New concepts in bacterial vaginosis
Barbara A. Majeroni
Numerous studies performed during the past decade have contributed to a better understanding of the nature, etiology and treatment of bacterial vaginosis. Previously, the term “nonspecific vaginitis” was used to describe a vaginal infection in the absence of Trichomonas vaginalis or Candida albicans.
In the 1950s, the condition now known as bacterial vaginosis was found to be frequently associated with a gram-variable coccobacillus that was called both Haemophilus vaginalis and Corynebacterium vaginalis. The organism was renamed Gardnerella vaginalis in honor of Gardner, who, along with Dukes, did much of the work in isolating and identifying the organism. G. vaginalis was thought to be the causative agent in bacterial vaginosis. However, the organism was also found in 40 to 50 percent of women who did not have signs or symptoms of vaginal disease. 
The condition formerly attributed to G. vaginalis is actually a polymicrobial, superficial vaginal infection caused by an increase in anaerobic organisms and a concomitant decrease in lactobacilli, which normally are the predominant bacteria in the vagina. [2,3] It is uncertain whether overgrowth of anaerobic organisms causes a reduction in lactobacilli or whether the reduced number of lactobacilli permits the overgrowth of anaerobic organisms. [4,5] Because of the superficial nature of the infection, the condition is termed vaginosis rather than vaginitis.
Risk Factors and Presentation
Risk factors for bacterial vaginosis include multiple sexual partners, concurrent or prior T. vaginalis infection, and the use of an intrauterine device or other nonbarrier contraception method.  The presence of G. vaginalis in vaginal secretions should not be considered evidence of sexual contact. 
In pregnant women, bacterial vaginosis has been associated with an increased incidence of premature rupture of the membranes, preterm labor and amniotic infection.  Bacterial vaginosis may also increase the risk of pelvic inflammatory disease. 
Patients with bacterial vaginosis usually have a homogeneous, malodorous vaginal discharge of variable color. The odor, which is often described as “fishy,” is caused by amines released by anaerobic flora. Release of the amine odor is accentuated with alkaline pH, which would occur with potassium hydroxide (KOH), and also in the presence of semen.
Signs and symptoms of bacterial vaginosis are unreliable as diagnostic criteria, and the history is not helpful in differentiating this condition from other causes of vaginal symptoms.  The diagnosis of bacterial vaginosis is based on at least three of the five criteria listed in Table 1.
Microscopic examination of vaginal secretions in bacterial vaginosis typically reveals clue cells. These are epithelial cells studded with coccobacilli, which obscure the borders of the cells so that the edges appear fuzzy rather than smooth (Figure 1). In bacterial vaginosis, more than 20 percent of the vaginal epithelial cells will be clue cells, which corresponds to at least one clue cell per low-power field.  Vacuolated epithelial cells, which are sometimes mistaken for clue cells, have some central stippling, but the edges of the cells appear sharp (Figure 2).
A few drops of 10 percent KOH added to a smear of vaginal secretions produces an amine or fishy odor, which gives rise to the term “sniff” test. The test can be done using the same slide to which KOH has
Diagnostic Criteria for Bacterial Vaginosis (*)
Homogeneous vaginal discharge
Presence of clue cells (>20 percent) on wet mount
Amine odor with addition of 10 percent
postassium hydroxide (positive “sniff” test)
Vaginal pH >4.7
Absence of lactobacilli on wet mount
(*)–The presence of three criteria is required for
been added to assist in the identification of C. albicans.
While normal vaginal secretions have a pH of less than 4.5, the pH of vaginal secretions in bacterial vaginosis is greater than 4.7. Direct application of pH paper to secretions taken from the vaginal pool will give an accurate result. The specimen should not be taken from the cervix, since the pH of cervical mucus is normally higher than that of vaginal secretions.
Long, thin lactobacilli are abundant in normal vaginal flora (Figure 3). However, lactobacilli are rare or absent in the wet mount of vaginal secretions from patients with bacterial vaginosis.
Because of the superficial nature of bacterial vaginosis, few leukocytes are seen in the wet mouth.  If numerous leukocytes are present, the possibility of concurrent trichomonal vaginitis and/or gonorrheal or chlamydial cervicitis should be considered.
In patients with vaginal discharge, pain or itching, the physical examination, wet mount, KOH preparation and vaginal pH measurement will usually differential the four major causes of vaginal symptoms (Table 2).
Other studies that may be indicated include a Gram stain of an endocervical specimen, which should be examined for the presence of leukocytes and gram-negative intracellular diplococci. Cultures or studies for gonorrheal and chlamydial infection also may be useful. Gonorrheal and chlamydial infections generally do not cause vaginitis, but they may cause a purulent cervicitis that produces leukorrhea. Cultures[TABULAR DATA OMITTED]
for Bacterial Vaginosis
Metronidazole (Flagyl, Protostat), 500 mg twice
daily for seven days or a single 2-g dose (*)
If metronidazole is contraindicated, clindamycin
(Cleocin), 300 mg twice daily for seven days
(*)–Precautions for using metronidazole are given
in the text.
for G. vaginalis are no longer considered useful.
Current treatment recommendations for bacterial vaginosis are summarized in Table 3. The drug of choice is metronidazole (Flagyl, Protostat), which is usually administered in a dosage of 500 mg twice daily for at least seven days.  A single 2-g dose of the drug is also effective,  although the recurrence rate may be higher with single-dose therapy than with the longer regimen. Single-dose therapy is also associated with an increased incidence of adverse effects, primarily nausea and dysgeusia. 
The efficacy of metronidazole, which is 95 percent or greater, is thought to be related to the activity of its hydroxy metabolites against G. vaginalis and to the elimination of anaerobes. Two randomized, double-blind trials [15,16] have demonstrated no benefit from treating the sexual partner.
Metronidazole can interfere with the metabolism of warfarin (Coumadin, Panwarfin) and anticonvulsants. The halflife of metronidazole is shortened in patients taking barbiturates, and a higher dose may be required.
Patients must be warned to avoid alcohol while taking metronidazole. A disulfiram-like reaction may occur if alcohol is ingested within 24 hours of metronidazole administration.
Metronidazole is contraindicated during the first trimester of pregnancy, and many clinicians are reluctant to use the drug at all in pregnant patients.
A recent study  has shown clindamycin (Cleocin) to be an effective alternative to metronidazole. Clindamycin may be safer for pregnant women and for patients who are unlikely to abstain from alcohol during treatment. However, clindamycin is considerably more expensive than metronidazole, and it is associated with an increased incidence of diarrhea and colitis.
Erythromycin and the sulfonamides are ineffective in the treatment of bacterial vaginosis. Oral tetracycline has a cure rate ranging from 13 to 65 percent. Although the cure rate for ampicillin ranges from 50 to 70 percent, ampicillin destroys the remaining lactobacilli and may delay repopulation of normal vaginal flora. 
 Eschenbach DA, Hillier S, Critchlow C, Stevens C, DeRouen T, Holmes KK. Diagnosis and clinical manifestations of bacterial vaginosis. Am J Obstet Gynecol 1988;158:819-28.
 Weaver CH, Mengel MB. Bacterial vaginosis. J Fam Pract 1988;27:207-15.
 Erkkola R, Jarvinen H, Terho P, Meurman O. Microbial flora in women showing symptoms of nonspecific vaginosis: applicability of KOH test for diagnosis. Scand J Infect Dis 1983;40 (Suppl):59-63.
 Fredricsson B, Englund K, Weintraub L, Olund A, Nord CE. Bacterial vaginosis is not a simple ecological disorder. Gynecol Obstet Invest 1989;28:156-60.
 Sobel JD. Bacterial vaginosis–an ecologic mystery. Ann Intern Med 1989;111:551-3.
 Bump RC, Buesching WJ 3d. Bacterial vaginosis in virginal and sexually active adolescent females: evidence against exclusive sexual transmission. Am J Obstet Gynecol 1988;158:935-9.
 Holst E, Wathne B, Hovelius B, Mardh PA. Bacterial vaginosis: microbiological and clinical findings. Eur J Clin Microbiol 1987;6:536-41.
 Martius J, Krohn MA, Hillier SL, Stamm WE, Holmes KK, Eschenbach DA. Relationships of vaginal Lactobacillus species, cervical Chlamydia trachomatis, and bacterial vaginosis to preterm birth. Obstet Gynecol 1988;71:89-95.
 Reed BD, Huck W, Zazove P. Differentiation of Gardnerella vaginalis, Candida albicans, and Trichomonas vaginalis infections of the vagina. J Fam Pract 1989;28:673-80.
 Thomason JL, Gelbart SM, Anderson RJ, Walt AK, Osypowski PJ, Broekhuizen FF. Statistical evaluation of diagnostic criteria for bacterial vaginosis. Am J Obstet Gynecol 1990;162:155-60.
 Gardner HL, Dukes CD. Haemophilus vaginalis vaginitis. Am J Obstet Gynecol 1955; 69:962-76.
 CDC releases updated guidelines for STD treatment. Am Fam Physician 1989;40(6):199-202.
 Mengel MB, Berg AO, Weaver CH, et al. The effectiveness of single-dose metronidazole therapy for patients and their partners with bacterial vaginosis. J Fam Pract 1989;28:163-71.
 McCue JD. Evaluation and management of vaginitis. An update for primary care practitioners. Arch Intern Med 1989;149:565-8.
 Moi H, Erkkola R, Jerve F, et al. Should male consorts of women with bacterial vaginosis be treated? Genitourin Med 1989;65:263-8.
 Vejtorp M, Bollerup AC, Vejtorp L, et al. Bacterial vaginosis: a double-blind randomized trial of the effect of treatment of the sexual partner. Br J Obstet Gynaecol 1988;95:920-6.
 Greaves WL, Chungafung J, Morris B, Haile A, Townsend JL. Clindamycin versus metronidazole in the treatment of bacterial vaginosis. Obstet Gynecol 1988;72:799-802.
BARBARA A. MAJERONI, M.D. is a clinical assistant professor of family medicine at the State University of New York at Buffalo School of Medicine and Biomedical Sciences. She also serves as medical director of the urban family care center at Erie County Medical Center in Buffalo. Dr. Majeroni graduated from the Medical College of Pennsylvania, Philadelphia, and completed a family practice residency at Hamot Medical Center in Erie, Pa.
COPYRIGHT 1991 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group