Abstract
BACKGROUND: HIV-infected women need access to safe contraception. We hypothesized that women using depomedroxyprogesterone acetate (DMPA) contraception would have faster HIV disease progression than women using oral contraceptive pills (OCPs) and nonhormonal methods. METHODS: In a previously reported trial, we randomized 599 HIV-infected women to the intrauterine device (IUD) or hormonal contraception. Women randomized to hormonal contraception chose between OCPs and DMPA. This analysis investigates the relationship between exposure to hormonal contraception and HIV disease progression [defined as death, becoming eligible for antiretroviral therapy (ART), or both]. RESULTS: Of the 595 women not on ART at the time of randomization, 302 were allocated to hormonal contraception, of whom 190 (63%) initiated DMPA and 112 (37%) initiated OCPs. Women starting IUD, OCPs, or DMPA were similar at baseline. Compared with women using the IUD, the adjusted hazard of death was not significantly increased among women using OCPs [1.24; 95% confidence interval (CI) 0.42-3.63] or DMPA (1.83; 95% CI 0.82-4.08). However, women using OCPs (adjusted hazard ratio (AHR) 1.69; 95% CI 1.09-2.64) or DMPA (AHR 1.56; 95% CI 1.08-2.26) trended toward an increased likelihood of becoming eligible for ART. Women exposed to OCPs (AHR 1.67; 95% CI 1.10-2.51) and DMPA (AHR 1.62; 95% CI 1.16-2.28) also had an increased hazard of meeting our composite disease progression outcome (death or becoming ART eligible) than women using the IUD. CONCLUSION: In this secondary analysis, exposure to OCPs or DMPA was associated with HIV disease progression among women not yet on ART. This finding, if confirmed elsewhere, would have global implications and requires urgent further investigation.