Probiotic and metronidazole treatment for recurrent bacterial vaginosis

Bacterial vaginosis is a polymicrobial syndrome. It arises when Lactobacillus bacteria, which are normally found in the vagina and produce hydrogen peroxide, are replaced by high concentrations of anaerobic bacteria, such as Gardnerella vaginalis and Mycoplasma hominis. It is frequently recurrent and may lead to pelvic inflammatory disease, postsurgical infections, increased risk of sexually transmitted infections and HIV, and preterm labor in pregnant women. Recurrent vaginosis in a patient is defined as 3 or more episodes occurring per annum [1]. Oral metronidazole and clindamycin vaginal cream is the treatment of choice, with a cure rate at 4 weeks of 60%–70% [2]. A randomized placebo-controlled trial for treatment of persistent bacterial vaginosis indicated that 0.75% metronidazole gel twice a week for 16 weeks followed by 12 weeks off therapy achieved a significant reduction in the recurrence rate of bacterial vaginosis [3]. Various trials have reported a positive association between absence of lactobacillus and recurrent bacterial vaginosis, but there is still no definite protocol for complete recovery [4]. We conducted a pilot study to compare the effectiveness of 0.75% metronidazole gel alone and in combination with an oral probiotic (Ecoflora; Tablets India Ltd, Chennai, India) in patients with recurrent bacterial vaginosis. The present pilot study included 16 patients who provided informed consent to undergo the treatment. All patients had had more than 3 episodes of bacterial vaginosis in the past year and had been treated repeatedly with 400 mg of oral metronidazole twice daily for 7 days with no success. The patients were allocated alternately into 2 groups. Group 1 received 0.75% metronidazole gel and were instructed to apply 5 g intravaginally twice per week for 3 months. Group 2 were instructed to take an oral probiotic tablet twice daily for 3 months in addition to following the same metronidazole regimen as for group 1. Bacterial vaginosis was assessed before and after treatment using the Amsel criteria; the presence of 3 out of 4 criteria was considered a confirmed diagnosis. Scoring based on Nugent criteria was also done, but the Amsel criteria were preferred to define cure rate. A definition of “fully cured” was considered when none of the 4 criteria was present, “improved” was defined when only 1 was present, “partial cure” when 2 criteria were present, and “failure” when 3 or 4 criteria were present. Patients attended after completion of a menstrual cycle and treatment was started. Use of the gel was restricted during menstruation. After each menstrual cycle the patients were evaluated using the Amsel criteria after 4, 8, and 12 weeks of treatment. The mean age of the 16 patients was 22.9 years and all were from a lower socioeconomic group (monthly income below Rs 5000). Comparative observations between the 2 groups of the treatment outcomes according to the Amsel criteria at 4, 8, and 12 weeks are shown in Table 1. All of the 8 (100%) patients taking the probiotic tablets (group 2) were defined as fully cured after 4 weeks of treatment. Prior to beginning the treatment, 3 patients from group 1 and 5 patients from group 2 had a Nugent score of more than 6, which was diagnostic of bacterial vaginosis. At 4 weeks, 2 patients (66.7%) from group 1 and 4 patients (80%) from group 2 had a score of less than 6. The Nugent score was the same at 8 weeks. At 12 weeks, 1 patient (33.3%) from group 1 and 3 patients (60%) from group 2 had a Nugent score of less than 6. Women with recurrent bacterial vaginosis have altered vaginal flora. If we try to replace the flora using a probiotic it will not be sufficient because the unnatural flora will prevent colonization. Therefore, it is necessary to fully kill the unnatural flora using a prolonged treatment of 0.75% metronidazole gel and then simultaneous treatment with a probiotic will improve the defense mechanism of the vagina. We preferred to use metronidazole gel rather than an oral tablet because we were interested in a local rather than a systemic effect. Further randomized controlled trials are needed to confirm these results. The authors have no conflicts of interest to disclose.