Bacterial vaginosis (BV) is a prevalent vaginal infection in which the
normal vaginal flora is altered to contain large numbers of anaerobes as
well as Gardnerella vaginalis and other species. Bacterial vaginosis
usually presents as a symptomatic malodorous discharge. It is
generally diagnosed by the presence of 3 of the following 4 signs: (1)
presence of clue cells; (2) vaginal fluid pH greater than 4.5; (3) fishy
odor arising from discharge before or after adding 10% potassium
hydroxide ("whiff test"); and (4) homogeneous noninflammatory
discharge adherent to vaginal walls.
Although controversial, there is some evidence for sexual transmission
of the microorganisms associated with BV. Coupled with the estimate
that BV carries a recurrence rate as high as 30%,2 this invokes the
question of whether treatment of the male sex partner influences
recurrence of BV.
There is no evidence that treating
male partners of women with recurring
BV infections improves cure or
recurrance rates
-Maura Hamrick, MD, PhD
Our literature search revealed 6 randomized placebo-controlled trials
addressing this question.3-8 All trials were clinic-based, employed the
diagnostic criteria for BV listed above, and excluded pregnant patients
and those with coexistent vaginal infections. Of the trials, Moi et al3
and Colli et al4 provided the longest follow-up periods. In a
double-blind study, Moi and colleagues randomized 241 couples after
which the female participants were given two 2-g doses of
metronidazole, 2 days apart. Their male partners were given either this
same medication regimen or a placebo. In an intention-to-treat analysis
performed at 12 weeks' follow-up, they found partner treatment to
have no effect on recurrence of symptoms or clinical BV. Colli and
colleagues also performed a double-blind study in which 139 females
received a 7-day course of 2% clindamycin vaginal cream. Their
partners were randomized to receive either a placebo or a 7-day course
of oral clindamycin. Again, an intention-to-treat analysis at 12 weeks'
follow-up revealed no difference between the groups in recurrence of
BV.
Other trials of shorter duration have shown similar results. In
double-blind trials, Vejtorp et al5 and Vutyavanich et al6 found partner
treatment to make no difference in 5- and 4-week recurrence rates,
respectively. Similarly, in a single-blind trial, Swedberg et al7 found
partner treatment to have no effect on 3-week recurrence rates.
Various metronidazole or tinidazole regimens were used in these 3
studies (though not all were in keeping with current Centers for
Disease Control and Prevention recommended regimens).
Only one study reported a statistically significant benefit of partner
treatment on BV outcomes. Mengel et al8 conducted an 8-week
double-blind trial designed primarily to determine whether a single dose
of metronidazole is as effective for BV as 7-day therapy. Additionally,
they analyzed the effect of single-dose metronidazole treatment for
partners on outcomes. After randomizing a total of 98 couples, the
authors found a statistically significant improvement in patients'
symptoms at 8-week follow-up when partners were also treated, but
no difference at 2- and 5-week follow-up. They also reported a
statistically significant improvement in cure rates at 2 weeks after
treatment but, unlike the other studies, this was determined by Gram
stain rather than wet-mount criteria. There was no difference in
2-week cure rates by wet mount criteria in this study. There was also
no difference in recurrence rates at the 5- and 8-week follow-ups,
although this was judged by Gram stain criteria only.
Conclusions from this last study must be drawn cautiously, as it had
several problem areas. The design was complex and multiple outcome
comparisons were made. Because the outcome data were presented
only as graphs, making numerical comparisons is difficult. Also, unless
patients had recurrent symptoms, they were examined only at 2-week
follow-up. Five- and 8-week follow-up consisted of telephone
interviews for symptoms and Gram stain evaluation of self-collected
vaginal fluid, which was mailed to the study group. Gram staining as a
diagnostic method in BV is at best controversial and is not commonly
used in clinical practice to diagnose BV. The improvement in
telephone-reported symptoms at 8 weeks was not reflected in a
decreased number of return visits for recurrent symptoms.
The majority of the evidence, therefore, supports the conclusion that in
the general BV population, partner treatment confers no benefit on
recurrence rates for up to 12 weeks after initial treatment. This
conclusion has held across 5 studies using a variety of treatment
regimens for both the patient and the partner. Two of these 5 studies4,
7 use a treatment regimen for females in accordance with current
Centers for Disease Control and Prevention guidelines,1 so the results
should be generalizable to what is currently practiced in most settings.
One could argue that the above trials do not answer the question at
hand, since they do not analyze a specifically recurrent BV cohort.
However, the largest of the above trials reported 20% to 30%
recurrence rates. Given such high rates, the fact that there was not
even a trend toward decreased recurrence rates makes it unlikely that a
study of a specifically recurrent cohort would show different results.
Despite this lack of evidence supporting it, several textbooks in various
fields recommend a trial of partner treatment for recurrent BV.2, 9, 10
The risks and benefits must be fully weighed before making such a
recommendation, however. Though treatment of the male partner
carries few physiologic adverse effects, there may be emotional
adverse effects to implying that BV is a sexually transmitted disease.
THE BOTTOM LINE :
There is no evidence that treating male partners of patients with
recurrent BV improves cure or recurrence rates. Although many
textbooks recommend considering partner treatment, the emotional
ramifications of implying that BV is sexually transmitted should be
taken into consideration before implementing a treatment of no proven
benefit.
AUTHOR INFORMATION
Maura Hamrick, PhD, MD
M. Lee Chambliss, MD, MSPH
Moses Cone Family Medicine Residency
1125 N Church St
Greensboro, NC 27401-1007
