Efficacy of Carraguard for prevention of HIV infection in women in South Africa: A randomised, double-blind, placebo-controlled trial

Document Type

Article (peer-reviewed)

Publication Date



Background: Female-initiated HIV-prevention options, such as microbicides, are urgently needed. We assessed Carraguard, a carrageenan-based compound developed by the Population Council, for its efficacy and long-term safety in prevention of HIV infection in women. Methods: We undertook a randomised, placebo-controlled, double-blind trial in three South African sites in sexually-active, HIV-negative women, aged 16 years and older. 6202 participants, who were randomly assigned by a block randomisation scheme to Carraguard (n=3103) or placebo (methylcellulose [n=3099]), were instructed to use one applicator of gel plus a condom during each vaginal sex act. Participants were followed up for up to 2 years. Visits every 3 months included testing for HIV presence and pregnancy, pelvic examinations, risk reduction counselling, and treatment for curable sexually transmitted infections and symptomatic vaginal infections. The primary outcome was time to HIV seroconversion. Analysis was in the efficacy population (a subset of the intention-to-treat population, excluding participants for whom efficacy could not be assessed). This study is registered with ClinicalTrials.gov, number NCT00213083. Findings: For the primary outcome (time to HIV seroconversion) we analysed 3011 women in the Carraguard group and 2994 in the placebo group. HIV incidence was 3·3 per 100 woman-years (95% CI 2·8-3·9) in the Carraguard group (134 events) and 3·8 per 100 woman-years (95% CI 3·2-4·4) in the placebo group (151 events), with no significant difference in the distribution of time to seroconversion (p=0·30). The covariate-adjusted hazard ratio was 0·87 (95% CI 0·69-1·09). Rates of self-reported gel use (96·2% Carraguard, 95·9% placebo) and condom use (64·1% in both groups) at last sex acts were similar in both groups. On the basis of applicator testing, however, gel was estimated to have been used in only 42·1% of sex acts, on average (41·1% Carraguard, 43·1% placebo). 1420 (23%) women in the intention-to-treat population had adverse events (713 Carraguard, 707 placebo), and 95 (2%) women had adverse events that were related to gel use (48 Carraguard, 47 placebo). Serious adverse events occurred in 72 (2%) women in the Carraguard group and 78 (3%) in the placebo group, only one of which was considered possibly related to gel use (placebo group). Interpretation: This study did not show Carraguard's efficacy in prevention of vaginal transmission of HIV. No safety concerns were recorded. Funding: US Agency for International Development, Bill & Melinda Gates Foundation.