In West Africa, Single-Dose Antibiotic Treats Vaginal Discharge as Effectively as Multiple-Dose Treatment
Among West African women visiting primary care clinics for treatment of vaginal discharge, those given a new single-dose antibiotic treatment are as likely as their counterparts given a standard multiple-dose antibiotic treatment to report relief of their symptoms, even if they are infected with HIV.1 In a randomized controlled trial, two-thirds of women reported complete resolution of their symptoms by two weeks and about one-third reported partial resolution—regardless of whether they had been treated with a single dose of tinidazole and fluconazole or with multiple doses of metronidazole and vaginal clotrimazole over seven days. The findings were similar among HIV-positive women and among women with different genital infections.
The study population consisted of women visiting primary care clinics in Ghana, Guinea, Mali and Togo during 2004–2005 because of vaginal discharge. At their initial visit, women completed a questionnaire asking about demographic and behavioral factors, symptoms and antibiotic use; underwent a gynecologic examination during which cervical and vaginal samples were collected for pathogen testing; and gave a blood sample for HIV testing. They were then assigned in roughly equal numbers to either a single-dose regimen (tinidazole and fluconazole) or a standard multiple-dose regimen (seven days of oral metronidazole and three days of vaginal clotrimazole). Those with signs of cervicitis or genital ulcers were given additional antibiotics. During a visit two weeks later, the women again completed questionnaires about their symptoms and underwent a gynecologic examination. All women who still reported having discharge were given the multiple-dose regimen as a second treatment.
A total of 1,570 women were enrolled in the trial. Three-fourths were aged 21 years or older. Thirty-seven percent were sex workers; among women who were not sex workers, 81% had had only one sexual partner in the past three months. In the majority of women (63%), the discharge had been present for two weeks or less. Substantial proportions of women with discharge also had vulvar itching (83%), low abdominal pain (35%) and pain during intercourse (31%) or urination (17%). About a third had already been treated for the discharge. Some 52% had signs of cervicitis and 5% had genital ulcers.
Tests for three vaginal infections showed that 46% of both sex workers and non–sex workers had bacterial vaginosis; 24% and 35%, respectively, had candidiasis; and 16% and 6%, respectively, had trichomoniasis. Tests for three cervical infections—chlamydia, gonorrhea and Mycoplasma genitalium—showed that 21% of sex workers and 5% of non–sex workers had at least one of these infections. In addition, 26% of sex workers and 8% of non–sex workers were HIV-positive.
In all, 1,242 women returned for the two-week follow-up visit. Among these women, the proportion who reported that their symptoms had completely resolved did not differ between those given the single-dose regimen (66%) and those given the multiple-dose regimen (64%). Similarly, there was no difference between the single-dose group and the multiple-dose group in the proportion reporting that their symptoms had partially resolved (33% and 34%).
In stratified analyses based on women's reports, the two regimens were equally effective, whether women initially had candidiasis, bacterial vaginosis or trichomoniasis, or a cervical infection. Moreover, among HIV-positive women, the single-dose and multiple-dose regimens yielded similar levels of complete resolution (71% vs. 72%) and partial resolution (28% vs. 25%); these were comparable to levels of complete resolution (68% vs. 65%) and partial resolution (32% vs. 33%) among HIV-negative women. Within each study country, women reported similar responses to the two regimens.
The presence of visible vaginal discharge at the two-week visit was associated with women's perceived response to treatment: In all, 93% of those reporting no resolution and 86% of those reporting a partial resolution in symptoms had visible discharge, compared with only 15% of those reporting complete resolution. Furthermore, the proportion with visible discharge did not differ by treatment regimen, in the study population overall or in the stratified subgroups.
Among women who initially had bacterial vaginosis, the proportion who still had it at the two-week visit did not differ between those who reported complete resolution of symptoms and those who did not (28% vs. 33%) or between those who did and did not have visible discharge (31% vs. 29%). In contrast, among women who initially had candidiasis, the proportion who still had it at the two-week visit was significantly higher among women who reported partial or no resolution of symptoms than among their counterparts who reported complete resolution (49% vs. 25%), and significantly higher among women who still had visible discharge than among those who did not (47% vs. 24%).
In the subset of women who had persistent symptoms and received the multiple-dose regimen as a second treatment, nearly all had complete resolution (81%) or partial resolution (18%) of their symptoms after an additional two weeks. These proportions were identical for women initially given the single-dose treatment and women initially given the multiple-dose treatment. Moreover, they did not differ between women who did and did not have bacterial vaginosis at two weeks, or between women who did and did not have candidiasis at two weeks. In addition, HIV-positive women were as likely as their HIV-negative counterparts to report relief of symptoms after the second treatment.
In a final analysis done to assess possible indicators of cervical infection, none of five signs of cervicitis found on initial examinations (cervical discharge, pus on a cervical swab, cervical bleeding after sampling, cervical inflammation or tenderness with motion of the cervix) was associated with the presence of one or more cervical infections (gonorrhea, chlamydia or M. genitalium). However, among women who were sex workers, the greater the number of white blood cells in cervical secretions, the more likely it was that a woman had a cervical infection; among women who were not sex workers, those who had spermatozoa in vaginal secretions were significantly more likely to have a cervical infection.
Commenting on the findings, the researchers argue that single-dose tinidazole-fluconazole is as effective as multiple-dose metronidazole-clotrimazole for relieving symptoms of vaginal discharge, regardless of underlying infections or HIV status. The single-dose regimen is inexpensive, easily adhered to and well tolerated, they point out; therefore, it "should be considered as a first-line treatment for vaginal discharge syndrome." They add that among women who are sex workers, the visit during which this treatment is dispensed should also be used to promote condom use and screen for cervical infections.—S. London
1. Pépin J et al., The syndromic management of vaginal discharge using single-dose treatments: a randomized controlled trial in West Africa, Bulletin of the World Health Organization, 2006, 84(9):729–738.