Prophylactic antibiotics for miscarriage surgery not beneficial

Référence

Lissauer D, Wilson A, Hewitt CA, et al. A randomized trial of prophylactic antibiotics for miscarriage surgery. N Engl J Med 2019;380(11):1012-1021.

Sommaire

This well-designed randomized controlled trial, called Antibiotics in Miscarriage Surgery (AIMS), sought to establish whether a prophylactic antibiotic regimen for miscarriage surgery would reduce the incidence of pelvic infection. Women and adolescents (N = 3412) were recruited from 13 hospitals in 4 relatively resource-poor countries (Malawi, Pakistan, Tanzania, and Uganda) if they had an incomplete spontaneous abortion at less than 22 weeks' gestation. Women were excluded if they had evidence of induced abortion, were younger than 16 years, had a current pelvic infection, had received antibiotics within the prior 7 days, or had contraindications to the antibiotics of the study arm. Surgery was performed according to local practice: sharp curettage in 70% of patients, manual vacuum aspiration in 23%, and suction curettage in 6%. Women were given four 400-mg capsules of doxycycline plus one 400-mg capsule of metronidazole—or 5 placebo capsules with identical appearance—2 to 4 hours before surgery. The primary outcome was pelvic infection within 14 days of surgery, defined as having at least 2 of the following criteria: purulent or foul-smelling vaginal discharge, temperature higher than 38.0 degrees centigrade, pelvic tenderness, a white blood cell count of more than 12,000 per microliter, or clinician judgment of pelvic infection that warrants treatment. The rate of postoperative infection was 4.1% in the antibiotic group and 5.3% in the placebo group (NS). Several prespecified subgroup analyses showed no differences based on patient age younger than 35 years or 35 years or older, gestational age less than 12 weeks or 12 weeks or greater, type of miscarriage (incomplete or missed), HIV infection status, or country of recruitment. Women who underwent manual vacuum aspiration had a lower rate of infection with antibiotic prophylaxis (1.3% vs 4.1%; risk ratio 0.32; 95% CI 0.12 - 0.86; not enough data to calculate number needed to treat). Rates were similar between groups with curettage procedures. Using a narrower definition of infection that excluded clinician judgment there were significantly fewer infections in the treated group (26/1700 [1.5%] vs 44/1704 [2.6%]; relative risk ratio 0.60; 0.37 - 0.96). One woman in the placebo group died of complications related to uterine perforation with bowel injury. Otherwise, adverse events were rare (0.9% and 1.5% in the antibiotic and placebo groups, respectively).