Comparative Effectiveness of Hysteroscopic Endometrial Resection Versus the 52-mg Levonorgestrel-Releasing Intrauterine System for Heavy Menstrual Bleeding: A Retrospective Cohort Study

Background and Objectives: To compare the effectiveness, durability, and patient-reported outcomes of hysteroscopic endometrial resection (ER) and the 52 mg levonorgestrel-releasing intrauterine system (LNG-IUS) in women with medically refractory heavy menstrual bleeding (HMB) without structural uterine pathology. Materials and Methods: This retrospective cohort study included 242 women treated between January 2021 and December 2024, of whom 117 underwent ER and 125 received a 52 mg LNG-IUS. The primary outcome was hysterectomy during follow-up. Secondary outcomes included bleeding-related symptoms, treatment satisfaction, the change in the Female Sexual Function Index-6 (FSFI-6) total score, LNG-IUS discontinuation/expulsion, and subsequent reintervention. Baseline bleeding severity was characterized using the Pictorial Blood Assessment Chart (PBAC), hemoglobin, and serum ferritin, and abnormal uterine bleeding etiologies were classified according to PALM–COEIN. To address non-randomized treatment allocation, inverse probability of treatment weighting (IPTW) was performed using prespecified baseline covariates. Hysterectomy was further evaluated using Firth penalized logistic regression, Kaplan–Meier analysis, and Cox proportional-hazards models. Results: Baseline PBAC, hemoglobin, ferritin, and PALM–COEIN distributions were comparable between groups, whereas age, gravidity, parity, and baseline FSFI differed. After IPTW, all prespecified baseline covariates achieved adequate balance (absolute standardized mean differences < 0.10). Hysterectomy occurred in 3.4% of women after ER and 16.0% after LNG-IUS treatment. In the IPTW-weighted analysis, LNG-IUS was associated with a higher risk of hysterectomy (RR = 8.9, 95% CI 2.8–28.4; p < 0.001) and any bleeding-related symptom (RR = 12.8, 95% CI 5.0–32.9; p < 0.001), while treatment satisfaction was lower (RR = 0.48, 95% CI 0.39–0.59; p < 0.001). Hysterectomy-free survival was lower after LNG-IUS treatment (log-rank p < 0.001); the age- and parity-adjusted Cox model showed an HR of 11.6 (95% CI 3.4–39.4; p < 0.001), and the IPTW-weighted Cox model showed an HR of 9.8 (95% CI 3.0–32.1; p < 0.001). Firth regression similarly demonstrated a higher adjusted odds of hysterectomy with LNG-IUS (aOR = 18.9, 95% CI 5.0–71.9; p < 0.001). The 12-month FSFI total score remained lower with LNG-IUS in the IPTW-weighted analysis (B = −2.1, 95% CI −2.5 to −1.6; p < 0.001). Among women initially treated with LNG-IUS, 39.2% experienced discontinuation/removal or expulsion and 25.6% underwent subsequent procedural reintervention. Conclusions: In this real-world retrospective cohort, ER was associated with fewer subsequent hysterectomies and bleeding-related symptoms, higher treatment satisfaction, and more favorable sexual-function outcomes than the 52 mg LNG-IUS. These associations remained consistent across propensity-weighted, penalized logistic, and time-to-event analyses. Because treatment allocation was non-randomized and preference-based, the findings should be interpreted as comparative associations rather than proof of causal treatment superiority.