Effectiveness and feasibility of hysteroscopic sterilization techniques: a systematic review and meta-analysis

Abstract

To assess whether hysteroscopic sterilization is feasible and effective in preventing pregnancy. Secondarily, to identify risk factors for failure of hysteroscopic sterilization. A systematic review and meta-analysis. Not applicable. Women undergoing hysteroscopic sterilization. Hysteroscopic sterilization with a commercially available system (Ovabloc Intra Tubal Device, Essure system, or Adiana permanent contraception system). Successful placement at first attempt, confirmed correct placement, complications, incidence of pregnancy, and risk factors for placement failure in hysteroscopic sterilization. Of the 429 citations identified, 45 articles were eligible for analyses. No randomized controlled trials (RCTs) were identified, just cohort studies. Six articles concerned Ovabloc, 37 Essure, and two Adiana sterilization. The probabilities for successful bilateral placement in a first attempt for Ovabloc, Essure, and Adiana, were, respectively, in the ranges 78%-84%, 81%-98%, and 94%. The probabilities of successful bilateral placement could not be pooled because of substantial heterogeneity. The 36 months' cumulative pregnancy rate of Adiana was 16 of 1,000. Reliable pregnancy rates after sterilization with Ovabloc or Essure method could not be calculated. For all three hysteroscopic techniques, the incidence of complications and their severity has not been studied adequately and remains unclear. We also found too little evidence to identify risk factors for placement failure. Sterilization by hysteroscopy seems feasible, but the effectiveness and risk factors for failure of sterilization remain unclear owing to the poor-quality evidence. Both currently applied hysteroscopic sterilization techniques and the coming new techniques must be evaluated properly for feasibility and effectiveness. Appropriate RCTs and observational studies with sufficient power and complete and long-term (>10 years) follow-up data on unintended pregnancies and complications are needed.