Six trials met the inclusion criteria for the meta-analysis (n=390 participants). The range of quality scores were 2 to 3 points out of a possible maximum of 5 on the Jadad criteria. Follow-up assessment of effect of treatment ranged from six to 36 months.
No statistically significant difference was observed between levonorgestrel-releasing intrauterine system and endometrial ablation in reducing menstrual blood loss: at six months (WMD -31.96, 95% CI -65.96 to 2.04), 12 months (WMD 7.45, 95% CI -12.37 to 27.26), and 24 months (WMD -26.70, 95% CI -78.54 to 25.15).
There was no significant difference in treatment failure between patients who received levonorgestrel-releasing intrauterine system and patients who received endometrial ablation (RR 1.40, 95% CI 0.89 to 2.20; six RCTs; n=390 participants).
There was no difference between treatment groups in overall quality of life scores at one year, although marked improvements were found between baseline and post-intervention quality of life scores. Also, there was no difference in the eight dimensions of the SF-36 scale evaluated between levonorgestrel-releasing intrauterine system and endometrial ablation.
No serious adverse events were reported (two RCTs).
These findings did not change in any of the planned sensitivity analyses.