WS05‐03Ultrasound in the diagnosis of endometrial abnormalities

There is a potential role of ultrasound examination in the management of women with postmenopausal (PMP) bleeding. Because women with an endometrium = 4 mm at transvaginal ultrasound examination rarely have endometrial pathology, it may be justified not to sample their endometrium. However, endometrial pathology is common among women with endometrium = 5 mm. In our prospective study, 105 consecutive women with PMP bleeding and endometrium = 5 mm underwent hydrosonography, hysteroscopy, D & C, and hysteroscopic resection. 80% of these women had endometrial pathology, including 24% with endometrial malignancy. 94% of the pathological lesions manifested a focal growth pattern. 87% of the focal lesions were only partly removed or remained in situ after D & C, D & C missing 25% of malignancies and 53% of benign pathology. Hydrosonography was as good as hysteroscopy at detecting focally growing lesions in the uterine cavity (agreement between hydrosnography and hysteroscopy being 96%; Kappa 0.77), but neither hysteroscopy nor hydrosnography could reliably discriminate between benign and malignant lesions. In a randomized trial, 49 women with PMP bleeding and endometrium = 4 mm were randomized to D & C and 48 to ultrasound follow up. 14 (29%) women in the D & C group and 22 (46%) in the ultrasound group had either rebleeding or manifested endometrial growth during a 12‐month follow‐up period. Endometrial pathology was not found in women with rebleeding unless the endometrium was = 5 mm. Conclusion Women with PMP bleeding and endometrium = 5 mm should undergo hydrosnography. If focal lesions are detected hysteroscopic resection should be carried out. In women with PMP bleeding, rebleeding during the first 12 months is an indication to sample the endometrium only if the endometrium measures = 5 mm at ultrasound examination.