OBJECTIVE: To determine whether antibiotic treatment for bacterial vaginosis or Trichomonas vaginalis during pregnancy decreases the risk of preterm birth and associated adverse outcomes.

DATA SOURCES: Pre-MEDLINE and MEDLINE (1966–2003), EMBASE (1980–2003), and the Cochrane Library were searched using the keywords “bacterial vaginosis,” “Trichomonas,” “Trichomonas vaginalis,” “Trichomonas vaginitis,” “Trichomonas infections,” “pregnancy,” “pregnant,” “antibiotics,” and “antibiotic prophylaxis.”

METHODS OF STUDY SELECTION: The search produced 1,888 titles, of which 1,256 abstracts were reviewed further. Of these, 1,217 were ineligible. Inclusion criteria were the following: randomized controlled trials in which antibiotics were compared with no antibiotic or placebo, for women in the second or third trimester of pregnancy with symptomatic or asymptomatic bacterial vaginosis or Trichomonas vaginalis, intact membranes, and not in labor. Exclusion criteria were as follows: published in a language other than English, dropout rate of more than 20% of women in either group, and lack of usable outcomes. Of the 39 papers reviewed in detail, 14 studies were included in the meta-analysis.

TABULATION, INTEGRATION, AND RESULTS: One of the authors reviewed titles obtained from the searches, and 2 reviewers independently reviewed the abstracts, excluded those that were ineligible, identified eligible papers, and abstracted the data. For women with bacterial vaginosis, antibiotics reduced the risk of persistent infection but did not reduce the risk of preterm birth or the incidence of associated adverse outcomes for the general population or for any subgroup analyzed. For women with Trichomonas vaginalis, metronidazole reduced the risk of persistent infection but increased the incidence of preterm birth.

CONCLUSION: Contrary to the conclusions of 3 recent systematic reviews, we found no evidence to support the use of antibiotic treatment for bacterial vaginosis or Trichomonas vaginalis in pregnancy to reduce the risk of preterm birth or its associated morbidities in low- or high-risk women.

Both bacterial vaginosis and Trichomonas vaginalis in pregnancy have been shown to be associated with preterm labor and preterm birth.4–9 With bacterial vaginosis, the proportion of normal lactobacilli is decreased, with resultant overgrowth of other organisms, including Gardnerella vaginalis, Mobiluncus, and anaerobes. Studies have shown that the risk of preterm birth is increased when bacterial vaginosis is diagnosed early rather than late in pregnancy.10 Trichomonas vaginalis is a sexually transmitted disease, with a prevalence of approximately 10%, depending on the population.9

If bacterial vaginosis and/or Trichomonas vaginalis are associated with preterm labor and birth, it is possible that antibiotic treatment of these organisms would reduce the risk of this adverse outcome. Three meta-analyses have been recently published on the treatment of bacterial vaginosis; all have concluded that there is no benefit to screening and treatment of bacterial vaginosis among the general obstetric population. However, all 3 continue to suggest that there is benefit to screening and treatment of women at high risk for preterm birth.11–13

The most recent Cochrane review on the treatment of Trichomonas vaginalis has concluded that there is no benefit from treatment with antibiotics and suggests that there may be a possible harmful effect because the largest trial was stopped early due to an increased risk of preterm birth with metronidazole treatment.14

More randomized controlled trials have since been published, perhaps altering the conclusions of the above reviews. We therefore undertook this systematic review and meta-analysis of randomized controlled trials of antibiotic treatment for pregnant women with either bacterial vaginosis or Trichomonas vaginalis to determine whether antibiotic therapy reduced the risk of preterm birth or associated adverse outcomes among pregnant women with either of these conditions.

Pre-MEDLINE and MEDLINE (1966–2003), EMBASE (1980–2003), and the Cochrane Library were searched using the keywords “bacterial vaginosis,” “Trichomonas,” “Trichomonas vaginalis,” “Trichomonas vaginitis,” “Trichomonas infections,” “pregnancy,” “pregnant,” “antibiotics,” and “antibiotic prophylaxis.” All searches were limited to female human subjects.

Studies were considered eligible for inclusion in the review if they were randomized controlled trials in which antibiotics were compared with no antibiotic or placebo, for women with symptomatic or asymptomatic bacterial vaginosis or Trichomonas vaginalis, as defined by each author's diagnostic criteria. Studies were eligible if women were not in labor, had intact membranes, and were in the second or third trimester of pregnancy at the time of randomization. The primary outcome measure was preterm birth before 37 weeks of gestation. Secondary outcome measures were preterm birth before 35 weeks, 34 weeks, 32 weeks, or 28 weeks of gestation, low birth weight (< 2,500 g), very low birth weight (< 1,500 g), preterm prelabor rupture of the membranes (PPROM), chorioamnionitis, endometritis, peripartum infection, perinatal death, neonatal sepsis, and admission to a neonatal intensive care unit (NICU). Eradication of the organism from the vagina was also assessed. Studies were considered eligible if data were presented for eligible women, even if not all women met the eligibility criteria.

Studies were excluded if they were not reported in English and/or if the study design was not a randomized controlled trial. Studies were also excluded if more than 20% of women in either group in the trial were lost to follow-up or if there were no usable outcomes reported.

One of the authors (K.A.G.) reviewed titles obtained from the searches. Case reports, descriptive studies, studies published in abstract form only, commentaries, and letters to the editor were excluded. Two reviewers (K.A.G. and N.O.) independently reviewed all remaining abstracts. If the reviewers agreed that an abstract described a study that was ineligible for the review, the abstract was excluded and the paper was not reviewed. If the reviewers were not in agreement or if they agreed that the study might be eligible, the paper was reviewed. Both reviewers (K.A.G. and N.O.) independently came to an assessment as to whether the paper was eligible for inclusion in the review. Disagreements were resolved by consensus or by discussion with a third party (M.E.H.). The reference lists of review papers and clinical practice guidelines were also reviewed for potentially eligible studies. Two reviewers (K.A.G. and N.O.) independently abstracted data from eligible papers.

The search produced 1,888 titles. Of these 1,256 abstracts were selected for further review. Of those abstracts reviewed, 1,217 were excluded because they clearly did not meet eligibility criteria.

The papers for the remaining 39 abstracts were reviewed in detail. Of these, 25 studies were excluded for the following reasons: 4 studies were not randomized,9,15–17 in 4 studies more than 20% of randomized subjects were lost to follow up,18–21 in 5 studies there were no data available on the outcomes of interest or data were not presented separately for bacterial vaginosis or Trichomonas vaginalis,22–26 4 studies excluded pregnant women,27–30 1 study included only women with ruptured membranes or women in preterm labor,31 in 2 studies all culture-positive women were treated,32,33 and 4 studies presented data that were published elsewhere.34–37 A recently published study of Trichomonas vaginalis treatment during pregnancy 38 was excluded because it was a subanalysis of a larger trial of mass treatment of sexually transmitted diseases to prevent human immunodeficiency virus (HIV) acquisition.39 The remaining 14 papers met our eligibility criteria and were included in this systematic review.40–53

Baseline data for women enrolled in the trials were presented descriptively. The rates of preterm birth at less than 37 weeks of gestation and other outcomes were compared for the antibiotic and no antibiotic or placebo groups for all women who had bacterial vaginosis, irrespective of whether they were symptomatic or asymptomatic or were at low or high risk of preterm birth. Subgroup analyses for the outcome of preterm birth at less than 37 weeks were undertaken for women with bacterial vaginosis who were treated with different types of antibiotics, for women who had confirmed bacterial vaginosis at the time of randomization, and for women who were at high risk for preterm birth. Women at high risk for preterm birth were those with a history of spontaneous preterm birth (or with a history of any preterm birth, if information on spontaneous preterm birth was not available) or second-trimester miscarriage.

The rates of preterm birth at less than 37 weeks and other outcomes were also compared between the antibiotic and the no antibiotic or placebo groups, among women who had Trichomonas vaginalis. Lastly, comparisons of each specific type of antibiotic with no antibiotic were undertaken, for women with either bacterial vaginosis or Trichomonas vaginalis in terms of the risk of preterm birth at less than 37 weeks.

Statistical analyses were undertaken with Review Manager 4.1.1 (The Cochrane Collaboration, Oxford, UK). We calculated a relative risk (RR) and 95% confidence interval (CI) for dichotomous variables using the DerSimonian and Laird random-effects model.54 A random-effects model incorporates between-study variation and gives wider confidence intervals than a fixed-effects model, which ignores between-study variation in estimating CI. Results were considered statistically significant if the 95% CI did not encompass one for RR or if the P value was less than .05. Tests of heterogeneity among pooled results were conducted using simple [chi]² analysis.

Overall, the studies were of high methodological quality in that the randomization sequence was likely concealed to some extent in all studies, because loss to follow-up was less than 5% in most studies and because all except one 41 were placebo-controlled, thus minimizing bias in the assessment of outcomes. Concealment of the randomization sequence was likely to have occurred in 12 studies as double-masked allocation, with identical antibiotics and placebo, was described.40–50,52,53 Four studies specified that pharmacists had dispensed the medications in identical bottles.47,48,50,52 In one study,51 tablets in numbered, sealed, opaque envelopes were picked from a box. In one study,44 tubes had coded allocation numbers that were kept by the data manager in a central location. Lastly, in 2 studies,45,52 allocations were kept in sealed opaque envelopes until the trial ended.

The type of placebo varied in the studies. Two studies used vitamin C tablets as placebo,50,51 whereas the other 11 studies used a placebo that was described as appearing identical to the antibiotic preparation.40,42–49,52,53

The selection criteria and antibiotic regimens are shown in Table 1. Twelve studies evaluated the effect of antibiotics principally among women with bacterial vaginosis.40,42–45,47–53 In 2 of these, some women also had Trichomonas vaginalis.43,44 Two studies evaluated the effect of antibiotics principally among women with Trichomonas vaginalis,41,46 and in one of these studies some women also had bacterial vaginosis.46 In some studies, screening was performed in the first trimester as well as in the second trimester of pregnancy.42,45,46,50 Enrollment and randomization were limited to women in the second trimester of pregnancy in all but 2 studies.45,50

Table 1. Selection Criteria and Antibiotic Regimens

Of the 11 studies that assessed antibiotics principally for bacterial vaginosis, 5 used metronidazole, 5 used clindamycin, and 1 study used a combination of metronidazole and erythromycin (Table 1). The 2 studies that assessed antibiotics principally for the treatment of Trichomonas vaginalis used metronidazole alone (Table 1). The antibiotic regimens used are outlined in Table 1. Baseline characteristics of antibiotic and control groups were similar in all studies (Table 2).

Table 2. Numbers Analyzed and Baseline Characteristics

Five studies, as well as the pooled data from these studies, reported that antibiotics reduced the risk of persistent bacterial vaginosis (RR 0.32, 95% CI 0.20–0.52, P < .001) (Table 3). 42–45,48

Table 3. Treatment of Bacterial Vaginosis With Any Antibiotic Compared With No Antibiotics

Eleven studies that assessed the effect of antibiotic treatment for women with bacterial vaginosis included data on rates of preterm birth at less than 37 weeks of gestation.42–47,49–53 When results for all women were pooled, there was no difference between the antibiotic and placebo groups on the risk of preterm birth at less than 37 weeks (RR 0.93, 95% CI 0.70–1.22, P = .6) (Table 3 and Fig. 1). Antibiotic treatment also did not reduce the risk of preterm birth at less than 35 weeks,42 less than 34 weeks,40,50,51 less than 32 weeks,42,44,52 or less than 28 weeks 51 among women with bacterial vaginosis (Table 3).

Fig. 1. Effect of antibiotics compared with placebo for women with bacterial vaginosis on the outcome of preterm birth less than 37 weeks.Okun. Antibiotics for BV and Trichomonas vaginalis. Obstet Gynecol 2005.

Among women with bacterial vaginosis considered to be at high risk because of a previous spontaneous preterm birth, a previous preterm birth, or a previous second-trimester miscarriage, pooled data revealed no effect of antibiotics on preterm birth at less than 37 weeks (RR 0.75, 95% CI 0.45–1.24, P = .30) (Table 4). 42,43,47,50–52 Among women known to be positive for bacterial vaginosis at the time of both randomization and treatment (thus excluding women who spontaneously cleared the organism subsequent to screening), there was no effect of antibiotics on the risk of preterm birth at less than 37 weeks (RR 0.91, 95% CI 0.66–1.26, P = .60) (Table 4).42–44,46,47,50,51,53

Table 4. Treatment of Bacterial Vaginosis With Antibiotic Compared With No Antibiotics Among Subgroups of Women for the Outcome of Preterm Birth (< 37 wk)

When only those trials that assessed the effect of metronidazole for women with bacterial vaginosis were included, the pooled data showed no effect of metronidazole on preterm birth at less than 37 weeks (RR 1.08, 95% CI 0.73–1.59, P = .70) (Table 4).42,46,47,50,51 One study that assessed the combined effect of metronidazole and erythromycin did find a reduction in the risk of preterm birth at less than 37 weeks among women at high risk for preterm birth (RR 0.68, 95% CI 0.49–0.93, P = .01) (Table 4).43

Similarly, when only those trials that assessed the effect of clindamycin for women with bacterial vaginosis were included, there was no overall effect of clindamycin on preterm birth at less than 37 weeks (RR 0.82, 95% CI 0.45–1.50, P = .50).44,45,49,52,53 However, 2 recently published studies of clindamycin reported a reduction in preterm birth. Lamont et al 53 included 409 women randomized to a 2-day course of vaginal clindamycin compared with placebo between 13 and 20 weeks of gestation (RR 0.41, 95% CI 0.18–0.91, P = .03). Ugwumadu et al 52 included 485 women at a mean gestational age of 15.6 weeks and randomized them to a 5-day course of oral clindamycin compared with placebo (RR 0.39, 95% CI 0.20–0.76, P = .005).

Ugwumadu et al 52 found a significantly lower rate of second-trimester miscarriage in the clindamycin group (RR 0.20, 95% CI 0.04–0.89, P = .04) among women with bacterial vaginosis detected and treated between 12 and 16 weeks of gestation (Table 4). Antibiotic treatment did not reduce the risk of low birth weight (< 2,500 g),42,44,49,50,52,53 or very low birth weight (< 1,500 g) infants,42,52,53 admission to a NICU,47,52 perinatal death,51,52,53 prelabor rupture of the membranes (RR 0.57 95% CI 0.22–1.47, P = .06),50 or peripartum infection among women with bacterial vaginosis (Table 4).45

Three studies reported on this outcome,41,43,46 but the Ross and Van Middlekoop 41 study sustained a greater than 20% loss to follow-up in the untreated group, and the data were therefore not included in our meta-analysis. The pooled data from the 2 remaining studies 43,46 demonstrated a significant reduction in the persistence of Trichomonas vaginalis with antibiotic treatment (RR 0.18, 95% CI 0.07–0.48, P < .001).

Of the 2 studies that assessed the effect of antibiotics for women with Trichomonas vaginalis,41,46 only the Klebanoff et al 46 study reported on the rates of preterm birth at less than 37 weeks of gestation. In that study, metronidazole was associated with an increased risk of preterm birth at less than 37 weeks (RR 1.78, 95% CI 1.19–2.66, P = .005). There was no significant effect of metronidazole on risk of preterm birth at less than 35 weeks (RR 1.30, 95% CI 0.74–2.29, P = .40) or at less than 32 weeks of gestation (RR 1.33, 95% CI 0.63–2.83, P = .50).

Among women with Trichomonas vaginalis who were at high risk for preterm delivery, Klebanoff et al 46 reported that metronidazole was associated with a higher risk of preterm birth at less than 37 weeks (RR 1.84, 95% CI 1.07–3.18, P = .03). The Ross and Van Middlekoop 41 study did not report on rates of preterm birth but did find that the mean gestational age was not significantly different between women treated with metronidazole compared with placebo (n = 200).

No effect was found in the 2 studies assessing the effect of metronidazole on the risk of low birth weight (< 2,500 g) (RR 1.31, 95% CI 0.92–1.87, P = .61).41,46 Klebanoff et al 46 assessed the effect of metronidazole for Trichomonas vaginalis in terms of other outcomes. Metronidazole had no effect on the risk of very low birth weight (<1,500 g) (RR 1.42, 95% CI 0.68–2.98, P = .40), admission to a NICU (RR 1.21, 95% CI 0.84–1.75, P = .30), neonatal sepsis (RR 1.17, 95% CI 0.74–1.86, P = .50), perinatal death (RR 1.28, 95% CI 0.52–3.14, P = .60), chorioamnionitis (RR 0.93, 95% CI 0.54–1.59, P = .80), or endometritis (RR 1.00, 95% CI 0.46–2.15, P = 1.0).

Significant heterogeneity was found when data were pooled from trials for all women with bacterial vaginosis who were treated with any antibiotic compared with no antibiotic for the outcomes of persistent bacterial vaginosis (P < .001) and preterm birth at less than 37 weeks of gestation. For the outcome of preterm birth at less than 37 weeks, the heterogeneity persisted among both unselected women (P < .001) and women at high risk of preterm birth (P = .001). Significant heterogeneity was also found when data were pooled from trials for only women known to have bacterial vaginosis at both randomization and treatment for the outcome of preterm birth at less than 37 weeks (P < .001), for all women with bacterial vaginosis treated only with metronidazole (P = .002), and for those treated only with clindamycin (P = .007).

We found that antibiotic treatment for women with bacterial vaginosis consistently reduced the risk of persistent bacterial vaginosis, compared with treatment with placebo or no treatment. However, we did not find that antibiotic treatment with either metronidazole or clindamycin reduced the risk of preterm birth at less than 37 weeks, either among all pregnant women with bacterial vaginosis or among any subgroup examined, including those women at high risk of preterm birth. We found no significant reduction in risk of other adverse perinatal outcomes with antibiotic treatment for women with bacterial vaginosis. We believe these findings are valid because the review selected for all high-quality trials that were published as full reports before December 31, 2003.

There have been 3 recently published systematic reviews on antibiotic treatment of bacterial vaginosis, however, that have disagreed with these conclusions.11–13 Guise et al 11 included 7 trials, all of which were included in our analysis.40,42–44,47,49,50 However, 7 studies in our analysis were not included in the review by Guise.41,45,46,48,51,52,53 Four of the studies we included were published after the Guise review was undertaken,45,51,52,53 1 study reported on eradication of bacterial vaginosis only,48 and the remaining 2 reported principally on the outcomes of treatment in women with Trichomonas vaginalis rather than bacterial vaginosis.41,46

The review by Leitich et al 12 included 10 trials,37, 40,42–45,47,49,50,55 two of which we excluded.37,55 One of these studied patients in preterm labor,55 which was one of our exclusion criteria, and another included data from a duplicate publication.37 Leitich did not include 6 studies that we included.41,46,48,51–53 Two of these reported outcomes in women with Trichomonas vaginalis.41,46 One reported on the primary outcome of eradication of bacterial vaginosis.48 Three studies were published 51–53 after the Leitich review was undertaken.

The most recent Cochrane review included 10 trials (Kekki M, Kurki T, Kurkinen-Raty M, Pelkonen J, Paavonen J. Recurrent bacterial vaginosis in pregnancy predisposes to infectious morbidity: a double-blind, placebo-controlled multicenter intervention trial with vaginal clindamycin [abstract]. Int J Gynecol Obstet 1999;67 suppl 2:S42; Porter K, Rambo D, Jazayeri A, Jazayeri M, Prien S. Prospective randomized trial of once versus twice a day metronidazole-vaginal in obstetrical population identified with bacterial vaginosis [abstract]. Am J Obstet Gynecol 2001;184:S166),18,40,42–44,47,50,51 7 of which were included in our analysis.40,42–44,47,50,51 Three trials included in the Cochrane review were excluded in our review for the following reasons: one had an excessive dropout rate,18 one was the abstract (Kekki et al, 1999) of a subsequently published trial,45 and one was published in abstract form only (Porter et al, 2001). This latter trial did not include a placebo group but compared once daily with twice daily metronidazole. The 7 studies in our analysis not included in the Cochrane review included 2 studies of women with Trichomonas vaginalis,41,46 the full trial 45 of one of their included abstracts (Kekki et al, 1999), 1 study reporting on the outcome of eradication of bacterial vaginosis only,48 one study that they stated is “awaiting assessment,”49 and 2 studies published after this review was undertaken.52,53

All 3 reviews agree that there is no demonstrated benefit from antibiotic treatment of women with bacterial vaginosis who are not at high risk of preterm birth. However, all 3 suggest that there is potential benefit from screening and treatment of high-risk women. Guise et al 11 based this conclusion on pooled data from 3 trials (n = 305)43,47,50 demonstrating a reduced incidence of preterm birth at less than 37 weeks, as well as pooled data from two of these studies (n = 128)47,50 demonstrating a reduced incidence of preterm premature rupture of the membranes and low birth weight. They propose that screening and treatment of a “more selected” high-risk group (previous preterm delivery and bacterial vaginosis) may reduce the incidence of preterm birth. Leitich et al 12 also propose treatment of high-risk women with an oral preparation for longer duration, based on those studies that demonstrate benefit.43,47,50 Finally, McDonald et al (Cochrane review)13 conclude similarly that screening and treating high-risk women may lower the incidence of the more important outcomes of premature rupture of the membranes and low birth weight, again based on the studies of McDonald and Morales.47,50

Because our review is more current and we were unable to demonstrate any benefit to antibiotic treatment, even among various subgroups of women, we believe there is inadequate evidence to justify a policy of screening and antibiotic treatment of high-risk women with bacterial vaginosis to reduce the risk of preterm birth. Our review of antibiotic treatment of bacterial vaginosis includes 2 potentially important recent trials: those of Ugwumadu et al (using oral clindamycin for 5 days)52 and Lamont et al (using vaginal clindamycin for 3 days)53 (combined n = 894), which demonstrate a reduced incidence of preterm birth at less than 37 weeks in a general population of pregnant women. Both of these trials treated women earlier than many of the previously published trials. Thus, it is perhaps earlier treatment of bacterial vaginosis, and not necessarily a given preparation, such as the addition of erythromycin 43 or treatment of a longer duration,50 that alters birth outcomes in both low- and high-risk pregnant women.

We restricted our review to studies that randomized women in the second or third trimester of pregnancy because that is the conventional time to consider antibiotic treatment for bacterial vaginosis and Trichomonas vaginalis. However, it is possible that initiation of treatment in the first trimester of pregnancy may have a higher likelihood of being effective. We are unaware of any randomized controlled trials that have evaluated treatment this early in pregnancy.

With regard to the treatment of asymptomatic Trichomonas vaginalis, the data available at this time suggest no benefit to treatment for the prevention of preterm birth. Of significance is the fact that one randomized controlled trial 46 and a subanalysis of a larger trial 39 indicate a harmful effect with metronidazole treatment. Although it is not clear why treatment of Trichomonas vaginalis with metronidazole would be harmful, it has been speculated that dying Trichomonas vaginalis organisms elicit an inflammatory response or release a virus from the organism that increases the risk of preterm birth.46

As with other systematic reviews, the trials included in this meta-analysis are limited by the significant heterogeneity of the studies. Sources of heterogeneity may include differences in the diagnostic criteria used for detecting bacterial vaginosis, differences in the timing and length of treatment, the different antibiotics used, and the differences in background risk profiles of the women enrolled. It is not clear how this heterogeneity affected the results, but we believe that earlier antibiotic treatment for women with bacterial vaginosis deserves further study.

Of importance, there was no evidence that treatment of bacterial vaginosis reduced perinatal or neonatal mortality or serious neonatal morbidity, which are more important outcomes than birth at 37 weeks. Justification for screening and treatment strategies should be based on beneficial effects of these more meaningful clinical outcomes and should include evidence that there are no harmful effects on the mother or fetus/neonate. None of the trials to date have adequately addressed these outcomes.

We conclude that there is no evidence to support screening and antibiotic treatment of pregnant women with bacterial vaginosis in the second or third trimester of pregnancy, neither in the general population nor in high-risk women. However, further randomized controlled trials of antibiotic treatment initiated early in pregnancy for bacterial vaginosis, which are powered to assess clinically important outcomes, are warranted. Finally, there is no evidence to support the treatment of asymptomatic Trichomonas vaginalis in pregnancy and some evidence that this treatment may be harmful.

1. Hack M, Fanaroff AA. Outcomes of extremely premature infants: a perinatal dilemma. N Engl J Med 1993;329:1649–50. [Context Link]

2. Mueller-Heubach E, Rubinstein DN, Schwarz SS. Histologic chorioamnionitis and preterm delivery in different patient populations. Obstet Gynecol 1990;75:622–6. [Context Link]

3. Romero R, Mazor J. Infection and preterm labor. Clin Obstet Gynecol 1988;31:553–84. Bibliographic Links [Context Link]

4. Lamont RF, Taylor-Robinson D, Newman M, Wigglesworth J, Elder MG. Spontaneous early preterm labour associated with abnormal genital bacterial colonization. Br J Obstet Gynaecol 1986;93:804–10. Bibliographic Links [Context Link]

5. Gravett MG, Nelson HP, DeRouen T, Critchlow C, Eschenbach DA, Holmes KK. Independent associations of bacterial vaginosis and Chlamydia trachomatis infection with adverse pregnancy outcome. JAMA 1986;256:1899–903. Bibliographic Links [Context Link]

6. Eschenbach DA, Hillier S, Critchlow C, Stevens C, DeRouen T, Holmes KK. Diagnosis and clinical manifestations of bacterial vaginosis. Am J Obstet Gynecol 1988;158:819–28. Bibliographic Links [Context Link]

7. McGregor JA, French JI, Richter R, Franco-Buff A, Johnson A, Hillier S, et al. Antenatal microbiologic and maternal risk factors associated with prematurity. Am J Obstet Gynecol 1990;163:1465–73. Bibliographic Links [Context Link]

8. Hillier SL, Nugent RP, Eschenbach DA, Krohn MA, Gibbs RS, Martin DH, et al. Association between bacterial vaginosis and preterm delivery of a low-birth-weight infant. The Vaginal Infections and Prematurity Study Group. N Engl J Med 1995;333:1737–42. Bibliographic Links [Context Link]

9. Cotch MF, Pastorek JG 2nd, Nugent RP, Hillier SL, Gibbs RS, Martin DH, et al. Trichomonas vaginalis associated with low birth weight and preterm delivery. The Vaginal Infections and Prematurity Group. Sex Transm Dis 1997;24:353–60. Ovid Full Text Bibliographic Links [Context Link]

10. Riduan JM, Joesoef MR, Hillier SL, Utomo B, Wiknojosastro G, Linnan M, Kandun N. Bacterial vaginosis and prematurity in Indonesia: association in early and late pregnancy. Am J Obstet Gynecol 1993;169:175–8. Buy Now Bibliographic Links [Context Link]

11. Guise JM, Mahon SM, Aickin M, Helfand M, Peipert JF, Westhoff C. Screening for bacterial vaginosis in pregnancy. Am J Prev Med 2001;20:62–72. [Context Link]

12. Leitich H, Brunbauer M, Bodner-Adler B, Kaider A, Egarter C, Husslein P. Antibiotic treatment of bacterial vaginosis in pregnancy: a meta-analysis. Am J Obstet Gynecol 2003;188:752–8. Buy Now Bibliographic Links [Context Link]

13. McDonald H, Brocklehurst P, Parsons J, Vigneswaran R. Antibiotics for treating bacterial vaginosis in pregnancy (Cochrane Review). In: The Cochrane Library, Issue 2, 2003. Oxford: Update Software. [Context Link]

14. Gulmezoglu AM. Interventions for trichomoniasis in pregnancy (Cochrane Review). In: The Cochrane Library, Issue 3, 2002. Oxford: Update Software. [Context Link]

15. McGregor JA, French JI, Parker R, Draper D, Patterson E, Jones W, et al. Prevention of premature birth by screening and treatment for common genital tract infections: results of a prospective controlled evaluation. Am J Obstet Gynecol 1995;173:157–67. Buy Now Bibliographic Links [Context Link]

16. Müller E, Berger K, Dennemark N, Oleen-Burkey M. Cost of bacterial vaginosis in pregnancy: decision analysis and cost evaluation of a clinical study in Germany. J Reprod Med 1999;44:807–14. Bibliographic Links [Context Link]

17. Dennemark N, Meyer-Wilmes M, Schlüter R. Screening and treatment of bacterial vaginosis in the early second trimester of pregnancy: a sufficient measure for prevention of preterm deliveries? Int J STD AIDS 1997;8(suppl):38–39. [Context Link]

18. Duff P, Lee ML, Hillier SL, Herd LM, Krohn MA, Eschenbach DA. Amoxicillin treatment of bacterial vaginosis during pregnancy. Obstet Gynecol 1991;77:431–5. Bibliographic Links [Context Link]

19. Paul VK, Singh M, Buckshee K. Erythromycin treatment of pregnant women to reduce the incidence of low birth weight and preterm deliveries. Int J Gynaecol Obstet 1998;62:87–8. [Context Link]

20. Rosenstein IJ, Morgan DJ, Lamont RF, Sheehan M, Dore CJ, Hay PE, et al. Effect of intravaginal clindamycin cream on pregnancy outcome and on abnormal vaginal microbial flora of pregnant women. Infect Dis Obstet Gynecol 2000;8:158–65. [Context Link]

21. Vermeulen GM, van Zwet AA, Bruinse HW. Changes in the vaginal flora after two percent clindamycin vaginal cream in women at high risk of spontaneous preterm birth. BJOG 2001;108:697–700. [Context Link]

22. McGregor JA, French JI, Richter R, Vuchetich M, Vaundeen B, Seo K, et al. Cervicovaginal microflora and pregnancy outcome: results of a double-blind, placebo-controlled trial of erythromycin treatment. Am J Obstet Gynecol 1990;163:1580–91. Bibliographic Links [Context Link]

23. Klebanoff MA, Regan JA, Rao AV, Nugent RP, Blackwelder WC, Eschenbach DA, et al. Outcome of the Vaginal Infections and Prematurity Study: results of a clinical trial of erythromycin among pregnant women colonized with group B streptococci. Am J Obstet Gynecol 1995;172:1540–5. Buy Now Bibliographic Links [Context Link]

24. Gray RH, Wabwire-Mangen F, Kigozi G, Sewankambo NK, Serwadda D, Moulton LH, et al. Randomized trial of presumptive sexually transmitted disease therapy during pregnancy in Rakai, Uganda. Am J Obstet Gynecol 2001;185:1209–17. Buy Now Bibliographic Links [Context Link]

25. Eschenbach DA, Nugent RP, Rao AV, Cotch MF, Gibbs RS, Lipscomb KA, et al. A randomized placebo-controlled trial of erythromycin for the treatment of Ureaplasma urealyticum to prevent premature delivery. Am J Obstet Gynecol 1991;164:734–42. Bibliographic Links [Context Link]

26. McCormack WM, Rosner B, Lee YH, Munoz A, Charles D, Kass EH. Effect on birth weight of erythromycin treatment of pregnant women. Obstet Gynecol 1987;69:202–7. Bibliographic Links [Context Link]

27. Schwebke JR, Weiss HL. Interrelationships of bacterial vaginosis and cervical inflammation. Sex Transm Dis 2002;29:59–64. Ovid Full Text Bibliographic Links [Context Link]

28. Schwebke JR. Asymptomatic bacterial vaginosis: response to therapy. Am J Obstet Gynecol 2000;183:1434–9. Buy Now Bibliographic Links [Context Link]

29. Ahmed-Jushuf IH, Shahmanesh M, Arya OP. The treatment of bacterial vaginosis with a 3-day course of 2 percent clindamycin cream: results of a multicentre, double blind, placebo controlled trial. Genitourin Med 1995;71:254–6. Bibliographic Links [Context Link]

30. Livengood CH, Thomason JL, Hill GB. Bacterial vaginosis: treatment with a topical intravaginal clindamycin phosphate. Obstet Gynecol 1990;76:118–23. Bibliographic Links [Context Link]

31. Ogasawara KK, Goodwin TM. Efficacy of azithromycin in reducing lower genital Ureaplasma urealyticum colonization in women at risk for preterm delivery. J Matern Fetal Med 1999;8:12–6. [Context Link]

32. Gjerdingen D, Fontaine P, Bixby M, Santilli J, Welsh J. The impact of regular vaginal pH screening on the diagnosis of bacterial vaginosis in pregnancy. J Fam Pract 2000;49:39–43. [Context Link]

33. French JI, McGregor JA, Draper D, Parker R, McFee J. Gestational bleeding, bacterial vaginosis, and common reproductive tract infections: risk for preterm birth and benefit of treatment. Obstet Gynecol 1999;93:715–24. Ovid Full Text Bibliographic Links [Context Link]

34. McDonald HM, O'Loughlin JA, Vigneswaran R, Jolley PT, Harvey JA, Bof A. Metronidazole treatment of bacterial vaginosis flora and its effect on preterm birth. Int J STD AIDS 1997;8(suppl):37–8. [Context Link]

35. Goldenberg RL, Klebanoff M, Carey JC, MacPherson C. Metronidazole treatment of women with a positive fetal fibronectin test result. Am J Obstet Gynecol 2001;185:485–6. Buy Now Bibliographic Links [Context Link]

36. Kurkinen-Räty M, Vuopala S, Koskela M, Kekki M, Kurki T, Paavonen J, et al. A randomised controlled trial of vaginal clindamycin for early pregnancy bacterial vaginosis. BJOG 2000;107:1427–32. [Context Link]

37. Odendaal HJ, Popov I, Schoeman J, Grové D. Preterm labour: is Mycoplasma hominis involved? S Afr Med J 2002;92:235–7. [Context Link]

38. Kigozi GG, Brahmbhatt H, Wabwire-Mangen F, Wawer MJ, Serwadda D, Sewankambo N, et al. Treatment of trichomonas in pregnancy and adverse outcomes of pregnancy: a subanalysis of a randomized trial in Rakai, Uganda. Am J Obstet Gynecol 2003;189:1398–400. [Context Link]

39. Wawer MJ, Sewankambo NK, Serwadda D, Quinn TC, Paxton LA, Kiwanuka N, et al. Control of sexually transmitted diseases for AIDS prevention in Uganda: a randomized community trial. Lancet 1999;353:525–35. Bibliographic Links [Context Link]

40. Vermeulen GM, Bruinse HW. Prophylactic administration of clindamycin 2% vaginal cream to reduce the incidence of spontaneous preterm birth in women with an increased recurrence risk: a randomized placebo-controlled double-blind trial. Br J Obstet Gynaecol 1999;106:652–7. Bibliographic Links [Context Link]

41. Ross SM, Van Middelkoop A. Trichomonas infection in pregnancy: does it affect perinatal outcome? S Afr Med J 1983;63:566–7. [Context Link]

42. Carey JC, Klebanoff MA, Hauth JC, Hillier SL, Thom EA, Ernest JM, et al. Metronidazole to prevent preterm delivery in pregnant women with asymptomatic bacterial vaginosis. N Engl J Med 2000;342:534–40. Bibliographic Links [Context Link]

43. Hauth JC, Goldenberg RL, Andrews WW, DuBard M, Copper RL. Reduced incidence of preterm delivery with metronidazole and erythromycin in women with bacterial vaginosis. N Engl J Med 1995;333:1732–6. Bibliographic Links [Context Link]

44. Joesoef MR, Hillier SL, Wiknjosastro G, Sumampouw H, Linnan M, Norojono W, et al. Intravaginal clindamycin treatment for bacterial vaginosis: effects on preterm delivery and low birth weight. Am J Obstet Gynecol 1995;173:1527–31. [Context Link]

45. Kekki M, Kurki T, Pelkonen J, Kurkinen-Räty M, Cacciatore B, Paavonen J. Vaginal clindamycin in preventing preterm birth and peripartal infections in asymptomatic women with bacterial vaginosis: a randomized, controlled trial. Obstet Gynecol 2001;97:643–8. Ovid Full Text Bibliographic Links [Context Link]

46. Klebanoff MA, Carey JC, Hauth JC, Hillier SL, Nugent RP, Thom EA, et al. Failure of metronidazole to prevent preterm delivery among pregnant women with asymptomatic Trichomonas vaginalis infection. N Engl J Med 2001;345:487–93. Bibliographic Links [Context Link]

47. McDonald HM, O'Loughlin JA, Vigneswaran R, Jolley PT, Harvey JA, Bof A, et al. Impact of metronidazole therapy on preterm birth in women with bacterial vaginosis flora (Gardnerella vaginalis): a randomised, placebo controlled trial. Br J Obstet Gynaecol 1997;104:1391–7. [Context Link]

48. McDonald HM, O'Loughlin JA, Vigneswaran R, Jolley PT, McDonald PJ. Bacterial vaginosis in pregnancy and efficacy of short-course oral metronidazole treatment: a randomized controlled trial. Obstet Gynecol 1994;84:343–8. Bibliographic Links [Context Link]

49. McGregor JA, French JI, Jones W, Milligan K, McKinney PF, Patterson E, et al. Bacterial vaginosis is associated with prematurity and vaginal fluid mucinase and sialidase: results of a controlled trial of topical clindamycin cream. Am J Obstet Gynecol 1994;170:1048–60. Buy Now Bibliographic Links [Context Link]

50. Morales WJ, Schorr S, Albritton J. Effect of metronidazole in patients with preterm birth in preceding pregnancy and bacterial vaginosis: a placebo-controlled, double-blind study. Am J Obstet Gynecol 1994;171:345–9. Buy Now Bibliographic Links [Context Link]

51. Odendaal HJ, Popov I, Schoeman J, Smith M, Grové D. Preterm labour: is bacterial vaginosis involved? S Afr Med J 2002;92:231–4. Bibliographic Links [Context Link]

52. Ugwumadu A, Manyonda I, Reid F, Hay P. Effect of early oral clindamycin on late miscarriage and preterm delivery in asymptomatic women with abnormal vaginal flora and bacterial vaginosis: a randomised controlled trial. Lancet 2003;361:983–8. Bibliographic Links [Context Link]

53. Lamont RF, Duncan SLB, Mandal D, Bassett P. Intravaginal clindamycin to reduce preterm birth in women with abnormal genital tract flora. Obstet Gynecol 2003;101:516–22. Ovid Full Text Bibliographic Links [Context Link]

54. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–88. Bibliographic Links [Context Link]

55. McGregor JA, French JL, Seo K. Adjunctive clindamycin therapy for preterm labor: result of a double-blind, placebo-controlled trial. Am J Obstet Gynecol 1991;165:867–75. Bibliographic Links [Context Link]