Bacterial vaginosis (BV) is the most common cause of vaginal symptoms among women of childbearing age.1 In the United States, it affects approximately 30% of women and more than 4 million women are treated for BV annually.2 Bacterial vaginosis increases risk of sexually transmitted, pelvic inflammatory disease, infertility, spontaneous abortions, and delivery a premature baby.3–6 It impairs quality of life, has a negative impact on job performance, and impairs intimate partner relations.7,8 One of the treatment approaches to the symptomatic BV, recommended by the Centers for Disease Control, includes metronidazole tablets 500 mg taken orally twice a day for 7 days.9 However, this protocol is associated with 58% recurrence rate with approximately more than 40% of cases occurring during the first 4 months.10
There is no consensus on the prevention of BV. In a series of small randomized trials, it has been shown that lactobacilli taken per os improve vaginal bacterial ecology and decrease recurrence rate of BV in heterogeneous groups of women.11–14 The rationale behind the oral use of lactobacilli has served as indirect evidence for the present understanding that colonic bacteria can translocate into the vagina.15 Women with BV are less likely to have rectal and vaginal colonization with Lactobacillus crispatus, a Lactobacillus species that is often absent in the vaginal microbiota of women with BV.16 Use of DNA hybridization techniques has shown the same genetic types of lactobacilli in the rectum and vagina.17,18
The effect of oral use of lactobacilli has not been yet assessed in a sufficiently powered randomized clinical trial. The objective of the study was to explore a role of oral intake of a mixture of 3 Lactobacillus species in recurrence of BV, recently treated with antibiotics.
SUBJECTS AND METHODS
This was a phase 2 randomized (1:1 allocation ratio) interventional parallel group prospective placebo controlled multicenter clinical study. It was performed from February to August 2018 at the following 7 clinical settings with gynecology units: 3 public tertiary care city or state hospitals (The 3rd Lviv Community Clinical Hospital, Lviv; Lviv Community Emergency Care Clinical Hospital, Lviv; Lviv State Center for the Reproductive Health, Lviv), 3 universities (Zaporizhzhia State Medical University, Zaporizhzhia; Odessa National Medical University, Odessa; Ivano-Frankivsk National Medical University, Ivano-Frankivsk); 1 private clinic (Medical Center “MediVit,” Ternopil). All included centers are located in Ukraine. The study received approval at the ethical committee at the Lviv Community Emergency Care Clinical Hospital (ref. #20-1.1. of September 28, 2017). Before enrollment, patients provided signed informed consent for participation in the trial. The privacy rights of participants were safeguarded throughout the study and the trial was fully compliant with the Declaration of Helsinki. The study was registered at ClinicalTrials.gov (Identifier: NCT03601429).
Women were included in the study if they signed the informed consent form, were 18 to 45 years old, able to swallow capsules, had recent BV cured with metronidazole tablets in a dose of 500 mg 2 times daily for 7 days, less than 48 hours after cure of BV, gave specific consent to communicate intimate history, agreed not to use spermicides for 4 months after starting the test dietary supplement (TDS), and stated the availability of a mobile phone as well as personal availability throughout the study. Bacterial vaginosis, or its recurrence, was defined as the presence of 3 of 4 Amsel criteria and one or combination of self-reported symptoms typical to the disease (vaginal discharge, vaginal odor).19 Recovery from BV was defined as absence of all self-reported symptoms plus 3 of 4 negative Amsel criteria.
Exclusion criteria included postmenopausal period, unprotected intercourse to achieve pregnancy, breast feeding, sexually transmitted disease (STD), vulvovaginal candidiasis, diabetes mellitus, inflammatory bowel disease, alcohol or drug addiction as suspected by the investigator, absence of a refrigerator in the home, known moderate to severe disease of any systems, difficulty in comprehending the study requirements as judged by physician, mucosal ulcerations, herpes-like lesions, cervicitis during vaginal examination, treatment of BV with measures other than metronidazole antibiotic therapy, use of any investigational drug within the previous 30 days, and use of drugs that suppress the immune system. Screening for STDs included polymerase chain reaction assays for chlamydiosis, gonorrhea, HIV, and medical records were reviewed for genital herpes, papillomatosis. There were no changes made to the eligibility criteria after the study commencement.
Study chronology is present in the Figure 1.
The patients with symptomatic BV entered metronidazole treatment period after preassessment of eligibility criteria. Within 48 hours after finishing metronidazole, if a woman had no symptoms and 3 of 4 negative Amsel criteria, she was invited for the baseline visit. During this visit, informed consent was documented with a signed and dated form, vaginal smears were evaluated for Amsel criteria and Nugent score,20 and recovery from BV was reassessed.
At the end of the baseline visit, the investigator randomly dispensed the TDS as described hereinafter, provided the 120-day study diary (120-DSD), and requested participants to report if symptoms of BV recurred.
Random numbers were generated by the Random Allocation Software v. 1.0.0 with no restrictions.21 Random allocation implemented sticking the labels with sequential random numbers to and sequential selection of the packs by the investigator. Random numbers were generated by the technician who did not participate in distribution of packs among centers or data analysis. The TDS looked the same and had the same smell and consistency. Investigators and patients were concealed which type of TDS was given.
The verum group received the TDS with a combination of lactobacilli in a dose of 5.4 billion (L. crispatus LMG S-29995, Lactobacillus brevis, and Lactobacillus acidophilus in proportion of 60%, 20%, and 20%, respectively; Lactogyn, Vésale Pharma, Belgium) 2 times daily for the first 7 days and 1 time daily for the next 8 to 120 days continuously after meals. The placebo group received a similar TDS (maltodextrin) without lactobacilli in the same manner.
At the baseline visit, women were advised to avoid any foods or dietary supplements that contained prebiotics or probiotics, vitamins, or trace elements during intake of the TDS. In the structured 120-DSD, women recorded TDS intake, symptoms of BV, and food intolerance events (abdominal distension, abdominal pain, bloating, constipation, diarrhea, nausea, pruritus, rash, urticaria). A woman was considered compliant with TDS intake, if she received more than 90% of the planned number of doses. The compliance was assessed by counting remaining capsules in bottles returned during the final visit.
On every visit, the investigator asked about symptoms of BV and performed a standard gynecologic examination, and vaginal discharge was analyzed for Amsel criteria and Nugent score. Between the study visits, an SMS notification to switch from 2 capsules to 1 capsule per day dosing mode was sent. Two standard phone calls were performed to check if any of BV symptoms appeared after previous visits, correctness of records in the 120-DSD, compliance with TDS intake, number of doses missed, food intolerance events, storage of the TDS, and to remind to come back for the interim or final visits.
The primary outcome measure was percent of recurrences of BV in the verum and placebo groups during 16 weeks of intervention. The secondary outcomes were Nugent scores at weeks 0 (baseline visit), 8 (interim visit), 16 (final visit), time to recurrence of BV, and rates of survival without BV. There were no changes made to the outcome measures after the study commencement.
A 43% recurrence rate of BV for the first 3 months after acute manifestation of the disease has been shown previously.10 Assuming 20% absolute reduction due to use of the TDS, sample size was calculated as of 83 women per arm (166 two arms) with 1-β error 0.80 and type 1 error based on 0.05 hypothesis for a 2-tailed distribution. Although the sample size was calculated based on the previously mentioned reference, the time frame was extended to 4 months as another study showed 2 distinct waves of recurrences at weeks 8th and 16th.22 There were no interim analyses and stopping guidelines in this study.
Absolute number and percent were used to describe race, education, marital status, employment, smoking status, gynecologic diseases, features of intimate habits, food intolerance events, items of Amsel criteria, and range for Nugent score. Mean and standard deviation described age, time to recurrence of BV. Normality of data was assessed with Kolmogorov-Smirnov test. Difference between the groups in percent data was assessed in Z test for proportions and between the means in 2-tailed t test for independent variables. Kaplan-Meier plot was used to describe survival without BV, and Cox F test assessed significance in survival rates in the groups. p values of less than 0.05 were considered significant. Tests were performed with the Statistica 9 (StatSoft., Inc, Tulsa, OK) software package.
The study was supported by Vésale Pharma (Belgium). The sponsor did not participate in the study design, analysis, construction, or decision as to the submission of the article.
Four hundred forty-eight women with BV were enrolled in the prestudy metronidazole treatment. All women had 4 of 4 positive Amsel criteria for BV before antibiotic therapy. Forty-five women of 276 excluded (16%) failed to achieve a cure of BV with metronidazole (see Figure 2).
At the baseline visit, 172 women were cured of BV and randomly allocated to the verum or placebo supplementation. During the study, 3 women were lost to follow-up with no further wish to participate. Three women were excluded from analysis because of poor compliance with TDS intake, use of spermicide, or antibiotic. Enrollment of patients was stopped when data from 166 women could be evaluated for the study purposes.
The verum and placebo groups were comparable on major demographic and medical characteristics (see Table 1).
Most participants were young women, half of them graduated university. The remainder had secondary education with a small portion being current students or having interrupted university education. Approximately, the half of the study population was married, one tenth divorced, and approximately 20% lived single. Employment status was variably reported, but most of women were employed full-time. History of BV, genital papillomatosis, and abnormal cervical cytology were similar in both groups. History of STD is not mentioned in the Table 1 as it was negative in all women, serving an exclusion criterion. Almost all women had one male sexual partner. Women reported more than one male partner or female partner very rarely. Pregnancy control measures were comparable in the groups.
At baseline visit, all included women had no symptoms and 3 of 4 negative Amsel criteria (see Table 2).
At the baseline visit, no women had clues cells in the vaginal smear. Few women occasionally presented with one of other Amsel criteria. Most of the participants had Nugent score 0 to 3 with no significant difference between groups on all the parameters. At the interim visit, the verum and the placebo groups were disproportional in number of patients with each positive Amsel criterion, except for amine test. Bacterial vaginosis was diagnosed 3 times less often in the verum group that was associated with greater proportion of women with Nugent score 7 to 10 in the placebo group. At the final visit, BV recurred more often in the placebo group, in which the number of women with Nugent score 0 to 3 was also lesser.
The mean (SD) time to recurrence of BV was 97.3 (26.7) days in the verum group versus 74.7 (27.7) days in the placebo group (p = .014). Survival analysis showed that women in the verum group survived longer without BV than did controls (see Figure 3).
Cox F test showed that time dependence of survival rates was significant (p = .018).
Nineteen (23.2%) of 82 women from the verum and 17 (20.2%) of 84 women from the placebo group missed at least one dose of the TDS (not significant [NS]). The mean (SD) number of doses missed was 2.1 (1.3) and 2.4 (1.6) in the verum and placebo groups, respectively (NS). All evaluable women received more than 90% of the course dose and were assumed compliant.
Both verum and placebo TDS were tolerated well. In the verum group 1 (1.2%) of 82 women had abdominal pain and 1 (1.2%) of 84 women had constipation. In the placebo group, a total of 6 intolerance events were documented by one for every item (bloating, abdominal pain, nausea, distension, pruritus, constipation). The intensity of the events was mild or moderate with no consistent relationship attributed to the TDS.
Our study showed that 4-month oral use of lactobacilli, started immediately after a cure by antibiotic therapy, significantly decreased the number of recurrences of BV. Previous studies differed by a smaller sample size (22–60 patients), a shorter-term (7–28 days) application of lactobacilli, variations in inclusion criteria (pregnant with herpetic infection, transsexual, postmenopausal undergoing breast cancer chemotherapy), or simultaneous initiation antibiotic therapy with lactobacilli.11–14 In all listed trials, benefits were achieved independently of the type of the protocol and included decrease in BV incidence, reduction of symptoms, and improvement of vaginal bacterial ecology.
A total dose and daily dosing of lactobacilli across the studies were also different. Kaufmann et al.11 administered lactobacilli 2.5 × 109 CFU x 2 times daily for 7 days (total dose 35 billion); Anoshina et al.12 400 × 109 CFU x 1 time daily for 7 days (total dose 280 billion); Marschalek et al.13 2.5 × 109 CFU x 2 times daily for 14 days (total dose 70 billion); Laue et al.14 1.25 × 109 CFU x 2 times daily for 28 days (total dose 70 billion). In our study, lactobacilli were provided in a total dose of approximately 686 billion, which was higher than the total dose in the previous studies. Double daily dosing during the first week could potentially have helped faster restore vaginal colonization, and our longer-lasting supplementation could have increased chance of colonization; however, this effect was not specifically evaluated. However, repeat use of Nugent score, which is based on counting the number of morphotypes potentially beneficial or pathogenic bacteria, allows us to speculate that long-term oral use of lactobacilli can influence vaginal microecology and increased concentration of large gram-positive rods, supposed to be of Lactobacillus genus.
As with our study, all referenced intervention studies used L. crispatus strains.11,13,14 Interest in this Lactobacillus species stems from the earlier epidemiologic finding that L. crispatus is the most common commensal flora in vagina of healthy women.23 Choice of L. crispatus for formulation was prompted by its direct relationship to BV. When L. crispatus was present in the rectum and vagina, BV was diagnosed in 9% of women, whereas its absence increased prevalence of BV up to 44%.18 Domination of L. crispatus was effective in suppression of BV-associated anaerobes. More than other lactobacilli, L. crispatus produced lactic acid and other antimicrobial substances.24
We documented variable but significantly fewer number of women with vaginal discharge, clue cells, or increased pH at the follow-up visits. Similar variability was found in the study by Laue et al.,14 who reported Amsel criteria separately and noted their uneven change of the individual items to the intervention. Investigators did not find significant change in pH of vaginal discharge, whereas amine tests were more likely negative in the verum group. Another discrepancy in the referenced article occurred along assessment of Nugent score, which was not different between the groups at the end of intervention at the background of significant decrease of clue cell positive smears.14 At the interim visit, we found significant decrease in percent of smears with Nugent score 7 to 10 and insignificant increase smears with the score 0 to 3 in the verum group, and vice versa during the final visit. This discrepancy could be resolved with the modified Nugent criteria, when scores 4 to 6 were used in combination with presence of clue cells for diagnosis of BV. This approach allowed us to find significantly less prevalence of BV in the verum group at both interim and final visits (data not shown). The selected Amsel criteria together with clinical manifestation of BV consistently showed statistical difference between the groups at every follow-up visit.
In our study, the verum supplementation with lactobacilli prolonged mean time to recurrence of BV by 22.6 days. Similarly, in the study by Bohbot et al.25 who performed vaginal applications of lactobacilli for 14 days immediately after metronidazole therapy and for 14 days after the end of 3 consecutive menstruation circles (approximately 3.5 months), time to recurrence was longer by 24.6 days in the verum group. In the mentioned study, the recurrence rates, 16 (41.0%) of 39 patients in the placebo group versus 8 (20.5%) of 39 patients in the verum group, were somewhat greater than these in our study, 27 (32.1%) of 84 patients in the placebo versus 15 (18.3%) of 82 patients in the verum group, respectively.
The strength of our study was its randomized double-blind controlled prospective parallel group design with involvement of 7 clinical centers and predefined sample size. This allowed adequate evaluation of a primary outcome measure, a percent of recurrences of BV in the verum and placebo groups. The most meaningful limitation of our study was use of less strict criteria to diagnose BV compared with others. In the present study with dietary supplement, BV was defined as at least 3 4 Amsel associated with symptoms, whereas recent guidelines for industry issued by the Food and Drug Administration recommended to consider 4 of 4 criteria for reliable diagnosis of BV.26 After re-evaluation of the database, we found that before commencement of metronidazole treatment, all patients had 4 of 4 Amsel criteria being positive. The same guidelines suggested definition of clinical cure as normalization of vaginal discharge, a negative whiff test, and absence of clue cells on microscopic examination of the saline wet mount. We defined cure after metronidazole therapy as absence more than 3 of any 4 Amsel criteria and symptoms of BV that are not in complete accordance with Food and Drug Administration recommendations for evaluation of drugs. Another limitation of the study was also the absence of microbiological end points describing evolution of the colonic and vaginal flora before and during follow-up visits. Indeed, 2 of 4 trials exploring oral use of lactobacilli in BV, incorporated such an analysis and reported the increase in lactobacilli count in vagina.11,12 At the same time, we measured morphotypes of bacteria that might serve a surrogate end point instead of detailed identification of bacterial species. Another limitation was absence of description of types (vaginal, oral, anal) and frequency of sexual activity that also might be different in the groups and influence the study outcomes.27 Long-term benefits of probiotic supplementation remains unknown as follow-up for women was 4 months, as we targeted at the early recurrences of BV at the background of the short-term TDS intake. We did not assess benefits of probiotic supplementation after cessation of TDS intake.
The results of this study indicate that 4-month use of lactobacilli in a dose of 5.4 billion, L. crispatus LMG S-29995, L. brevis, and L. acidophilus in proportion of 60%, 20%, and 20%, respectively, is safe, decreases significantly rates of exacerbations of BV, and prolongs time to the recurrence in fertile age women.
1. Allsworth JE, Peipert JF. Prevalence of bacterial vaginosis
: 2001–2004 National Health and Nutrition Examination Survey data. Obstet Gynecol
2. Koumans EH, Sternberg M, Bruce C, et al. The prevalence of bacterial vaginosis
in the United States, 2001-2004; associations with symptoms, sexual behaviors, and reproductive health. Sex Transm Dis
3. Cherpes TL, Meyn LA, Krohn MA, et al. Association between acquisition of herpes simplex virus type 2 in women and bacterial vaginosis
. Clin Infect Dis
4. Wiesenfeld HC, Hillier SL, Krohn MA, et al. Bacterial vaginosis
is a strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection. Clin Infect Dis
5. Cohen CR, Lingappa JR, Baeten JM, et al. Bacterial vaginosis
associated with increased risk of female-to-male HIV-1 transmission: a prospective cohort analysis among African couples. PLoS Med
6. Donders GG, Van Calsteren K, Bellen G, et al. Predictive value for preterm birth of abnormal vaginal flora, bacterial vaginosis
and aerobic vaginitis during the first trimester of pregnancy. BJOG
7. Payne SC, Cromer PR, Stanek MK, et al. Evidence of African-American women's frustrations with chronic, recurrent bacterial vaginosis
. J Am Acad Nurse Pract
8. Bilardi JE, Walker S, Temple-Smith M, et al. The burden of bacterial vaginosis
: women's experience of the physical, emotional, sexual and social impact of living with recurrent bacterial vaginosis
. PLoS One
9. Centers for disease control and prevention. 2015 Sexually Transmitted Diseases Treatment Guidelines. Bacterial vaginosis
. Available at: https://www.cdc.gov/std/tg2015/bv.htm
. Accessed July 18, 2018.
10. Bradshaw CS, Morton AN, Hocking J, et al. High recurrence rates of bacterial vaginosis
over the course of 12 months after oral metronidazole therapy and factors associated with recurrence. J Infect Dis
11. Kaufmann U, Domig KJ, Lippitsch CI, et al. Ability of an orally administered lactobacilli preparation to improve the quality of the neovaginal microflora in male to female transsexual women. Eur J Obstet Gynecol Reprod Biol
12. Anoshina TM. Role of microbiota correction in complex treatment of pregnant women with herpesvirus infection. Perinatol Pediatr
13. Marschalek J, Farr A, Marschalek ML, et al. Influence of orally administered probiotic Lactobacillus
strains on vaginal microbiota in women with breast cancer during chemotherapy: a randomized placebo-controlled double-blinded pilot study. Breast Care
14. Laue C, Papazova E, Liesegang A, et al. Effect of a yoghurt drink containing Lactobacillus
strains on bacterial vaginosis
in women – a double blind, randomised, controlled clinical pilot trial. Benef Microbes
15. Cribby S, Taylor M, Reid G. Vaginal microbiota and the use of probiotics. Interdiscip Perspect Infect Dis
2008;2008:256490. Published online 2009 Mar 29. Available at: https://www.ncbi.nlm.nih.gov/pmc/
articles/PMC2662373. Accessed September 11, 2019.
16. Marrazzo JM, Fiedler TL, Srinivasan S, et al. Extravaginal reservoirs of vaginal bacteria as risk factors for incident bacterial vaginosis
. J Infect Dis
17. Ahrne S, Nobaek S, Jeppsson B, et al. The normal Lactobacillus
flora of healthy human rectal and oral mucosa. J Appl Microbiol
18. Song Y, Kato N, Matsumiya Y, et al. Identification of and hydrogen peroxide production by fecal and vaginal lactobacilli isolated from Japanese women and newborn infants. J Clin Microbiol
19. Amsel R, Totten PA, Spiegel CA, et al. Nonspecific vaginitis: diagnostic criteria and microbial and epidemiologic associations. Am J Med
20. Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial vaginosis
is improved by a standardized method of gram stain interpretation. J Clin Microbiol
21. Random Allocation Software. Available at: https://random-allocation-software.software.informer.com/1.0/
. Assessed July 18, 2018.
22. Turner AN, Carr Reese P, Fields KS, et al. A blinded, randomized controlled trial of high-dose vitamin D supplementation to reduce recurrence of bacterial vaginosis
. Am J Obstet Gynecol
23. Antonio MA, Hawes SE, Hillier SL. The identification of vaginal Lactobacillus
species and the demographic and microbiologic characteristics of women colonized by these species. J Infect Dis
24. Tachedjian G, Aldunate M, Bradshaw CS, et al. The role of lactic acid production by probiotic Lactobacillus
species in vaginal health. Res Microb
25. Bohbot JM, Daraï E, Bretelle F, et al. Efficacy and safety of vaginally administered lyophilized Lactobacillus
crispatus IP 174178 in the prevention of bacterial vaginosis
recurrence. J Gynecol Obstet Hum Reprod
26. Food and Drug Administration, Center for Drug Evaluation and Research. Bacterial vaginosis
: developing drugs for treatment guidance for industry. [Docket Number: FDA-2016-D-1659]. Available at: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/bacterial-vaginosis-developing-drugs-treatment-guidance-industry
. Accessed July 18, 2018.
27. Bradshaw CS, Vodstrcil LA, Hocking JS, et al. Recurrence of bacterial vaginosis
is significantly associated with posttreatment sexual activities and hormonal contraceptive use. Clin Infect Dis