Endometrial thickness as measured by transvaginal ultrasound (TVUS) is being increasingly used as a first-line method to evaluate patients with vaginal bleeding. Our study aims to examine correlation between the histopathologic diagnosis and the results of TVUS and find a threshold that could reliably exclude carcinoma. We included women, age 55 years and above, who presented with postmenopausal bleeding and had a TVUS within 30 days of their endometrial biopsy. Total of 304 patients met our criteria and were divided into 4 groups. Patients in group A (n=198) had benign/atrophic endometrium, group B (n=44) had polyps, group C (n=30) had hyperplasia, and group D (n=32) had carcinoma. The endometrial thickness obtained by TVUS was compared with the histopathologic finding of the endometrial biopsy. The mean endometrial thickness was 7.5, 12.1, 14.8, and 16.9 mm for groups A to D, respectively. Statistical analysis showed that very low endometrial thickness (3 to 4 mm) would be ideal to use as a threshold to maximize sensitivity. Three of 32 patients in group D had an endometrial thickness ≤4 mm. At a threshold of 4 mm, the sensitivity is 90.6% and increases to 96.9% when decreasing the threshold to 3 mm. However, other parameters such as test accuracy, specificity, and positive predictive values are very low at these thresholds. Sensitivity can be maximized to 96.9% using a threshold of 3 mm. However, this would call into question the cost-effectiveness of this method. Postmenopausal bleeding remains the most reliable indicator of endometrial pathology.