmenopausal women (median age, 58 years; range, 47-83 years) presenting with uterine bleeding and a thickened endometrium (greater than or equal to 5-mm double-layer endometrial thickness) on transvaginal sonography were included in this prospective study. Endometrial blood flow distribution was assessed in all patients by power Doppler immediately after B-mode transvaginal sonography. Three different vascular patterns were defined: Pattern A: multiple-vessel pattern, Pattern B: single-vessel pattern and Pattern C: scattered-vessel pattern. Histological diagnoses were obtained in all cases. No patient taking tamoxifen citrate or receiving hormone replacement therapy was included. Results Histological diagnoses were as follows: endometrial cancer: 33 (36%), endometrial polyp: 37 (41%), endometrial hyperplasia: 14 (15%), endometrial cystic atrophy: 7 (8%). Blood flow was found in 97%, 92%, 79% and 85% of cases of carcinoma, polyp, hyperplasia and endometrial cystic atrophy, respectively. A total of 81.3% of vascularized endometrial cancers showed Pattern A, 97.1% of vascularized polyps exhibited Pattern B and 72.7% of vascularized hyperplasias showed Pattern C. Sensitivity and specificity for endometrial cancer were 78.8% and 100%. For endometrial polyp these respective values were 89.2% and 87% and for hyperplasia they were 57.1% and 88.3%. Conclusions Transvaginal power Doppler blood flow mapping is useful to differentiate benign from malignant endometrial pathology in women presenting with postmenopausal bleeding and thickened endometrium at baseline sonography. Copyright (C) 2003 ISUOG. Published by John Wiley Sons, Ltd.