Evaluation of endometrium by transvaginal ultrasound and baseline factors as a predictor for endometrial abnormalities in asymptomatic postmenopausal women

OBJECTIVE: The cut-off value of endometrial thickness (ET) in asymptomatic postmenopausal women, beyond which intervention is required, is still debated. This study aimed to provide a clinical decision support tool to guide the management of asymptomatic postmenopausal women with incidentally detected endometrial thickening. METHODS: A retrospective observational study was conducted involving 280 asymptomatic postmenopausal women with ET > 5 mm who underwent hysteroscopy or dilation and curettage. Participants were stratified by menopausal duration and history of endocrine therapy for breast cancer. We analysed associations between ET, transvaginal ultrasound (TVUS) features, baseline characteristics and pathological outcomes. Based on univariate and multivariate analyses, we developed a prediction model to help predict endometrial lesions. RESULTS: The pathological negativity rate was 35.36% among the whole cohort. We propose two candidate diagnostic thresholds for ET in asymptomatic postmenopausal women: a lower cut-off of 7 mm suggestive of any endometrial pathology (primary outcome), and a higher cut-off of 9.5 mm warranting suspicion for clinically significant pathology, including atypical endometrial hyperplasia (AEH) and endometrial cancer (EC) (secondary outcome). Our findings indicate a statistically significant difference in ET between healthy women within 5 years of menopause and those beyond 5 years. Nevertheless, no such significant disparity was observed in women with endometrial lesions. No significant difference was observed in ET between the overall population and patients receiving postoperative endocrine therapy for breast cancer. Using the presence or absence of endometrial lesions as the outcome, a prediction model was established. The model incorporates age, height, TVUS vessel pattern, and echogenicity. It provides a more holistic risk assessment tool than relying on a single millimetre measurement. CONCLUSION: Adopting a 7 mm candidate cut-off for intervention could spare over one‑third of asymptomatic women from unnecessary invasive procedures. Risk assessment should integrate TVUS morphological features with clinical factors, not merely rely on a single ET measurement. External validation in diverse populations is required before clinical implementation.