Background: This study aimed to compare the levonorgestrel-releasing intrauterine system (LNG-IUS) with oral cyclic medroxyprogesterone acetate (MPA) in endometrial hyperplasia therapy using randomized controlled trials (RCTs).
Methods: The study searched MEDLINE, EMBASE, CENTRAL, and other databases. All regression outcomes were calculated for dichotomous outcomes in terms of relative risk (RR) and 95% confidence intervals (CIs) using a Mantel-Haenszel random effects model.
Results: The search found 543 articles but selected 342 articles after the removal of duplicates. A meta-analysis found five RCTs (377 patients). The study did not analyze RR for total outcome because of high heterogeneity (I 2 = 87%). In a subgroup analysis of studies with non-obese women, the LNG-IUS treatment appeared to have a higher regression rate than oral MPA (RR 1.41; 95% CI 1.23-1.62; 4 trials, 265 patients; I 2 = 0%). In a subgroup analysis of studies with obese women, LNG-IUS appeared to have a regression rate similar to that of oral MPA (RR 1.03; 95% CI 0.94-1.13; 1 trial, 60 patients). In a subgroup analysis according to histology in the non-obese group, the LNG-IUS treatment appeared to have a higher regression rate than oral cyclic MPA in a meta-analysis of women with non-atypical endometrial hyperplasia (RR 1.36; 95% CI 1.07-1.73; 2 trials, 92 patients; I 2 = 6%) and mixed endometrial hyperplasia (atypical and non-atypical) (RR 1.44; 95% CI 1.21-1.71; 2 trials, 173 patients; I 2 = 0%).
Conclusions: The LNG-IUS treatment has a higher regression rate than cyclic MPA in non-atypical endometrial hyperplasia and mixed endometrial hyperplasia therapy for non-obese women but has a similar regression rate, albeit limited, for obese women.