A prospective study was performed to assess the efficacy of sonohysterography (SHG) in identifying endometrial pathologies among asymptomatic, postmenopausal breast cancer patients treated with tamoxifen. In this study the uterine cavity of 68 such patients with endometrial thickness of > or = 8 mm was prospectively evaluated by SHG. Forty-six (67.6%) patients in whom SHG did not identify any findings in the uterine cavity (negative group) were followed by diagnostic hysteroscopy. Another 22 (32.4%) who were identified by SHG to have abnormal endometrial findings, such as an echogenic or polypoid mass (positive group), were followed by operative hysteroscopy and by postoperative SHG. In the positive group the basal transvaginal sonogram revealed an endometrial echogenic mass in only 10 (45.5%). In the remaining 12 (54.5%) patients, the transvaginal sonogram identified only thick endometrium. In these latter 12 patients, histological assessment confirmed endometrial polyps in 8 (66.7%) and fibroid in 1 (8.3%). Four (18.2%) patients in the positive group had no histological endometrial pathology. Two (50%) of them had a uterine septum as diagnosed during hysteroscopy, in one (25%) operative hysteroscopy failed to resect the endometrial polyp, and in another (25%) there was a false-positive SHG diagnosis. Overall, SHG accurately diagnosed endometrial and/or other intrauterine pathology in 95.5% of these patients. In the 46 patients with "negative" basal SHG features, diagnostic hysteroscopy confirmed this diagnosis. Thus, there was no SHG false-negative diagnosis. Comparing the results of the basal SHG with those of operative hysteroscopy and/or the histopathological findings in the positive group, the sensitivity of SHG was 1.0, the specificity 0.0, positive predictive value 95.5%, and negative predictive value 0.0. It is suggested that SHG is a useful diagnostic tool for the assessment of specific endometrial pathologies in asymptomatic postmenopausal breast cancer patients treated with tamoxifen who were diagnosed by transvaginal sonography to have thick endometrium.