Selective Endometrial Screening versus Routine Curettage in Postmenopausal Women Undergoing Intrauterine Device Removal: A Retrospective Cohort Study

Abstract Background The choice between selective and routine endometrial screening during postmenopausal intrauterine device (IUD) removal remains controversial. Universal blind curettage frequently yields insufficient endometrial tissue and results in overtreatment. This study aimed to evaluate a selective, risk-stratified endometrial screening strategy for postmenopausal women undergoing IUD removal and develop an individualized endometrial screening strategy for this patient population. Methods A total of 512 naturally postmenopausal women (amenorrhea ≥ 12 months) who underwent IUD removal were retrospectively enrolled. According to the presence or absence of postmenopausal abnormal uterine bleeding (AUB), patients were divided into a bleeding group (n = 128) and a non-bleeding group (n = 384). All patients with AUB underwent endometrial sampling; asymptomatic patients were sampled only when high-risk transvaginal ultrasound findings were present (endometrial thickness > 8 mm, heterogeneous echogenicity, or uterine cavity effusion; n = 218). The remaining 166 low-risk asymptomatic patients underwent IUD removal alone. Baseline clinical characteristics, preoperative transvaginal ultrasound findings, and postoperative pathological outcomes were analyzed. Results Patients with AUB had significantly higher body mass index (BMI), higher prevalence of diabetes mellitus, hormone replacement therapy (HRT) history, and tamoxifen exposure history (all P < 0.05). Endometrial thickening, heterogeneous echogenicity, and uterine cavity effusion were all significantly more common in the AUB group (all P < 0.001), whereas the rate of IUD incarceration did not differ (P = 0.121). Among the 346 sampled patients, the overall detection rate of significant endometrial lesions (hyperplasia with or without atypia and carcinoma) was 10.40%. The lesion detection rate in the bleeding group was 14.06%, which was significantly higher than 8.26% in the asymptomatic ultrasound high-risk group (P = 0.040). Low-risk asymptomatic patients with normal ultrasound presented an extremely low risk of severe endometrial lesions. Blind curettage yielded invalid tissue specimens in a proportion of cases. Conclusions Postmenopausal abnormal uterine bleeding, abnormal endometrial ultrasound features, diabetes, obesity, and hormone-related exposure are significant risk factors for endometrial lesions in postmenopausal women undergoing IUD removal. Routine universal curettage is unnecessary for unselected postmenopausal women undergoing IUD removal. A selective endometrial screening strategy integrating clinical symptoms, transvaginal ultrasound and systemic risk factors is feasible, safe and effective. It avoids unnecessary curettage in around one-third of patients and reduces overtreatment.