[show abstract][hide abstract] ABSTRACT: HIV-infected women need highly effective contraception to reduce unintended pregnancies and mother-to-child HIV transmission. Previous studies report conflicting results regarding the effect of hormonal contraception on HIV disease progression.
HIV-infected women in Uganda and Zimbabwe were recruited immediately after seroconversion; CD4 testing and clinical examinations were conducted quarterly. The study end point was time to AIDS (two successive CD4 200 cells/mm or less or World Health Organization advanced Stage 3 or Stage 4 disease). We used marginal structural Cox survival models to estimate the effect of cumulative exposure to depot-medroxyprogesterone acetate and oral contraceptives on time to AIDS.
Three hundred three HIV-infected women contributed 1408 person-years. One hundred eleven women (37%) developed AIDS. Cumulative probability of AIDS was 50% at 7 years and did not vary by country. AIDS incidence was 6.6, 9.3, and 8.8 per 100 person-years for depot-medroxyprogesterone acetate, oral contraceptive, and nonhormonal users. Neither depot-medroxyprogesterone acetate (adjusted hazard ratio, 0.90; 95% confidence interval, 0.76-1.08) nor oral contraceptives ( adjusted hazard ratio, 1.07; 95% confidence interval, 0.89-1.29) were associated with HIV disease progression. Alternative exposure definitions of hormonal contraception use during the year before AIDS or at the time of HIV infection produced similar results. Sexually transmitted infection symptoms were associated with faster progression, whereas young age at HIV infection (18-24 years) was associated with slower progression. Adding baseline CD4 level and set point viral load to models did not change the hormonal contraception results, but Subtype D infection became associated with disease progression.
Hormonal contraceptive use was not associated with more rapid HIV disease progression, but older age, sexually transmitted infection symptoms, and Subtype D infection were.
[show abstract][hide abstract] ABSTRACT: High levels of HIV-1 viremia exist in peripheral blood during acute and early infection; however, data on HIV-1 viral loads in female genital secretions during this period are sparse.
Prospective cohort of 188 African women with primary HIV-1 infection.
HIV-uninfected and infected women were followed quarterly; we tested serial plasma specimens by HIV PCR to estimate infection dates. We used the Loess procedure to estimate the magnitude and timing of viral setpoints in plasma and cervical secretions and generalized estimating equations (GEE) to identify predictors of plasma and cervical viral setpoints.
We estimated the mean HIV-1 plasma setpoint to be 4.20 log10 HIV-1 RNA copies/ml [95% confidence interval (CI) 4.04-4.35] at 121 days (95% CI 91-137) from infection; an analogous mean cervical viral setpoint was 1.64 log10 HIV-1 RNA copies/swab (95% CI 1.46-1.82) at 174 days (95% CI 145-194) from infection. Cervical loads were significantly higher (0.7-1.1 log10 copies/swab) during acute infection than subsequently. Subtype D infection, pregnancy, breastfeeding, and older age at the time of infection were associated with higher plasma viral setpoint. Subtype C infection, nonviral sexually transmitted infections, having a partner spending nights away from home, recent unprotected sex, and shorter time since infection were associated with higher cervical HIV-1 loads. Hormonal contraception was not associated with either the HIV-1 plasma setpoint or cervical loads during early infection.
Cervical HIV-1 viral loads were highest during acute infection and then declined up to 6 months following infection, when a 'setpoint' was attained. The prognostic value of a cervical 'setpoint' on future transmission risk remains unclear.
AIDS (London, England) 02/2010; 24(4):573-82. · 4.91 Impact Factor
[show abstract][hide abstract] ABSTRACT: Trichomoniasis vaginalis is the most common nonviral sexually transmitted infection (STI) worldwide, with a particularly high prevalence in regions of human immunodeficiency virus (HIV) endemicity. However, its impact as a cofactor for HIV acquisition is poorly understood.
Samples from 213 women who experienced HIV seroconversion (cases) during a longitudinal study involving 4450 women in Uganda and Zimbabwe were matched with samples from HIV-uninfected women (controls). All samples underwent polymerase chain reaction (PCR) analysis for Trichomonas vaginalis DNA. For cases, analyzed samples were from the visit in which HIV seroconversion was detected and the visit preceding detection of seroconversion; for controls, one analyzed sample was from the visit matched by follow-up duration to the cases' seroconversion visit, and the other sample was from the visit immediately preceding the matched visit.
The prevalence of T. vaginalis infection before HIV infection was 11.3% in cases and 4.5% in controls (P = .002). In multivariable analysis controlling for hormonal contraception, other STIs, behavioral, and demographic factors, the adjusted odds ratio for HIV acquisition was 2.74 (95% confidence interval, 1.25-6.00) for T. vaginalis-positive cases. The presence of behavioral risk factors for HIV infection, study recruitment from a referral population at high-risk for HIV, primary sex partner-associated risk for HIV infection, and herpes simplex virus type 2 seropositivity were also predictive of incident HIV infection.
T. vaginalis infection is strongly associated with an increased risk for HIV infection in this general population of African women. Given the high prevalence of T. vaginalis infection in HIV-endemic areas, T. vaginalis control may have a substantial impact on preventing HIV acquisition among women.
The Journal of Infectious Diseases 03/2008; 197(4):548-54. · 5.85 Impact Factor