Mifepristone treatment prior to insertion of a levonorgestrel releasing intrauterine system for improved bleeding control - a randomized controlled trial

STUDY QUESTION: Does pre-treatment with a low dose of mifepristone improve irregular vaginal bleeding patterns during the initial 3 months after LNG-IUS placement? SUMMARY ANSWER: Mifepristone treatment prior to LNG-IUS insertion results in significantly lower bleeding and spotting rates for the first month post insertion but no significant reduction at 3 months post insertion. WHAT IS KNOWN ALREADY: One of the leading causes of premature discontinuation of the LNG-IUS is unscheduled bleeding in the first months following its insertion. Up to now, there has been no effective treatment to prevent this side effect which reduces continuation rates for one of the most effective contraceptives. STUDY DESIGN, SIZE, DURATION: This randomized, double blinded, controlled trial was conducted between 2009 and 2015. In total, 68 women opting for a LNG-IUS were screened for eligibility, of whom 58 were randomized at a ratio of 1: 1 in blocks of 10 to pre-treatment with mifepristone or a comparator. The main outcome was the rate of bleeding and spotting days reported during the first 3 months post LNG-IUS 52 mg placement. PARTICIPANTS/MATERIALS, SETTING, METHODS: Healthy women with regular and normal menstrual cycles aged 18-43 years were enrolled at a single center in a university hospital; 29 were allocated to 2 months pre-treatment with a low dose mifepristone and 29 to the comparator prior to insertion of the LNG-IUS. Women were advised to use barrier methods during the pre-treatment period. Bleeding diaries were collected for the pre-treatment period and for the first 6 months after the LNG-IUS placement. MAIN RESULTS AND THE ROLE OF CHANCE: There were no differences in demographic or baseline characteristics between the study groups. Data for analysis of the main outcome were contributed by a per protocol population of 19 women per group. There was a significant lower bleeding and spotting rate in the mifepristone group (-17.8% points, P < 0.001) after 2 months of pre-treatment and during the first month after the LNG-IUS insertion. In the following months, no significant difference could be seen. While no pregnancies occurred prior to LNG-IUS insertion in the mifepristone group, there were three unintended pregnancies in the comparator group which emphasizes the need for a reliable contraceptive potential in any pre-treatment regimen used in clinical practice. LIMITATIONS, REASONS FOR CAUTION: The use of mifepristone prior to the LNG-IUS insertion in this trial was used as prophylaxis against unscheduled bleeding after the placement of the device. Although this side effect constitutes a major concern in a clinical setting, only a subset of women are at risk. This is the first study using pre-treatment to attempt improved bleeding control. The differences were small and the effect was short lasting but the reduced rate of bleeding and spotting observed during the first month following LNG-IUS insertion indicates that this approach may be further explored. The fact that there were three pregnancies in the comparator group stresses the need for any pre-treatment to also protect against unplanned pregnancy. WIDER IMPLICATIONS OF THE FINDINGS: Modified treatment protocols of mifepristone could be used prior to the LNG-IUS insertion to investigate possible further improvement of the outcome. The effect size of the current dose used might have been more prominent in women with LNG-IUS if the treatment also continued some weeks after the placement of the device. Although the low dose of mifepristone used in this trial is not available in Europe, other progesterone receptor modulators currently available could be investigated in larger clinical trials. To avoid unintended pregnancy in the pretreatment period, the dosage used should, ideally, also be effective for contraception and the pretreatment period should be kept as short as possible. STUDY FUNDING/COMPETING INTEREST(S): This study was funded by the Swedish Research Council (2012-2844, 2017-00932), Stockholm County Council and Karolinska Institutet (ALF). Conflicts of interests for K.G.D. and H.K.K. are listed at the end of the paper. The other authors have no conflicts of interest. TRIAL REGISTRATION DATE: 20 July 2009. DATE OF FIRST PATIENT'S ENROLMENT: 24 November 2009.