Interpretation
The main findings of our study that LNG-IUS achieved higher 16-week CR rate than MA in AEH patients were consistent with findings from previous retrospective studies 12, 13. A meta-analysis evaluating 24 observational studies showed that oral progestin achieved lower pooled regression rate (69% vs. 90%, P=0.03) compared with LNG-IUS in AEH patients 13. Although our study for the first time provides evidence from prospectively randomized and controlled trial, the sample size was not large enough to draw a conclusion. Further confirmation is needed in phase III study with sufficient sample size.
Our study also found that LNG-IUS was associated with fewer adverse events than MA or MA+LNG-IUS. This is important because fertility preserving treatment takes long time which is at least four to six months 7-9. Long-term usage of MA might cause many adverse events such as weight gain, edema, vomiting that affect quality of daily life, and even cause thrombosis which is life-threatening4. In this context, LNG-IUS instead of oral progestin might provide patients with higher life-quality for less and milder side-effects, and thus, might increase the patient compliance of fertility-sparing treatment 35-37.
Data in our study suggested that the efficacy of LNG-IUS alone might be better than MA alone in AEH patients with BMI≥28 kg/m2. This is important because obesity has been shown to be the most important factor adversely affecting the fertility-preserving treatment in AEH and EEC patients24, 38, 39. Long term oral progestin usage might also lead to higher risk of thromboembolism in obese women. Our results support that LNG-IUS might be more suitable in AEH patients with BMI≥28 kg/m2.
Our data did not find better treatment effect using LNG-IUS plus MA compared with LNG-IUS or MA alone, which was an unexpected result. It might be because LNG is a highly effective progestin, and the drug concentration of LNG using LNG-IUS could reach nearly a thousand times in endometrium than oral MA 40. Thus, LNG-IUS alone is effective enough on endometrial lesion, and adding systemic MA could not add more value on the treatment effect in endometrial lesion.
In our study, differences in the CR rate between groups became less significant from 16 weeks to 32 weeks of treatment. The reason might be that all patients received hysteroscopic evaluation and treatment which had been shown to effectively increase the CR rate in AEH and EEC patients 8. With the prolongation of treatment time, the effect of lesion removal by hysteroscopy plays important role in improving treatment effect regardless of the different type of progestin treatment.