A randomized trial of prophylactic doxycycline for curettage in incomplete abortion

To determine whether prophylactic doxycycline at suction curettage for incomplete abortion decreases the rate of postoperative pelvic infection.

We randomized 240 patients to receive intravenous doxycycline or placebo at curettage. Cervical specimens for gonorrhea and chlamydia were obtained preoperatively. Two weeks post-procedure, we evaluated all patients for infectious morbidity and repeated gonorrhea and chlamydia cultures. Statistical analysis used Mann-Whitney U test, McNemar test, or Fisher exact test, as appropriate.

There were no statistically significant differences in age, parity, gestational age, history of sexually transmitted disease, pelvic inflammatory disease, or multiple sex partners between the doxycycline and placebo groups. Preoperative gonorrhea or chlamydia isolates were positive in five (4.2%) and six (5%) of 120 doxycycline patients and four (3.3%) and eight (6.6%) of 120 controls (not significant). All preoperative gonorrhea isolates remained positive postoperatively. Seven (5.8%) controls had positive postoperative chlamydia isolates, as did one (0.8%) in the doxycycline group (P = .06). We diagnosed eight (6.6%) of 120 doxycycline patients and seven (5.8%) of 120 controls with infectious morbidity (not significant).

In our population of patients with incomplete abortion, the prevalence of gonorrhea and chlamydia was low, and prophylactic doxycycline did not decrease the rate of postoperative febrile morbidity.