Bacterial vaginosis: Initial treatment

INTRODUCTION — Bacterial vaginosis (BV) is the most common cause of abnormal vaginal discharge in reproductive-age females. Treatment is aimed at relieving symptoms, although many individuals are asymptomatic. Of those with symptoms, abnormal vaginal discharge and fishy odor are typical.

This topic will present the treatment options for various groups of patients with BV. Related topics on the presentation and diagnosis of BV and management of patients with recurrent BV, as well as the approach to the individual with vaginal discharge or cervicitis, are presented separately.

●(See "Bacterial vaginosis: Clinical manifestations and diagnosis".)

●(See "Bacterial vaginosis: Recurrent infection".)

●(See "Vaginitis in adults: Initial evaluation".)

●(See "Acute cervicitis".)

In this topic, when discussing study results, we will use the terms "women" or "patients" as they are used in the studies presented. We encourage the reader to consider the specific counseling and treatment needs of transgender and gender-expansive individuals.

RATIONALE FOR TREATMENT — Patients with confirmed BV who are symptomatic or who are undergoing a gynecologic procedure involving the vagina (regardless of symptoms) are treated [1,2]. Aside from those undergoing gynecologic surgery, we generally do not treat other asymptomatic individuals because BV is self-limiting in approximately one-third of nonpregnant persons and one-half of pregnant persons [3-6]. (See 'Individuals undergoing gynecologic procedures' below.)

Benefits of treatment include:

●Relief of symptoms (if present). (See "Bacterial vaginosis: Clinical manifestations and diagnosis", section on 'Clinical features'.)

●Reduction of postoperative infection – For individuals with asymptomatic infection prior to surgical abortion or hysterectomy, treatment can reduce the risk of postoperative infection. (See 'Individuals undergoing gynecologic procedures' below.)

●Reduction of sexually transmitted infections – Treatment of BV may reduce the risk of acquiring sexually transmitted infections (STIs), including HIV [7,8]. For this reason, some experts support the concept of treating all patients with BV regardless of presence or absence of symptoms. However, we agree with recommendations to not treat asymptomatic individuals unless they are undergoing gynecologic surgery [1,9]. Asymptomatic pregnant individuals with previous preterm births may benefit from treatment, but screening and treatment of these individuals has overall not been shown to be of benefit in averting adverse consequences of BV as it relates to pregnancy. (See 'Asymptomatic pregnant persons' below.)

DRUG DESCRIPTION AND SELECTION

Metronidazole — Metronidazole is given as an oral or vaginal medication over multiple days (table 1 and algorithm 1).

●Dosing – The choice of oral versus vaginal therapy depends upon patient preference because the efficacy is similar for multiday oral or vaginal treatments.

•Oral pills – We suggest oral metronidazole 500 mg twice daily for seven days [1]. Treatment with a single oral dose of 2 grams of metronidazole has lower efficacy [10] and thus is not recommended for treatment of BV.

•Vaginal gel – Vaginal therapy is given as metronidazole 0.75% gel; a 5-gram applicator is inserted into the vagina once daily for five days (5 grams of gel contains 37.5 mg of metronidazole) [1,11,12]. The multiday vaginal gel has similar efficacy to seven days of oral metronidazole [13-15].

•Single-dose treatment options – A newer single-dose metronidazole gel (1.3%) is available [16]. While the single-dose 1.3% treatment is superior to placebo gel [17], it is not known if the 1.3% single-day dose is as efficacious as the multiday oral or multiday 0.75% vaginal metronidazole treatments, and therefore we prefer the multiday treatments.

As above, single-dose oral metronidazole is not advised because of lower efficacy compared with multiday dosing [10].

●Adverse effects

•Common – Side effects of metronidazole (oral or vaginal) include a metallic taste, nausea (in 10 percent of patients), transient neutropenia (7.5 percent), prolongation of international normalized ratio in patients taking vitamin K antagonists (eg, warfarin), and peripheral neuropathy [18]. Gastrointestinal side effects are less common with vaginal administration [14]. Patients who are being treated with disulfiram should not use metronidazole as psychotic reactions have been reported [18].

•Alcohol consumption – For patients receiving oral nitroimidazoles, manufacturer packaging advises abstaining from alcohol consumption during treatment and for 24 hours after treatment [18]. However, data supporting this statement are sparse, and the magnitude of risk appears to be low [19]. We do not counsel patients taking oral or vaginal metronidazole to avoid alcohol during treatment [1].

•Drug allergy – Allergy to metronidazole is uncommon; it manifests as rash, urticaria, pruritus, and, rarely, anaphylaxis. The majority of patients can successfully undergo oral desensitization [20,21].

●Efficacy – The efficacy of metronidazole has been established in randomized trials using either placebo or active drug controls [22,23]. Published cure rates can vary widely according to the investigators' diagnostic criteria, definition of cure and treatment failure, and length of time post-therapy before the follow-up visit.

•Multiday therapy – Treatment of symptomatic nonpregnant patients with multiday metronidazole administered either orally (seven days) or intravaginally (five days) results in a high rate of clinical cure (70 to 80 percent at four weeks of follow-up) [22,24-27]. Most comparative studies using divided-dose oral regimens for one week achieved early cure rates in excess of 90 percent, and cure rates (by Amsel criteria) at four weeks of up to 80 percent [13,14,22,25,28-30].

•Longer-course treatment – There is little evidence of benefit from prolonging therapy longer than seven days. Although early cure rates are significantly higher when the initial course of metronidazole therapy is 14 rather than 7 days, long-term cure rates (21 days after completion of therapy) are similar for both treatment regimens [31].

Clindamycin — When selected for treatment of BV, clindamycin is typically given as a vaginal cream for seven nights (table 1 and algorithm 1) [1]. While oral clindamycin is an alternative treatment option, it is less preferred because of potential risk of Clostridioides (formerly Clostridium) difficile-associated diarrhea [32].

●Dosing

•Preferred – The preferred regimen is clindamycin cream 2% given vaginally as 5 grams of cream daily for seven days (5 grams of cream contains 100 mg of clindamycin phosphate) [1].

•Alternative regimens – Alternative regimens include [1,33]:

-Clindamycin 300 mg twice daily orally for seven days

or

-Clindamycin 100 mg vaginal suppositories at bedtime for three days

•Less-preferred regimens – Clindamycin 2% is also available as a single-dose bioadhesive cream (commercial name Clindesse) or gel (commercial name Xaciato). Both are given as a single 5 g intravaginal dose of product containing 100 mg of clindamycin phosphate [34,35]. The products can be inserted at any time of day.

-Cream – While the 21-day BV cure rates of single-dose and multidose clindamycin cream were similar in one study (64.3 versus 63.2 percent), methodologic limitations prevented definitive conclusions regarding efficacy and therefore we prefer multiday regimens [1,36].

-Gel – Clindamycin 2% vaginal gel (commercial name Xaciato) is an effective single-day treatment for BV; data comparing the efficacy of single-day gel with multiday regimens are not yet available. In a trial comparing single-dose clindamycin 2% gel with placebo in 307 individuals with BV, 21-day cure rates were 86 versus 21 percent for clinical cure, 53 versus 3 percent for biological cure, and 45 versus 3 percent for therapeutic cure [35,37]. Treatment-emergent adverse events were more common in patients receiving clindamycin gel than placebo (15.3 versus 9.7 percent, respectively); 9.4 percent of the adverse events in the treatment group were related to vulvovaginal candidiasis.

●Adverse effects

•Common – The most common adverse effects are vulvovaginal candidiasis (topical and oral formulations) and gastrointestinal side effects (oral formulation) [38,39]. Although uncommon, pseudomembranous colitis has been reported with both formulations.

•Clindamycin resistance – Intravaginal clindamycin therapy has been associated with an increased prevalence of clindamycin-resistant anaerobic bacteria in the vagina post-treatment (17 percent of bacterial isolates before versus 53 percent of isolates after therapy) [40]. This effect persisted in most women for at least 90 days after clindamycin treatment. However, given the generally similar cure rates between clindamycin and metronidazole regimens, the clinical significance of this finding is unclear.

•Impact on latex condoms or diaphragms – Latex condoms or diaphragms should not be used concurrently, or within 72 hours of treatment, with clindamycin cream. Clindamycin cream may weaken latex.

●Efficacy – The efficacy of oral and vaginal clindamycin preparations have been demonstrated in a meta-analysis of randomized trials, both comparative and placebo controlled [22].

Tinidazole — Tinidazole is a second-generation nitroimidazole that is considered an alternative regimen if metronidazole and clindamycin are unavailable or not tolerated [1]. It has a longer half-life than metronidazole (12 to 14 hours versus 7 to 8 hours) and fewer side effects reported in some, but not all, studies [41,42].

●Dose – If used, we suggest 1 gram orally once daily for five days, as efficacy is slightly higher and side effects are slightly less frequent than with shorter-course therapy (tinidazole 2 grams orally daily for two days) [43].

●Adverse effects – In initial clinical trials with variable designs and treatment indications, 11 percent of patients reported adverse effects with a single 2-gram dose of tinidazole [44]. The most common were a metallic/bitter taste (3.7 percent), nausea (3.2 percent), and weakness or fatigue (2.1 percent).

●Efficacy – Randomized trials have shown that it is at least as effective as metronidazole, but not superior [43,45-47], and a single-dose regimen appears to be as effective as vaginal clindamycin cream [48].

Secnidazole — Secnidazole is a nitroimidazole antibiotic with a longer half-life than metronidazole (approximately 17 to 19 hours versus 7 to 8 hours) that is considered an alternative regimen for BV [1,49].

●Dose – Secnidazole is prescribed as a single 2-gram packet of granules that are taken once orally, typically mixed into a soft food (ie, applesauce, yogurt, or pudding) [50]. The mixture should be consumed within 30 minutes, and the granules should not be crunched or chewed.

●Adverse effects – Compared with placebo, treatment with secnidazole is more likely to result in vulvovaginal candidiasis (nearly 9.6 percent), headache (3.6 percent), nausea (3.6 percent), diarrhea (2.5 percent), and abdominal pain (2 percent) [50].

●Efficacy – In a phase 3 trial comparing a single 2-gram dose of oral secnidazole with placebo, single-dose secnidazole was superior to placebo, with clinical outcome responder rates of 53 versus 19 percent [49]. A single 1-gram oral dose of secnidazole appears to be effective as well [51,52]. Although compliance is enhanced by the convenience of a single-dose regimen, there is no evidence demonstrating that single-dose secnidazole is superior to multidose metronidazole therapy, which typically is less expensive. While previously available in other countries, secnidazole was approved for the treatment of BV by the US Food and Drug Administration (FDA) in 2017 [50].

Dequalinium chloride — Dequalinium chloride is a quaternary ammonium compound that functions as a local antiseptic and is available in Canada and Europe (not available in the US) [53,54]. One dequalinium chloride 10 mg tablet is inserted vaginally at bedtime for six consecutive nights [53]. A clinical trial comparing dequalinium treatment (n = 163) with a standard clindamycin cream (n = 152) reported early clinical cure rates (ie, 3 to 14 days after end of therapy) of 81.5 and 78.4 percent, respectively [55]. The role in primary treatment of BV in diverse populations is unclear [56].

Product use is limited by the following:

●Patient age – Patients should be aged 18 to 55 years of age as data on patients <18 years and >55 years are not yet available [53].

●Menses – The product should not be used during menses because local drug concentrations cannot be ensured [53].

●Sexual intercourse – As data on the impact of vaginal intercourse on drug efficacy are insufficient, patients are advised to refrain from vaginal sexual activity during treatment [53]. Additionally, data are insufficient to assess the drug's impact on condoms, diaphragms, menstrual cups, and related devices. Thus, use of these products is not advised during treatment with dequalinium chloride.

●Pregnancy, lactation, and birth – Data on drug safety in pregnancy are limited, although reassuring to date, and drug excretion in breast milk is not known [53]. Additionally, patients are advised to avoid use of dequalinium within 48 hours of intercourse if attempting to conceive and to avoid the drug within 12 hours of vaginal delivery.

●Limited data from diverse populations – The vast majority of participants in the above trial were White [55]. Larger clinical trials that include diverse patient populations are needed to fully understand the efficacy and optimal use of this medication [56].

PROBIOTICS AND OTHER THERAPIES

●Probiotics – Probiotics (live microorganisms that confer a health benefit on the host when administered in adequate amounts) have been used alone and as adjunctive therapy to antibiotics for treatment of BV and prevention of relapse. Systemic reviews of clinical trials of probiotics for treatment of BV have not found sufficient evidence for or against efficacy [57,58]. Although some trials have reported very promising results, particularly for treatment with a human-derived L. crispatus strain, larger trials that demonstrate treatment efficacy are needed before use of this therapy is considered [59]. In addition, further investigation is needed to determine the optimum route of administration (oral or vaginal), which strains or combination of strains are most effective (eg, Lactobacillus rhamnosus GR-1, Lactobacillus reuteri RC-14, Lactobacillus acidophilus), the timing of administration relative to antibiotic therapy, and the dose and duration of use [60].

●Less effective – Cure rates with ampicillin and amoxicillin are mediocre. Boric acid has not been shown to be effective in treating isolated or initial cases of BV. While boric acid may play a supportive role in treatment with primary antibiotic therapy, it is not to be used for primary treatment. There is a role for boric acid vaginal suppositories in patients with relapsing or recurrent BV when used in combination with antimicrobial agents. A novel form of boric acid is under investigation [61]. (See 'Persistent or repeat symptoms' below.)

●Ineffective – Triple-sulfa creams, erythromycin, tetracycline, ampicillin, amoxicillin, lactic acid gel, acetic acid gel, ascorbic acid, azithromycin, chlorhexidine, hydrogen peroxide, vaginal boric acid as an isolated therapy, and povidone-iodine vaginal douches are significantly less effective for the treatment of BV than metronidazole and clindamycin and should not be used [31,62-68].

For patients with recurrent BV, long-term suppressive oral or topical clindamycin regimens are not advised because of toxicity (oral) and lack of documented efficacy (topical) [69]. In our experience, long-term clindamycin therapy is frequently associated with vaginal yeast coinfections compared with metronidazole gel. Clindamycin should not be used in patients with a history of C. difficile infection.

PATIENTS WHO REQUIRE TREATMENT — We treat all patients with confirmed BV who are symptomatic or who are undergoing a gynecologic procedure involving the vagina (regardless of symptoms) [1,2].

Symptomatic — Patients with confirmed BV infection and symptoms are treated because symptoms are significantly bothersome and treatment may reduce the risk of acquiring sexually transmitted infections. (See 'Rationale for treatment' above.)

Nonpregnant — Symptomatic nonpregnant patients are treated as below. Treatment is extrapolated to patients with a surgical neovagina as data supporting need for an alternate approach are lacking.

Preferred treatments — Both metronidazole and clindamycin, as oral or vaginal formulations, are preferred therapies for BV and have similar treatment efficacy (algorithm 1) [1,2]. The choice of medication is based on availability, patient preferences, side effects, cost, and history of response or adverse reactions [27]. Treatment does not vary with HIV infection. This treatment approach is extrapolated to individuals who have undergone vaginoplasty. (See "Gender-affirming surgery: Male to female", section on 'Genital (bottom) surgery (vaginoplasty)'.)

The following treatment recommendations are consistent with those of the World Health Organization (WHO) and the United States Centers for Disease Control and Prevention (CDC) (algorithm 1) [1,2]. Guidelines from other organizations are available elsewhere [70,71]. Oral metronidazole is commonly used because oral treatment is often preferred by patients rather than intravaginal therapy and metronidazole is less commonly associated with Clostridioides difficile infection compared with oral clindamycin [32].

●Metronidazole 500 mg orally twice daily for seven days [1,2]

or

●Metronidazole gel 0.75% one full applicator (5 g) intravaginally, once daily for five days [1]

or

●Clindamycin cream 2% one full applicator (5 g) intravaginally at bedtime for seven days [1]. We prefer vaginal clindamycin over metronidazole gel if the patient has only been treated with metronidazole to date [56].

Alternative therapy — If the preferred therapies are not available or if the patient prefers a shorter duration of treatment, clindamycin intravaginal ovules, oral tinidazole, and oral secnidazole are reasonable treatment alternatives. Efficacy appears similar to metronidazole and clindamycin although supporting data are limited. For tinidazole and secnidazole, the benefits of a shorter treatment duration may be offset by higher cost and lack of longer-term outcomes data, which is the rationale for making these drugs alternates [1].

●Clindamycin 300 mg orally twice daily for seven days [1,2]

or

●Clindamycin ovules 100 mg intravaginally at bedtime for three days [1]. Use of this product may weaken latex or rubber contraceptives such as condoms or diaphragms. Patients should wait 72 hours after use of clindamycin ovules before using latex or rubber condoms or diaphragms.

or

●Secnidazole 2 g oral granules in a single dose (sprinkled onto unsweetened applesauce, yogurt, or pudding before ingestion) [1].

or

●Tinidazole 2 g orally once daily for two days [1]

or

●Tinidazole 1 g orally once daily for five days [1]

Pregnant or lactating persons

●Pregnancy – Pregnant persons with symptomatic BV are treated to relieve bothersome symptoms. In a meta-analysis of 10 trials comparing antibiotic therapy with placebo for individuals with confirmed BV or intermediate vaginal microbiota while pregnant, antibiotic therapy effectively reduced BV during pregnancy (average risk ratio 0.42, 95% CI 0.31-0.56, 10 trials, 4403 women) [72]. While oral and vaginal regimens are available, oral metronidazole is commonly used as some data indicate greater efficacy of oral metronidazole against upper tract infection compared with other options [73-75]. However, not all guidelines agree [1,2].

Routine screening of asymptomatic pregnant persons for BV has not been shown to reduce the risk of preterm birth [76,77]. (See "Spontaneous preterm birth: Overview of interventions for risk reduction", section on 'Routinely screening for cervicovaginal and sexually transmitted infections'.)

•Oral treatment regimens (preferred) – Oral treatment is effective and has not been associated with adverse fetal or obstetric effects [72,78-82]. Therapeutic options include:

-Metronidazole 500 mg orally twice daily for seven days [1]

or

-Metronidazole 250 mg orally three times daily for seven days [1,2,74,75]

or

-Clindamycin 300 mg orally twice daily for seven days [1,2,83]

•Alternative topical therapy – The author and other experts prefer oral therapy in pregnant patients because some data indicate oral treatment is more effective against potential upper genital tract infection [73-75]. However, as topical therapy is not inferior to oral therapy in effecting cure or preventing adverse pregnancy outcomes such as preterm birth, the World Health Organization and US Centers for Disease Control and Prevention (CDC) recommend either oral or topical therapy for treatment of symptomatic pregnant individuals [1,2].

The regimens are the same as for nonpregnant patients and include [1]:

-Metronidazole 0.75% gel one full applicator (5 g) intravaginally, once daily for five days (5 grams of gel contains 37.5 mg of metronidazole) [1,2,84]

or

-Clindamycin cream 2% one full applicator (5 g) intravaginally at bedtime for seven nights [1,85]

•Drug safety in pregnancy

-Metronidazole – Some clinicians avoid use of metronidazole in the first trimester because it crosses the placenta, and thus has a potential for teratogenicity. However, meta-analyses of observational data in pregnant persons have not found any relationship between metronidazole exposure during the first trimester of pregnancy and congenital anomalies [86-88]. The CDC no longer discourages the use of metronidazole in the first trimester [1]; the World Health Organization continues to advise caution with first trimester use [2]. An additional concern is that the drug is mutagenic in bacteria and carcinogenic in mice, but there is no evidence of harm in humans.

-Drugs to avoid – Limited data are available to inform use of tinidazole or secnidazole in pregnant persons. However, as animal data suggests tinidazole is associated with risk, use of this drug for pregnant persons is not advised [1,89]. As the available data on secnidazole is inadequate to determine safety during pregnancy, we and others also avoid using it for pregnant patients [1].

●Lactating individuals – Lactating individuals with confirmed BV and symptoms are offered treatment while asymptomatic individuals are not. While the treatment options as the same as during pregnancy, some preference is given to metronidazole as clindamycin has been associated with infant gastrointestinal symptoms.

Additional discussions of drug safety during lactation are available online.

•Metronidazole (preferred) – Our preferred treatment for lactating individuals is metronidazole. Although vaginal metronidazole has not been extensively studied during breastfeeding, plasma levels are less than 2 percent of those after a 500 mg oral dose, so vaginal metronidazole is unlikely to be a safety concern in lactating individuals [90].

-Metronidazole 500 mg orally twice daily for seven days [1,2]

or

-Metronidazole gel 0.75% one full applicator (5 g) intravaginally, once daily for five days [1]

Compared with metronidazole treatment of infection in neonates, breastfed infants receive lower doses of metronidazole through breastmilk secretion [1]. Available human data do not support an association between metronidazole use and cancer although an association with carcinogenesis in rodents has been demonstrated [90]. Maternal metronidazole use does not appear to be associated with increased adverse events compared with use of other antimicrobials, although a cohort study found a nonsignificant trend toward more loose stools and more candidal colonization in metronidazole-exposed infants. Since the relative infant dose of metronidazole is high (29 percent) with use of the maternal 2-gram single dose, a cautious approach for mothers receiving this dose is to express and discard their milk for 12 to 24 hours. This recommendation has not been extended to other metronidazole regimens [1].

•Clindamycin – While oral clindamycin is a reasonable therapeutic choice, we prefer oral metronidazole because clindamycin has the potential to cause adverse effects on the breastfed infant's gastrointestinal microbiota. Infants of patients treated with oral clindamycin should be monitored for diarrhea, candidiasis (thrush, diaper rash) or, rarely, blood in the stool indicating possible antibiotic-associated colitis [90]. Infant side effects are less likely with vaginal clindamycin than oral use since only approximately 30 percent of a vaginal dose is absorbed.

•Alternative drugs – For lactating individuals who are administered tinidazole or secnidazole, interruption of breastfeeding is recommended during treatment and for three days after the last dose based on animal data (tinidazole only) [50].

Individuals undergoing gynecologic procedures — For individuals with confirmed BV (with or without symptoms), we suggest antibiotic treatment prior to transvaginal procedures or surgery (eg, pregnancy termination, dilation and curettage, hysterectomy, etc) to reduce the risk of postprocedure infection (cuff infection after hysterectomy, endometritis after termination) [91-93]. Reported reductions in postoperative infectious complications range from 10 to 75 percent [94-98]. There is insufficient evidence to make a conclusion regarding screening for BV prior to intrauterine device insertion [99]. The medication treatment options are the same as for symptomatic nonpregnant patients.

TREATMENT NOT INDICATED — Unless a patient with documented BV infection is undergoing gynecologic surgery involving the vagina, we do not routinely treat individuals with asymptomatic BV. BV is self-limiting in approximately one-third of nonpregnant persons and one-half of pregnant persons [3-6]. In addition, sex partners (female or male) are not treated unless symptoms develop.

Asymptomatic nonpregnant — We suggest observation, and not antibiotic treatment, for asymptomatic nonpregnant individuals [1,9]. Treatment of asymptomatic BV is typically avoided since patients often spontaneously improve over a period of several months and any antibacterial therapy can be followed by symptomatic vaginal yeast infection. Although some experts recommend treatment of asymptomatic BV because affected individuals are more susceptible to acquiring other sexually transmitted infections (STIs, including HIV and herpes simplex virus), available evidence is insufficient to clearly support or exclude a benefit of treatment [93,100-103].

●Evidence – This practice is supported by a double-blind, placebo-controlled trial of 54 women with asymptomatic BV who were randomly assigned to receive intravaginal metronidazole or placebo gel [5]. There was no difference in the proportion of women in either group who noticed improvement in vaginal odor or discharge. In addition, 6 of 28 women receiving metronidazole developed symptomatic vaginal candidiasis compared with no women taking placebo.

●Exception – Asymptomatic individuals with confirmed BV who are planning gynecologic procedures or surgery that involves the vagina are treated. Preoperative treatment of asymptomatic BV has been associated with reduced risk of postoperative infection. (See 'Individuals undergoing gynecologic procedures' above.)

Asymptomatic pregnant persons — Asymptomatic pregnant individuals with previous preterm births may benefit from treatment, but screening and treatment of these individuals are controversial as the available data do not show a consistent benefit to this approach [1].

●Our approach – Based upon the data below, we and others, including the American College of Obstetricians and Gynecologists, United States Preventive Services Task Force, CDC, and Society of Obstetricians and Gynaecologists of Canada, suggest not routinely screening and treating all pregnant individuals with asymptomatic BV to prevent preterm birth and its consequences [1,104-107]. It is not possible to define specific features characterizing a subgroup of individuals who might respond favorably to a screening and treatment protocol. Defining these features is an active area of investigation.

However, as illustrated below, there may be benefits to early screening and treatment of asymptomatic pregnant individuals who have a history of a previous preterm delivery, but there are insufficient data to recommend this as a routine practice [79,108-111]. Further definition of high-risk subgroups is under investigation. As an example, individuals with polymorphisms of genes regulating cytokine production (eg, tumor necrosis factor variants) have a greater proinflammatory immune response to infectious stimuli, such as BV [112]. Enhanced induction of cytokines in these individuals could then lead to preterm labor or rupture of membranes. Other aspects of host response (eg, low levels of immunoglobulin A [IgA] to Gardnerella vaginalis) or the specific types of BV-associated bacteria involved (eg, bacteria that produce high levels of sialidase or protease) may also play a role in placing some individuals with BV at high risk of preterm birth. Additional information is available in related content:

•(See "Spontaneous preterm birth: Overview of risk factors and prognosis", section on 'Genetic variants'.)

●Risk for adverse pregnancy outcomes, including preterm birth – As many as one-third of pregnant individuals in the United States have BV [113]. Of concern, a 2007 meta-analysis reported a statistically significant increased risk of preterm birth in these women; the pooled odds ratio (OR) for prematurity was 2.16 (95% CI 1.56-3.00) [114]. The increased risk of preterm birth attributable to BV appears to be linked to preterm labor due to chorioamnionitis [115,116]. Other complications of BV include postpartum endometritis (OR 2.53, 95% CI 1.25-5.08) and an increased risk of late miscarriage (OR 6.32, 95% CI 3.65-10.90) [114].

●Conflicting data for screening and treatment – Despite the association between BV and adverse outcome, screening and treatment of asymptomatic BV during pregnancy are controversial and vary in part based on the individual's baseline risk for preterm delivery [107]. Meta-analyses of randomized trials performed in general obstetric populations have generally found that treatment of asymptomatic infection does not reduce the incidence of preterm labor or delivery in the overall obstetric population [72,85,104], but some subgroups of women, such as those at high risk for preterm birth, may benefit. The available evidence is discordant due to differences in selection of trials and differences between the included trials. Examples include:

•In a 2015 Cochrane meta-analysis of eight trials comparing the use of prophylactic antibiotics in the second or third trimester of pregnancy in women at risk for preterm delivery, antibiotic treatment reduced preterm delivery in the subgroup of women with a prior preterm delivery and current BV compared with no treatment (risk ratio 0.64, 95% CI 0.47-0.88, one trial, 258 women) [117]. Antibiotic prophylaxis did not reduce preterm delivery in women with a prior history of preterm birth who did not have BV.

•In a 2013 Cochrane meta-analysis including 21 trials involving 7847 pregnant women with BV (symptomatic or asymptomatic) detected through screening, antibiotic therapy was highly effective in eradicating infection but did not significantly reduce the odds of preterm birth at less than 37 weeks (OR 0.88, 95% CI 0.71-1.09) or the risk of preterm premature rupture of membranes (OR 0.74, 95% CI 0.30-1.84) [72]. Treatment initiated before 20 weeks of gestation also did not reduce the risk of preterm birth before 37 weeks (OR 0.85, 95% CI 0.62-1.17).

When the Cochrane reviewers separately analyzed the subgroup of women with a history of one or more prior preterm births (ie, women at high risk for preterm birth), the detection and treatment of BV still did not significantly reduce the risk of preterm birth (OR 0.78, 95% CI 0.42-1.48, three trials, 421 women).

•In a 2011 meta-analysis of five randomized trials of asymptomatic women with BV at <22 weeks of gestation treated with clindamycin or placebo/no treatment, clindamycin therapy was associated with a reduction in preterm birth <37 weeks (3.7 percent [44 out of 1183] versus 6.2 percent [72 out of 1163]; fixed effects relative risk [RR] 0.60, 95% CI 0.42-0.86, random effects 0.64, 95% CI 0.39-1.05) and late miscarriage (0.3 percent [2 out of 639] versus 1.9 percent [12 out of 631]; RR 0.20, 95% CI 0.05-0.76) [85]. Subgroup analysis revealed that oral, not vaginal, clindamycin therapy was associated with a significant reduction in preterm birth (oral therapy RR 0.39, 95% CI 0.20-0.76; vaginal RR 0.73, 95% CI 0.47-1.14). The analysis included a mixed population of women at both low and high risk of preterm birth.

●Therapeutic options if treatment is elected – When the decision is made to treat asymptomatic pregnant individuals who have confirmed BV because of prior history, the optimal choice of antibiotic, timing of therapy, duration of use, and harms of therapy are debated. Based on the data presented below, we use oral therapy when treatment is indicated and consider both metronidazole and clindamycin acceptable choices.

•Metronidazole – The first trials demonstrated a reduction in preterm birth in high-risk women treated with oral metronidazole or oral metronidazole and erythromycin (table 2) [73-75]. However, in two studies, metronidazole use in pregnancy appeared to increase the risk of preterm birth [118,119]. This association requires further investigation and confirmation before avoiding metronidazole for treatment of BV in pregnancy.

•Clindamycin – Trials suggest that oral clindamycin given early in pregnancy is an effective therapy [85]. Topical clindamycin given in the second half of pregnancy is less effective and even associated with an increase in low birth weight and neonatal infection [120].

Asymptomatic sex partners — Although sexual activity is a risk factor for transmission, the data do not support treatment of asymptomatic sex partners of individuals with confirmed BV. (See 'Nonpregnant' above.)

Males — There is no strong evidence that treatment of sex partners who have a penis impacts the treatment success or risk of recurrence for the patient with confirmed BV [1,121-123]. Until population-specific data are available, this thinking is extrapolated to patients who have undergone surgical phalloplasty as well. (See "Gender-affirming surgery: Female to male", section on 'Phalloplasty with genital reconstruction'.)

Relevant studies include:

●In a 2016 meta-analysis of seven trials assessing efficacy of BV treatment versus placebo in women and their male sex partners, antibiotic treatment of the male sex partners did not increase the rate of clinical or symptomatic improvement for the index patients nor did it reduce recurrence rates during a four-week study period [121]. The men in the treatment group reported more adverse events compared with men in the placebo group but most events were minor.

●In a trial of 214 individuals with recurrent BV, seven-day metronidazole treatment of male partners, compared with placebo treatment, did not reduce BV recurrence in intention-to-treat analysis [123]. Although not statistically significant, BV recurrence was less likely in individuals whose male partners were adherent to the study medication.

Females — Individuals with a vagina who are sex partners of those with confirmed BV should be educated about the signs and symptoms of BV given the high risk of concordant infection (25 to 50 percent) [124-126]. If symptoms subsequently develop in a sex partner and BV is confirmed, treatment is indicated for relief of symptoms. It has been hypothesized that behavioral interventions that reduce transfer of vaginal fluid between sex partners may be helpful (eg, cleaning sex toys between use, use of gloves during digital-vaginal sex); however, a small randomized trial evaluating this approach reported no reduction in BV persistence [127]. Further study is needed. (See "Sexual and gender minority women (lesbian, gay, bisexual, transgender, plus): Medical and reproductive care", section on 'Sexually transmitted infection'.)

This treatment approach is extrapolated to individuals who have undergone surgical vaginoplasty. (See "Gender-affirming surgery: Male to female", section on 'Genital (bottom) surgery (vaginoplasty)'.)

TREATMENT RESPONSE AND PATIENT FOLLOW-UP — Treatment response is anticipated in 70 to 80 percent of patients; response rates vary by choice of drug, duration of treatment, and patient population (eg, initial or recurrent BV). Efficacy of treatment by drug is discussed above. (See 'Drug description and selection' above.)

●Symptoms resolve – Routine follow-up visits are not required if symptoms resolve [1].

●Symptoms persist or return – Patients whose symptoms persist despite adequate treatment or whose symptoms return quickly (ie, within a few weeks) are re-evaluated for BV and other causes of abnormal vaginal discharge are excluded. If BV is confirmed as the likely cause of symptoms, the patient is treated for repeat infection. Identification of prognostic markers that reliably predict outcome, and thus provide opportunities to modify therapy, are an area of active research [128,129]. (See 'Persistent or repeat symptoms' below.)

PERSISTENT OR REPEAT SYMPTOMS — Patients may have symptoms that persist despite adequate treatment or may experience rapid (ie, within a few weeks) return of symptoms. We educate patients that periodic recurrences of BV are common but do not necessarily represent recurrent BV. Recurrent BV is defined as three or more confirmed episodes within one year. (See "Bacterial vaginosis: Recurrent infection".)

Definition and prevalence

●BV relapse versus reinfection – Potential explanations for confirmed repeat BV infection include relapse or reinfection. Relapse, the more likely mechanism, can reflect inadequate initial treatment (eg, incorrect use of medication or early cessation of therapy), failure to eradicate the offending organisms, or inability reestablish the normal protective vaginal microbiota dominated by lactobacillus. Infections involving biofilms can be more difficult to eradicate [130]. These mechanisms may explain the persistence of pathogenic bacteria, as documented by 16S rRNA sequencing and polymerase chain reaction (PCR) testing, in patients with documented infection who are treated with seven days of oral metronidazole therapy [131].

The only interventions proven to reduce development or recurrence of BV are chronic suppressive therapy and circumcision of male partners [132,133]. (See "Bacterial vaginosis: Recurrent infection".)

●Prevalence – Approximately 30 percent of patients with initial responses to therapy have a recurrence of symptoms within three months [28], and more than 50 percent experience a recurrence within 12 months [134]. The explanation for this high rate of repeat infection is unclear.

●Recurrent BV – Patients with three or more documented BV infections in one year are diagnosed with recurrent BV. Management of patients with recurrent BV is discussed below. (See "Bacterial vaginosis: Recurrent infection".)

Confirm BV — Patients who develop recurrent BV symptoms, including abnormal vaginal discharge and odor, are re-evaluated to confirm BV and exclude other etiologies of vaginal discharge. (See "Vaginitis in adults: Initial evaluation".)

Retreatment — Limited data are available regarding the optimal treatment strategy for patients with rapid return of BV [1].

●Induction therapy – We retreat patients with symptomatic confirmed repeat BV with a seven-day course of oral or vaginal metronidazole or clindamycin.

•If one drug class or route has not been previously prescribed, we start with that medication.

•If the patient has tried multiple drugs in the past, then we repeat whichever drug seemed most helpful or was better tolerated.

●Suppressive therapy – Patients with a propensity for repeated BV infections, but who do not meet the criteria for recurrent BV, may benefit from extended or adjunct therapy after initial antibiotic treatment. We take this approach for patients who experience repeat infections less than six months from initial treatment. We start with extended treatment with metronidazole gel because it is readily available. For patients in whom extended metronidazole gel has not been helpful or effective, we then add adjunct vaginal boric acid.

Management of patients with recurrent BV, defined as three or more documented infections in one year, is discussed separately. (See "Bacterial vaginosis: Recurrent infection".)

•Metronidazole gel suppressive therapy – After completion of induction antibiotic therapy, we treat with metronidazole gel 0.75% one full applicator (5 g) intravaginally twice-weekly for four to six months [1]. We re-evaluate patients yearly and use a symptom-driven approach to determine further treatment [56]. We counsel patients to avoid intravaginal metronidazole or intercourse starting one day prior to evaluation. We do not repeat molecular tests (ie, nucleic acid amplification tests [NAATs]) as they have not been validated for test of cure.

-Asymptomatic – Asymptomatic patients are continued on metronidazole gel twice weekly. The rationale is that we prefer to continue successful suppressive therapy rather than wait for repeat infection and need for more intensive treatment [56].

-Symptomatic – Symptomatic patients are retreated for repeat infection. Initial antibiotic therapy consists of the drug that has not yet been used or that resulted in the best past response. For patients who developed BV symptoms while on suppressive therapy with metronidazole gel, we typically add vaginal boric acid in combination with antibiotic treatment.

•Combination therapy with adjunct vaginal boric acid – Vaginal boric acid suppositories can be used simultaneously with induction antibiotic treatment [1,135]. For this approach, vaginal boric acid 600 mg once daily at bedtime is begun simultaneously with the antibiotic and continued for a total of 20 to 30 days [1]. Patients are seen for follow-up a day or two after their last vaginal boric acid dose; if they are in remission, we immediately begin metronidazole gel twice weekly for four to six months as suppressive therapy. Therapy is then discontinued once treatment has been completed.

Boric acid can cause death if consumed orally; patients should be told to store boric acid in a secure place that is inaccessible to children. Boric acid should not be used by individuals who are pregnant or attempting conception. Sex partners of patients treated with vaginal boric acid have reported skin irritation after exposure [136]. The magnitude and duration of risk from the time of boric acid use is not known.

●Patients with metronidazole allergy –Although supporting data are lacking, those with a metronidazole allergy can either undergo metronidazole desensitization (our preference) or be treated with topical clindamycin gel. In our experience, clindamycin therapy is frequently associated with vaginal yeast coinfections compared with metronidazole gel. Clindamycin should not be used in patients with a history of C. difficile infection.

Clinical expectations — We educate patients that repeat infection is common and is not necessarily diagnostic of recurrent BV. Studies have reported repeat BV episodes in up to 30 percent of patients at three months from treatment and in nearly 60 percent by 12 months [134,137,138]. Patients who responded well to initial treatment are anticipated to respond well again. As recurrent BV requires more intensive treatment and the diagnosis can be stigmatizing, it is important to limit this diagnosis to individuals with three or more confirmed episodes in one year. (See "Bacterial vaginosis: Recurrent infection".)

Ineffective therapies

●Vaginal acidifying agents – Vaginal acidifying agents, although popular and widely used, have no role in the treatment of acute or chronic BV, as they have never been shown to enhance cure rates. A newly available vaginal gel containing lactic acid, citric acid, and potassium bitartrate (commercial name Phexxi) has been shown to be an effective on-demand contraception; however, its role in maintaining normal vaginal pH as management for BV has not been studied. (See "Pericoital (on demand) contraception: Diaphragm, cervical cap, spermicides, and sponge", section on 'Vaginal spermicide and pH regulator gel'.)

Douching should be avoided.

●Probiotics – In a randomized trial, probiotics were not more effective than placebo for prevention of relapse [69].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Bacterial vaginosis".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Bacterial vaginosis (The Basics)")

●Beyond the Basics topics (see "Patient education: Bacterial vaginosis (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Who to treat and rationale – We treat all symptomatic persons with confirmed bacterial vaginosis (BV) to reduce symptoms (typically vaginal discharge and odor). We also treat patients with confirmed BV who are undergoing gynecologic procedures that involve the vagina, regardless of symptoms, to reduce the risk of postoperative complications. We typically do not treat other asymptomatic individuals because BV infection is self-limited and any antibacterial therapy can be followed by symptomatic vaginal yeast infection. (See 'Rationale for treatment' above.)

●Symptomatic patients

•Nonpregnant – For nonpregnant persons with confirmed symptomatic BV, we suggest multiday treatment with metronidazole or clindamycin rather than treatment with other antibiotics (algorithm 1) (Grade 2C). As efficacy is similar for oral or vaginal regimens of both metronidazole or clindamycin, choice of treatment route is driven by patient preference around treatment route, drug availability, and cost. Treatment of patients with a neovagina and/or HIV infection is the same. (See 'Preferred treatments' above.)

-Preferred treatment regimens include:

Metronidazole 500 mg twice daily orally for seven days.

or

Metronidazole gel 0.75% (5 grams containing 37.5 mg metronidazole) once daily vaginally for five days.

or

Clindamycin 2% vaginal cream once daily at bedtime for seven days. During therapy with clindamycin cream, latex condoms should not be used.

-Alternate regimens – Oral tinidazole and secnidazole have demonstrated efficacy similar to oral metronidazole and convenient dosing. However, as they also tend to be of higher cost and lesser availability, they are considered alternatives to the treatments above. (See 'Alternative therapy' above.)

•Pregnant or lactating – For pregnant or lactating patients with symptomatic BV, we suggest oral metronidazole or clindamycin rather than intravaginal formulations (algorithm 1) (Grade 2C). The author and other experts prefer oral therapy in pregnant persons because some data indicate oral treatment is more effective against potential upper genital tract infection. Treatment of lactating individuals is the same as for pregnant persons with some preference is given to metronidazole as clindamycin has been associated with gastrointestinal symptoms in infants. (See 'Pregnant or lactating persons' above.)

-Metronidazole 500 mg twice daily orally for seven days (preferred in lactating individuals)

or

-Clindamycin 300 mg orally twice daily for seven days

•Planned gynecologic surgery – For individuals who are undergoing gynecologic procedures involving the vagina, we suggest testing and treatment of confirmed BV rather than observation, even in the absence of symptoms (Grade 2C). Preoperative treatment decreases the frequency of postoperative infectious complications. (See 'Individuals undergoing gynecologic procedures' above.)

●Asymptomatic individuals – For asymptomatic individuals, we suggest observation rather than antibiotic treatment (Grade 2C), unless they are undergoing gynecologic surgery. Treatment is typically avoided since patients often spontaneously improve over a period of several months and any antibacterial therapy is often followed by symptomatic vaginal yeast infection. One exception is patients undergoing gynecologic surgery who are treated regardless of symptoms. (See 'Asymptomatic nonpregnant' above.)

Specific sub-populations for consideration include:

•Asymptomatic pregnant persons – For asymptomatic pregnant individuals, we suggest not screening or treating for BV (Grade 2C). There is no consistent evidence that screening and treatment of asymptomatic infection during pregnancy reduces the risk of preterm birth. (See 'Pregnant or lactating persons' above.)

•Asymptomatic sex partners – For sex partners of individuals with confirmed BV, we recommend not treating empirically with antibiotics (Grade 1B). Antibiotic treatment of sex partners has not been associated with increased rates of clinical or symptomatic improvement for the index patient nor reduction of recurrence rates and antibiotic therapy has been associated with side effects. Sex partners of a patient with confirmed BV who themselves have a vagina or surgical neovagina should be educated about the signs and symptoms of BV given the high risk of concordant infection (25 to 50 percent). (See 'Asymptomatic sex partners' above.)

●Symptom recurrence – Approximately 30 percent of patients with an initial response to therapy have a recurrence of symptoms within three months and nearly 60 percent experience a recurrence within 12 months. We treat symptomatic recurrence using a different antibiotic than that used for the initial episode when possible. (See 'Persistent or repeat symptoms' above.)

Individuals with three or more documented BV episodes in one year are defined as having recurrent BV. (See "Bacterial vaginosis: Recurrent infection".)