RESUMEN
Background: Postmenopausal bleeding is generally regarded as an ominous alarm of genital pathologies which requires a thorough evaluation clinically and pathologically to exclude carcinoma as the cause and ensure a benign pathology. This study aims at finding out whether clinical diagnosis and ultrasonographic features can be reliable parameters for the diagnosis of causes and whether the findings correspond with histopathology reports.Methods: This observational study was conducted in a tertiary care centre in Pondicherry between January 2018 to August 2019. 114 women were enrolled for whom detailed history taking and clinical examination was done. All the patients were subjected to transvaginal ultrasonography. Patients with clinically visible lesions on cervix and vulva were subjected to biopsy and the rest underwent fractional curettage and the sample was sent for histopathological examination. Finally, histopathology report was compared with clinical and ultrasonographic findings.Results: With endometrial thickness cut off of 4 mm, the sensitivity, specificity, positive predictive value and negative predictive value in predicting malignancy by ultrasonography were 100%, 12.3%, 4.5% and 100%. Histopathology showed atrophic endometrium (43.8%), endometrial hyperplasia (8%), endometrial polyp (7.9%) and endometrial carcinoma (3%). Clinical and ultrasonographic findings did not show any statistical correlation with histopathology.Conclusions: Authors conclude that clinical findings and ultrasonographic features do not correlate with histopathology in cases of postmenopausal bleeding for which atrophic endometrium was the commonest etiology. However, ultrasound should be done routinely before endometrial sampling as the sensitivity for predicting malignancy was 100% for endometrial thickness cut off of 4 mm.