Modest increase in rate of completed miscarriage with mifepristone before misoprostol

Référence

Chu JJ, Devall AJ, Beeson LE, et al. Mifepristone and misoprostol versus misoprostol alone for the management of missed miscarriage (MifeMiso): a randomized, double-blind, placebo-controlled trial. Lancet 2020;396(10253):770-778.

Sommaire

This was a multicenter trial in the United Kingdom to assess the potential benefit of adding oral mifepristone as pretreatment, 48 hours prior to the administration of misoprostol, for women with missed abortion who choose medical management rather than expectant or surgical management. Included women were at least 16 years old and had a nonviable pregnancy within the first 14 weeks of gestation as confirmed by ultrasound. Women were excluded if a gestational sac was not present, bleeding was life-threatening, they had contraindication to mifepristone or misoprostol, or were participants in other trials of medicinal products. Upon enrollment, women were randomized to receive a 200-mg oral dose of mifepristone or identical placebo. After 48 hours women in both groups returned to receive 800 mcg misoprostol that could be administered by oral, sublingual, or intravaginal route. After another 48 hours, women who had little or no bleeding were asked to contact the research office to discuss taking a second dose of misoprostol. Women returned 7 days after enrollment for an ultrasound to assess for the presence of a gestational sac. Women were instructed to contact their local research office if they had heavy bleeding or pain at any time during the study. Women who failed medical treatment were managed according to local hospital practice. All participants were followed up for at least 3 weeks and until they had a negative urine pregnancy test result. The sample size was calculated to detect a 10-percentage point decrease from an estimated 25% failure to pass the gestational sac within 7 days (control) to 15% in the mifepristone group. Analysis was by intention to treat. Multiple baseline characteristics were balanced between groups through a sophisticated computerized randomization process. The actual reduction was statistically significant, without reaching the predefined threshold for clinical importance, from 82/348 (24%) to 59/348 (17%) (relative risk 0.73; 95% CI 0.54 - 0.99; number needed to treat = 15; 8 - 159). There was no evidence for subgroup effects based on age, body mass index, parity, gestational age, or bleeding score. There were also no differences between groups for serious adverse effects.