Miscarriage surgery: Should antibiotics be prescribed for pelvic infection prophylaxis?
Administering prophylactic treatment with doxycycline and metronidazole prior to miscarriage surgery does not appear to significantly reduce the risk of pelvic infection, as defined by broad criteria, according to the results of a trial.
The trial enrolled 3,412 women scheduled to undergo surgical evacuation of the uterus due to spontaneous abortion at <22 weeks of gestation. They were randomized to receive a single preoperative dose of 400-mg oral doxycycline plus 400-mg oral metronidazole (n=1,705) or identical placebos (n=1,707).
Within 14 days after surgery, the primary outcome of pelvic infection occurred similarly in the two groups: 4.1 percent (68 of 1,676) of patients who received antibiotics and 5.3 percent (90 of 1,684) of those who received placebo (risk ratio, 0.77; 95 percent CI, 0.56–1.04; p=0.09). [N Engl J Med 2019;380:1012-1021]
Pelvic infection was defined by the presence of at least two of four clinical features—purulent vaginal discharge, pyrexia, uterine tenderness and leucocytosis—or by the presence of one of these features and the clinically identified need to administer antibiotics.
When a stricter definition of the outcome was used (ie, presentation of at least two of the clinical features, without reference to the administration of antibiotics), antibiotic prophylaxis showed a potential benefit. Pelvic infection was documented in 1.5 percent (26 of 1,700) of patients in the antibiotics group and 2.6 percent (44 of 1,704) of those in the placebo group (risk ratio, 0.60; 0.37–0.96).
In term of safety, the incidence of adverse events such as diarrhoea, vomiting and blood transfusion was similar in the two groups. None of the patients developed anaphylaxis. Serious AEs were uncommon and did not significantly differ in the antibiotics and placebo groups (0.9 percent vs 1.5 percent, respectively).
“Diagnosis [of pelvic infection] according to strict criteria has been common in the existing literature and is considered meaningful from the perspectives of patients and policy makers. However, we widened the criteria that we used for diagnosis during the course of the trial … in response to safety concerns of some trial clinicians that some pelvic infections were being missed when the strict criteria were used,” according to the investigators.
“Whereas inclusion of clinician judgment among the criteria for diagnosis would be expected to improve the sensitivity for identifying pelvic infection, it is also likely to have decreased specificity... The addition of clinician judgment to the pragmatic definition is likely to have diluted the observed treatment effect,” they added.
In an accompanying commentary, Dr David Serwadda from the Makerere University School of Public Health in Kampala, Uganda, noted that the change made in the definition of the primary outcome has major implications for the interpretation of the results. [N Engl J Med 2019;380:1075-1076]
Serwadda concurred with the investigators that the broader criteria would be expected to be more sensitive but less specific, by including women with mild or no pelvic infections. Changing the definition of the primary outcome to a less specific one may have reduced the probability of observing a true difference in event rates between the trial groups.
“Thus, although the investigators concluded that antibiotic prophylaxis before miscarriage surgery ‘did not result in a significantly lower risk of pelvic infection, as defined by pragmatic broad criteria, than placebo’ I would interpret the results as indicating that antibiotic prophylaxis prevented pelvic infections as defined by international diagnostic criteria,” he said.
In light of the risks associated with pelvic infections in low- and medium-income countries, data from the current trial provide reasonable support for prescribing prophylactic antibiotics in women undergoing surgical evacuation of the uterus, he added. “Antibiotic resistance, however, will need to be monitored.”