We compared endometrial thickness and volume in patients with postmenopausal bleeding, and examined the value of each parameter in differentiating between benign and malignant endometrial pathology. A total of 103 patients with a history of postmenopausal bleeding were recruited into the study. Patients who were taking hormone replacements therapy or other hormonal preparations with a known effect on the endometrium were excluded. Each patient underwent three-dimensional ultrasonography for the measurement of endometrial thickness and volume. In 97 cases both of these measurements were obtained and the results were compared to the histological diagnosis after endometrial biopsy or dilatation and curettage. Endometrial cancer was diagnosed in 11 patients. The mean endometrial thickness was 29.5 mm (SD 12.59) and the mean volume was 39.0 ml (SD 34.16). In the remaining 86 patients there were eight cases with endometrial hyperplasia and seven with endometrial polyps. The endometrial thickness and volume in patients with benign pathology was 15.64 mm (SD 5.26) and 5.47 ml (SD 6.32), respectively. In 71 patients with atrophic or normal endometrium the mean thickness and volume was 5.29 mm (SD 3.97) and 0.91 ml (SD 1.71), respectively. Receiver operating characteristic curves showed endometrial volume to be superior to endometrial thickness for the diagnosis of endometrial cancer. The optimal cut-off value of endometrial thickness for the diagnosis of cancer was 15 mm, with the test sensitivity of 83.3% and positive predictive value of 54.5%. With the cut-off level of 13 ml for endometrial volume measurement the sensitivity was 100% and the positive predictive value 91.7%. Both the thickness and volume were higher in patients with advanced and less differentiated cancers. The measurements of endometrial volume was superior to that of endometrial thickness as a diagnostic test for the detection of endometrial cancer in symptomatic postmenopausal women.