Oral and vaginal antifungal treatments are similarly effective for vulvovaginal candidiasis - meta-analysis

OBJECTIVE To compare the e¡ectiveness, safety, and cost of oral and vaginal antifungal treatments for uncomplicated vulvovaginal candidiasis. DESIGN Meta-analysis of randomized, controlled trials. DATA SOURCES Computerized searches of the Cochrane Controlled Trials Registers, MEDLINE, and EMBASE, hand searches of the reference lists of relevant articles, and consultation with drug companies. STUDY SELECTION Articles in any language that described randomized trials comparing oral and vaginal antifungal treatment (imidazole or triazole) in nonpregnant women with uncomplicated, symptomatic, acute, mycologically con¢rmed vulvovaginal candidiasis. Studies involving women with chronic infection were excluded unless such women represented o20% of the subjects or separate results were available for those with acute infection. Trials involving ketoconazole were excluded. Studies were assessed for inclusion and methodological quality independently by two reviewers. DATA EXTRACTION Data were extracted independently by two reviewers. MAIN OUTCOME MEASURES Overall odds ratio (OR, 95% CI) for short-term (5-15 days) or longterm (2-12 weeks) clinical or mycological cure in women receiving oral medication compared to those receiving vaginal medication. MAIN RESULTS The meta-analysis included 17 trials, published between 1989 and 1995, providing 19 comparisons involving 2919 women. The main oral drug used was fluconazole (15 comparisons), usually 150 mg single dose, but also 200 mg single dose or 50 mg daily for 3 or 6 days. The main vaginal drug used was clotrimazole (14 comparisons), 500 mg single dose, 200 mg daily for 3 days, or 100 mg daily for 6-7 days. Rates of clinical cure with both routes of treatment were 80% in the short term and >80% in the long term. From nine trials, the OR for short-term clinical cure was 1.0 (0.7-1.4) and from seven trials, the OR for long-term clinical cure was 1.0 (0.7-1.5). Rates of mycological cure were >80% with both routes in the short term and ~70% in the long term. From 17 trials, the OR for short-term mycological cure was 1.2 (0.9-1.7) and from 14 trials, the OR for long-term mycological cure was 1.3 (1.0-1.7). No significant difference for any outcome was found in subgroup analyses comparing oral fluconazole vs vaginal clotrimazole, single-dose oral vs single-dose vaginal, or single-dose oral vs multiple-dose vaginal. The incidence of side-effects was similar for the two routes of treatment. Oral treatment was preferred by the women in each of the 10 trials that reported this outcome. No trial reported information on costs. CONCLUSION Antifungal treatment for uncomplicated vulvovaginal candidiasis was similarly e¡ective when administered by either the oral or vaginal route.