Clinical outcomes of three hysteroscopic levonorgestrel-releasing intrauterine system placement methods in the management of adenomyosis: a retrospective cohort study

Background The levonorgestrel-releasing intrauterine system (LNG-IUS) is effective for adenomyosis, yet expulsion frequently limits its clinical application. This study aimed to compare clinical outcomes of three hysteroscopic LNG-IUS placement techniques for adenomyosis. Methods This retrospective cohort recruited 131 adenomyosis patients (non-suture fixation n = 32, suture fixation n = 23, conventional placement n = 76) at Fujian Maternity and Child Health Hospital from October 2024 to October 2025. The primary outcome was LNG-IUS expulsion or displacement. Follow-up assessments were performed at baseline, 1 month and 6 months after placement for uterine volume, endometrial thickness, hemoglobin, carbohydrate antigen 125, dysmenorrhea and menstrual blood loss. Intraoperative parameters and postoperative adverse events were recorded. Between-group comparisons used ANOVA or Kruskal-Wallis tests; within-group changes used repeated-measures ANOVA or the Friedman test, with Bonferroni correction for multiple comparisons. Subgroup analysis was stratified by uterine cavity depth (cut-off 9.0 cm). Propensity-score matching (1:1:1), generalized estimating equations (GEE) and Cox regression were applied for data analyses. Results Expulsion or displacement rates at 1 and 6 months were 0% in both fixation groups versus 19.74% and 27.63% in the conventional group ( P < 0.01). All groups achieved significant improvements in clinical parameters ( P < 0.001). For patients with uterine cavity depth > 9.0 cm, fixation techniques yielded lower expulsion or displacement rates and superior 6-month clinical improvements versus conventional placement; comparable efficacy was observed among all groups for depth ≤ 9.0 cm. Operation time was shorter for non-suture fixation and conventional placement than suture fixation ( P < 0.001), whereas intraoperative blood loss and adverse reactions were similar ( P > 0.05). In the PSM-matched cohort ( n = 23 per group), expulsion-displacement remained 0% in fixation groups vs. 69.56% conventional. GEE analysis in the matched cohort further revealed significant group × time interactions across the longitudinal clinical outcomes ( P < 0.05). Cox regression identified uterine volume > 150 cm³ as the only independent risk factor (HR = 8.822, 95% CI 3. 172–24. 530, P < 0.001). Conclusions All three methods effectively improve clinical outcomes in adenomyosis patients. For patients with cavity depth > 9.0 cm, hysteroscopic fixation methods were associated with significantly lower observed expulsion rates compared with conventional placement. Non-suture fixation has practical merits of shorter operative time and lower technical requirements. Further prospective studies are required for external validation.