Practical Guide to Diagnosing and Treating Vaginitis

, MD, FACOG, US Naval Medical Center


Medscape General Medicine. 1997;1(2) 

In This Article

Bacterial Vaginosis

BV is a polymicrobial, primarily anaerobic, infection associated with sometimes fishy-smelling increased vaginal discharge (Fig. 1), but not accompanied by leukorrhea, vulvar burning, or pruritis. Infection with BV can have significant sequelae, however. It has been associated with an increased risk of septic abortion, premature rupture of amniotic membranes, preterm labor, preterm delivery, post-Cesarean endomyometritis, and post-hysterectomy pelvic cellulitis.[4,5,6,7,8,9,10,11]

Figure 1. The characteristic milky or creamy vaginal discharge of bacterial vaginosis is associated with a high vaginal pH and a fishy odor. Figure courtesy of James A. McGregor, MD, University of Colorado Health Sciences Center.

Its presence represents a change in the vaginal ecosystem, specifically a decrease in lactobacilli, which is part of the normal flora; a proliferation of pathogenic inhabitants of the vagina; and an elevation of pH (>4.5). Despite exhaustive basic research, it is unclear whether the decline in lactobacilli, proliferation of pathogens, or rise in pH initiates the cascade. While BV was originally described by Curtis in 1911, it gained notoriety only in 1955 after Gardner and Dukes described the offending organism as Haemophilus vaginalis; the organism has since been renamed Gardnerella vaginalis in honor of its discoverer.[1] The current term, bacterial vaginosis, rather than vaginitis, indicates lack of an inflammatory reaction (absence of WBCs in the discharge) and is much more reflective of the true polymicrobial nature of this condition.

Although the majority of pathogens causing BV are derived from the endogenous flora, one genus--Mobiluncus--is unique to patients with BV; it has never been found in patients without vaginosis. Mobiluncus are comma-shaped, gram-variable or gram-negative, anaerobic rods with tapered ends that utilize subpolar flagella for a tumbling motility. G vaginalis is found in nearly 100% of women with symptomatic BV.[12] In fact, G vaginalis may be the organism primarily responsible for the premature rupture of membranes in women with BV. The organism has phospholipase A2 activity, which initiates labor.[13] Other factors may also play a role in early labor, however, as G vaginalis is also found in many normal women with uncomplicated pregnancies.

Whether BV is sexually transmitted remains unsettled. The majority of observations, such as those listed below, suggests that BV is probably not sexually transmitted.

  1. Urethral colonization rates of Gardnerella and Mobiluncus species among male partners of women with BV have been observed to be no higher than those for partners of normal women.[14]

  2. The incidence of BV has not been documented to rise with increasing number of lifetime sexual partners.[14]

  3. Cotreating the male partner has not been shown to reduce the recurrence risk in females (hence the basis for the current recommendation not to cotreat the male partner).[14,15]

However, one contrary piece of evidence, suggesting a sexually transmitted aspect, is that condom use is associated with a decreased incidence of BV among the female partners. Hence, if a patient experiences recurrences of BV, there is some basis to treat the partner.[16]

Bacterial vaginosis identified on wet mount (Fig. 2) represents the most common vaginal infection identified in women with inflammation or altered flora on Papanicolaou smear.[17]

Figure 2. On wet preparation of vaginal fluid, absence of WBCs and stippling of epithelial cells support a diagnosis of bacterial vaginosis. From Infect Med 9(1):50, 1992. © Copyright 1992, SCP Communications, Inc.

Diagnosis. The diagnosis of BV requires the presence of at least 3 of the following 4 criteria (Table I):[18]

  1. A homogenous noninflammatory discharge (not many WBCs).

  2. Vaginal pH >4.5.

  3. Clue cells (bacteria attached to the borders of epithelial cells, >20 % of epithelial cells; Fig. 3).

  4. Whiff test positive for fishy or musty odor when alkaline KOH solution added to smear.

Figure 3. Pap smear showing clue cells consistent with bacterial vaginosis. Courtesy of Abner P. Korn, MD. © Copyright 1996, SCP Communications, Inc.

Applying these diagnostic criteria will result in correctly diagnosing BV more than 90% of the time with a false-positive rate of less than 10%.[18,19] Note that neither Papanicolaou smear nor culture are diagnostic in and of themselves. The Papanicolaou smear may indicate a coccobacillary shift of flora, but this report should prompt patient evaluation for the formal diagnostic criteria rather than empiric treatment. Culture is not helpful since the flora of BV are largely derived from the normal commensal flora.

Treatment. For years, oral metronidazole has been the primary indicated regimen. While very effective, this regimen is fraught with such serious side effects as convulsive seizures and peripheral neuropathy, causing many patients to discontinue their treatment prior to completion of the 7-day regimen. Other systemic options include oral clindamycin. More recently, the advent of intravaginal preparations has afforded topical options that cause less-adverse systemic effects.[20] The amount of drug absorbed and delivered is also much less. For example, metronidazole requires an oral dose of 500mg, while only 37.5mg of metronidazole intravaginal gel is needed to achieve therapeutic vaginal tissue levels.[21,22] Similarly, oral and intravaginal preparations of clindamycin achieve equal efficacy, although a lower dose is delivered intravaginally.[20,23]

While the regimens listed in Table II have been shown to provide 90% or better efficacy, the single 2g dose of metronidazole affords only 70% efficacy and leads to more gastrointestinal adverse effects than clindamycin.[24,25,26,27] The oral clindamycin regimen is associated with a higher incidence of diarrhea than oral metronidazole.[28] A decade ago, ampicillin had traditionally been endorsed as first-line therapy for BV in pregnancy; however, this finding was based more on safety concerns than on efficacy. Given the more recent data suggesting that BV is associated with worse obstetric outcomes,[4,9,10,11] more efficacious therapy is justified. Oral metronidazole alone, or in combination with oral erythromycin, has been shown to reduce adverse obstetric sequelae in high-risk patients with BV.[29,30] Clindamycin cream, while effectively treating the BV, was associated by Amsel's criteria with worse obstetric outcomes.[31,32] This paradox may be explained by clindamycin reducing the endogenous lactobacillus population or by the overgrowth of Escherichia coli and Enterococcus.[23,33] Despite common opinion, exhaustive studies have not shown metronidazole to be teratogenic. It is mutagenic in bacteria and carcinogenic in rodents, but no human data have demonstrated adverse outcomes when subjected to analytic scrutiny. However, several sources, albeit without evidence-based data, state that metronidazole is contraindicated in the first trimester of pregnancy.[15,28]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: