Objectives: Three different treatment options for endometrial hyperplasia were evaluated in a prospective long-time follow-up study, comparing effects of intrauterine levonorgestrel impregnated device (LNG-IUD), low oral dose of medroxyprogesterone acetate (MPA) and no treatment (observation only). To select patients with high probability for co-existing or future carcinoma we used the objective morphometric algorithm, D-score, stratifying patients into three different risk groups. As far as we know, this is the first prospective long-time follow-up study in which treatment recommendation and outcome is based on the D-score assessment.
Methods: From a total of 370 patients initially diagnosed with endometrial hyperplasia from eight different hospitals in North Norway, 258 were available for long-time follow-up. After D-score classification, one of three different treatment options was chosen: LNG-IUD, low oral dose of MPA or observation only. Follow-up controls were performed and biopsies taken in the local hospitals.
Results: Among the 370 investigated cases with endometrial hyperplasia, only ten endometrial cancers were detected at the entrance of the study, all belonging to the high risk group (D-score <0). No further cancers were detected during follow-up, irrespective of risk group. After 6 months treatment with LNG-IUD proved significantly superior to oral treatment (p=0.001 for D-score >1 and p=0.003 for D-score 0-1 groups) and observation only (p=0.001 for D-score >1 and p=0.001 for D-score 0-1 groups). After 56 to 108 months the LNG-IUD proved significantly superior to oral treatment and to the observation group. Comparison of oral therapy to observation only showed no significant differences, neither after 6 months nor after long-time observation.
Conclusions: LNG-IUD is the optimal treatment for endometrial hyperplasia. Outcome after oral low-dose MPA regimen is comparable to expectation.