OBJECTIVE: We sought to describe the natural history of abnormal vaginal flora in pregnancy and estimate the efficacy of oral clindamycin in eradicating it and preventing relapse. METHODS: This was a subanalysis of a randomized trial of oral clindamycin for abnormal vaginal flora in pregnancy. All 494 enrolled women were asked to provide a vaginal smear 2 weeks after treatment and every second participant to provide further smears at 20, 24, 28, 32, and 36 weeks of gestation. We used Nugent score of Gram-stained smears to assess the cure rate among the clindamycin group and the rate of spontaneous resolution among the placebo group. RESULTS: Posttreatment smears were available for 462 women (231 in each of the clindamycin and placebo arms). The prevalence of abnormal flora posttreatment was 10%(22 of 231) in the clindamycin group compared with 93%(214 of 231) in the placebo group (P < .001). Two hundred nineteen women obtained 4 weekly smears; slides for 84 women were lost, and results were available for 135 women (69 clindamycin, 66 placebo). In the clindamycin group, the prevalence of abnormal flora was 15%at 20 weeks of gestation and 17%at 36 weeks of gestation compared with 69%at 20 weeks of gestation and 43%at 36 weeks of gestation in the placebo group. CONCLUSION: Oral clindamycin eradicated abnormal flora in 90%of treated pregnant women and maintained a normal flora in two thirds of women throughout pregnancy. Almost one third of untreated women in our study had spontaneous resolution of abnormal flora by 20 weeks of gestation. Because previous research has shown that spontaneous resolution does not modify the risk of preterm birth, early screening is essential. OBJECTIVE: We sought to describe the natural history of abnormal vaginal flora in pregnancy and estimate the efficacy of oral clindamycin in eradicating it and preventing relapse. METHODS: This was a subanalysis of a randomized trial of oral clindamycin for abnormal vaginal flora in pregnancy. All 494 enrolled women were asked to provide a vaginal smear 2 weeks after treatment and every second participant to provide further smears at 20, 24, 28, 32, and 36 weeks of gestation. We used Nugent score of Gram-stained smears to assess the cure rate among the clindamycin groups and the rate of spontaneous resolution among the placebo group. RESULTS: Posttreatment smears were available for 462 women (231 in each of the clindamycin and placebo arms). The prevalence of abnormal flora posttreatment was 10% (22 of 231 ) in the clindamycin group compared with 93% (214 of 231) in the placebo group (P <.001). Two hundred nineteen women obtained 4 weekly smears; slides for 84 women were lost, and results The clindamycin group, the prevalence of abnormal flora was 15% at 20 weeks of gestation and 17% at 36 weeks of gestation compared with 69% at 20 weeks of gestation and 43 % at 36 weeks of gestation in the placebo group. CONCLUSION: Oral clindamycin eradicated abnormal flora in 90% of treated pregnant women and maintained a normal flora in two thirds of women throughout pregnancy. Almost one third of untreated women in our study had spontaneous resolution of abnormal flora by 20 weeks of gestation. Because previous research has shown that spontaneous resolution does not modify the risk of preterm birth, early screening is essential.