How to Treat Bacterial Vaginosis and Stop It Coming Back

Bacterial vaginosis (BV) clears up with antibiotics in most cases, but the real challenge is keeping it gone. Roughly half of women who are successfully treated see symptoms return within a few months. That frustrating cycle has a biological explanation rooted in biofilm, menstrual-cycle timing, and sexual partners, and the strategies that actually reduce recurrence go well beyond the initial prescription.

First-Line Antibiotic Options

The standard treatments for BV are metronidazole and clindamycin, available in both oral and vaginal forms. A randomized double-blind trial found that vaginal metronidazole gel was as effective as oral metronidazole tablets, with a cure rate around 92% for the vaginal route compared with about 90% for oral, while producing fewer side effects like nausea and metallic taste.¹ Clindamycin vaginal cream performs similarly: one trial comparing it head-to-head with oral metronidazole found cure rates of about 68% and 67% respectively, with no significant difference between the two.² An earlier, smaller study showed a similar pattern, with clindamycin cream at 72% and oral metronidazole at 87%, though both groups dropped to about 61% cure a month later.³

The practical takeaway is that if you prefer not to take oral pills, the vaginal versions work just as well. The choice between metronidazole and clindamycin mostly comes down to availability, cost, and side-effect preference. Oral metronidazole interacts with alcohol, which some women find inconvenient during a treatment course. Vaginal clindamycin can weaken latex condoms and diaphragms, so timing matters if those are your contraception.

A newer option is secnidazole, a single-dose oral granule. A systematic review found that a single 2-gram dose of secnidazole was comparable in clinical cure to a five-day course of metronidazole taken twice daily.⁴ In a randomized controlled trial, the clinical cure rate for the 2-gram secnidazole group was about 68% compared with roughly 18% for placebo.⁵ The appeal is obvious: one dose, one time, done. For women who struggle with multi-day regimens, that convenience can improve adherence.

Why BV Keeps Coming Back

Understanding recurrence requires looking at what antibiotics cannot fully reach. The bacteria involved in BV, particularly Gardnerella species, form a polymicrobial biofilm that adheres to the vaginal lining.⁶ Think of biofilm as a protective shell: the bacteria embed themselves in it, and antibiotics can kill the free-floating organisms while leaving the embedded ones alive. When you stop the antibiotic, surviving bacteria within the biofilm repopulate the vagina within weeks.

Healthy vaginal flora is normally dominated by Lactobacillus species, especially Lactobacillus crispatus or Lactobacillus iners, which keep the environment acidic and hostile to invaders. In BV, this community shifts to a high-diversity mix of anaerobic bacteria including Gardnerella vaginalis and members of the Lachnospiraceae and Leptotrichiaceae families.⁷ Antibiotics knock back this anaerobic overgrowth, but they do not automatically restore the Lactobacillus population. If Lactobacillus does not recolonize quickly enough, the BV-associated bacteria creep back in.

The Menstrual Cycle Connection

If you have noticed that your BV symptoms flare around your period, you are not imagining it. Research has found that BV recurrence clusters within the first seven to ten days of the menstrual cycle and that the condition resolves spontaneously most often at mid-cycle.⁸ A study tracking women on suppressive therapy confirmed that most recurrences happened within ten days of menstruation.⁹ Menstrual blood raises vaginal pH and introduces iron that anaerobic bacteria thrive on, temporarily weakening the Lactobacillus-dominated environment that keeps BV in check.

This pattern has practical implications for maintenance therapy. Some clinicians time suppressive antibiotic doses around menstruation rather than using them throughout the entire month, concentrating protection during the window when the vaginal environment is most vulnerable.

Does Treating Your Male Partner Help?

This is one of the most debated questions in BV management, and the evidence just got more complicated. There is solid evidence that BV-associated bacteria are exchanged between sexual partners during sex.¹⁰ Gardnerella and other BV organisms have been found on the penile skin and under the foreskin of male partners of women with BV. The logic is straightforward: if you treat the woman but leave her partner carrying the same bacteria, reinfection during sex is likely.

In 2025, a landmark Australian trial (the StepUp RCT) was published in the New England Journal of Medicine. The trial was stopped early by its safety monitoring board because treating only the woman was clearly inferior to treating both the woman and her male partner. The combination of oral and topical antibiotics for the male partner resulted in substantially lower recurrence of BV within 12 weeks.¹¹

However, a subsequent meta-analysis pooling data from multiple randomized trials found no overall significant reduction in BV recurrence with male partner treatment compared with standard female-only therapy. When the authors excluded the StepUp trial, which was the only one to use combined oral and topical treatment for the male partner, heterogeneity disappeared and the remaining trials showed a neutral effect.¹² The implication is that earlier partner treatment trials may have failed not because the concept is wrong but because they used inadequate regimens, giving the male partner oral antibiotics alone without topical penile treatment. In practical terms, the combined oral-plus-topical approach used in the StepUp trial seems to be the specific protocol that works. Giving a male partner a standard course of oral metronidazole by itself probably is not enough.

This is still an evolving area. If you are in a stable sexual partnership and your BV keeps recurring, it is worth discussing concurrent partner treatment with your clinician, specifically the dual-route approach.

Suppressive Maintenance Therapy

For women with frequent recurrences, the most studied long-term strategy is suppressive vaginal metronidazole gel, applied twice a week after the initial treatment course. In a randomized trial, women using twice-weekly metronidazole gel had a recurrence rate of about 26% during the suppressive phase compared with roughly 59% on placebo.¹³ The probability of staying BV-free was about 70% on treatment versus 39% on placebo. After stopping, though, the protective effect faded: by 28 weeks, only about a third of the metronidazole group remained cured. One known downside is that prolonged vaginal antibiotic use increases the risk of vaginal yeast infections.

A longer-term suppressive regimen using maintenance metronidazole gel reported that about 70% of compliant patients remained free of symptomatic BV at both 6 and 12 months of follow-up.¹⁴ High-dose vaginal metronidazole suppositories have also been explored, with rare clinical recurrence during the period of active use, though recurrence climbed once suppression stopped.¹⁵ The overall picture is that suppressive therapy works well while you are on it but does not permanently reset the vaginal environment for most women. It buys time, and for some that time is enough for their Lactobacillus population to reestablish itself.

Boric Acid as an Add-On Treatment

Vaginal boric acid suppositories have gained popularity as an adjunct for recurrent BV, particularly targeting biofilm. The rationale is that boric acid disrupts the biofilm matrix that antibiotics cannot fully penetrate, making the bacteria more accessible to treatment.¹⁶ A retrospective study of women with recurrent BV (averaging about four prior episodes) who were treated with 600 mg of intravaginal boric acid daily for 14 days found meaningful improvements: roughly 89% achieved a normal Nugent score, and the presence of clue cells dropped dramatically.¹⁷

Boric acid is not an antibiotic, so it works by a different mechanism. It lowers vaginal pH and has antimicrobial and anti-biofilm properties. It is typically used either after a standard antibiotic course or as part of a multi-phase regimen: antibiotics first to knock down the infection, boric acid to help clear remaining biofilm, then maintenance therapy to prevent regrowth. Boric acid is toxic if swallowed, so vaginal suppositories should never be taken orally and must be kept away from children. It should also not be used during pregnancy.

Probiotics and Restoring Lactobacillus

The idea behind probiotic therapy is appealing: if the problem is that healthy Lactobacillus bacteria are missing, why not put them back? A randomized double-blind trial found that both oral and vaginal capsules containing Lactobacillus crispatus strains reduced BV signs and symptoms, improved Nugent scores, and reduced odor and discharge. The vaginal probiotics specifically increased Lactobacillus counts while decreasing some BV-related bacteria.¹⁸

The catch is that not all probiotic products are created equal. Many over-the-counter vaginal probiotics contain strains that have never been tested for BV specifically, or they contain Lactobacillus species that are not dominant in a healthy vagina. L. crispatus is considered the gold-standard species for vaginal health. Newer clinical trials are testing “live biotherapeutic products,” essentially pharmaceutical-grade vaginal tablets containing multiple L. crispatus strains, applied after antibiotic therapy to establish colonization.¹⁹ These are still in early-phase trials, so they are not yet available as approved treatments, but they represent where the field is heading. In the meantime, if you want to try probiotics, look for products that specifically contain L. crispatus and are designed for vaginal use.

Lactic Acid Gels and Non-Antibiotic Approaches

Another non-antibiotic strategy is vaginal lactic acid gel, which works by lowering vaginal pH to mimic the acidic environment created by healthy Lactobacillus. Research on asymptomatic women suggests that lactic acid-containing gels could serve as maintenance therapy to reduce recurrences in women with episodic or recurrent BV.²⁰ The appeal is that these gels are available without prescription in many countries and carry minimal side effects. They are unlikely to be strong enough to treat active BV on their own, but as part of a prevention strategy between episodes, they may help maintain a hostile environment for BV-associated bacteria.

Prevention Habits That Actually Matter

Vaginal douching is one of the few modifiable risk factors with strong evidence behind it. A large longitudinal study of over 3,600 women found that regular douching raised the risk of BV by about 21% compared with never douching, even after accounting for the possibility that women douche in response to symptoms rather than causing them.²¹ The vagina is self-cleaning, and introducing water, vinegar, or commercial douching products disrupts the pH and washes away protective Lactobacillus.

Condom use during sex reduces the transfer of BV-associated bacteria between partners, which may lower recurrence risk, particularly in partnerships where the bacteria are being ping-ponged back and forth. Scented soaps, bubble baths, and fragranced feminine hygiene products applied to the vulva or vagina can also shift pH and irritate the mucosal lining. Cleaning with plain warm water externally is all that is needed.

Contraception and BV Risk

Your choice of contraception can influence BV risk in ways that are worth knowing. Copper IUD users have a higher incidence of BV compared with women using hormonal methods like the pill, ring, or patch. One longitudinal study found BV incidence of about 37% among IUD users versus roughly 19% among users of combined hormonal contraceptives over six months.²² A separate prospective study confirmed that copper IUD users had about a 28% higher BV incidence than women using no contraception or other non-hormonal methods.²³

Injectable progestins like depot medroxyprogesterone acetate (the Depo shot) were actually associated with lower BV risk in that same study. Oral contraceptive pills and hormonal implants did not show a significant effect in either direction.²⁴ If you have a copper IUD and are struggling with recurrent BV, this does not necessarily mean you need to switch, but it is a factor worth discussing with your provider alongside other contraceptive priorities.

When BV Is Not Actually BV

Persistent or recurrent vaginal symptoms that do not improve with BV treatment may actually be something else. Aerobic vaginitis is a distinct condition that can look similar to BV: the Lactobacillus population is depleted and discharge may be present. But aerobic vaginitis involves different bacteria (streptococci, staphylococci, or E. coli rather than anaerobes) and triggers a much stronger inflammatory immune response, producing elevated levels of inflammatory cytokines that BV typically does not.²⁵ BV is not usually an inflammatory condition, which is why it can be present without redness or pain. Aerobic vaginitis, by contrast, often involves visible inflammation, yellowish discharge, and soreness. The distinction matters because the treatments are different: clindamycin can work for aerobic vaginitis, but metronidazole, the go-to for BV, targets anaerobic bacteria and would miss the aerobic culprits entirely.

Yeast infections and trichomoniasis can also overlap with or mimic BV symptoms. If you have been treated for BV multiple times without lasting improvement, getting a proper diagnosis is the first step. Newer molecular diagnostic tests using quantitative PCR are more sensitive than the traditional Amsel clinical criteria, which missed a substantial proportion of BV cases in at least one study.²⁶ If your provider has only been using the “sniff test” and clinical impression, it may be worth asking about lab-based testing.

BV During Pregnancy

BV carries particular significance during pregnancy. A meta-analysis found that having BV roughly doubles the odds of preterm birth.²⁷ An earlier large study associated BV with preterm delivery of a low-birth-weight infant, with an adjusted odds ratio of about 1.4.²⁸ Whether treating BV in pregnancy consistently prevents these outcomes has been harder to prove, and guidelines vary on screening asymptomatic pregnant women. But if you are pregnant and symptomatic, treatment is clearly indicated, and oral metronidazole is considered safe during pregnancy.

BV during pregnancy also elevates the risk of other complications including post-cesarean wound infections and chorioamnionitis. Women with a history of preterm birth may be screened for BV early in pregnancy as part of a broader prevention strategy.

Experimental Treatments on the Horizon

The most intriguing new approach targets the biofilm problem directly. Researchers have developed an engineered endolysin called PM-477 derived from bacteriophages (viruses that infect bacteria) that naturally prey on Gardnerella. In laboratory and clinical samples, PM-477 killed Gardnerella bacteria and physically dissolved BV biofilms in 13 out of 15 patient samples, all without harming beneficial Lactobacillus or other normal vaginal bacteria.²⁹ Follow-up work showed that Gardnerella strains resistant to metronidazole remained fully susceptible to PM-477, and the endolysin was effective against preformed biofilms at very low concentrations.³⁰

PM-477 is still in preclinical development and not available as a treatment. But it represents a fundamentally different strategy: instead of carpet-bombing the vaginal microbiome with broad-spectrum antibiotics that kill Lactobacillus along with everything else, an endolysin like PM-477 could selectively remove the specific troublemaker while leaving the healthy flora intact. If this approach works in clinical trials, it could be the first BV treatment that simultaneously clears the infection and preserves the protective environment, addressing both sides of the recurrence equation at once.

BV and STI Risk

BV is not itself a sexually transmitted infection, but having it increases vulnerability to acquiring STIs. Women with the highest Nugent scores, indicating the most severe disruption of vaginal flora, had roughly 2.7 times the risk of acquiring a sexually transmitted infection compared with women who had normal flora.³¹ The mechanism is likely that the loss of Lactobacillus and the resulting higher pH leave the vaginal mucosa more permeable and less defended against pathogens. This is an additional reason to treat BV rather than simply tolerating it, even when symptoms are mild or intermittent.