326
Predictors of High HIV-1 Cervical Viral Load during Early HIV-1 Infection among Women in Uganda and Zimbabwe
Charles Morrison*1, C Kwok1, K Demers2, A Rinaldi1, B Van Der Pol3, P L Chen1, T Chipato4, F Mmiro5, E Arts2, and R Salata2
1Family Hlth Intl, Research Triangle Park, NC, US; 2Case Western Reserve Univ, Cleveland, OH, US; 3Indiana Univ Sch of Med, Indianapolis, US; 4Univ of Zimbabwe, Harare; and 5Makerere Univ, Kampala, Uganda
Background: Several studies
have examined levels and correlates of HIV-1 cervical shedding associated with
prevalent HIV-1 infection. However, sparse information is available concerning HIV-1
cervical shedding during early HIV-1 infection. We analyzed levels and
predictors of HIV-1 cervical shedding during early (<6 months) HIV-1
infection using data from a large, prospective study of hormonal contraception and
the risk of HIV-1 acquisition and a follow-on study of women with primary HIV-1
infection.
Methods:
Within 6 months
of acquiring HIV-1 infection, 155 Ugandan and Zimbabwean women, aged 18 to 35
years, with primary HIV-1 infection, contributed 442 cervical specimens. We defined
infection date as the midpoint between the first HIV-1+ (enzyme
immunoassay [EIA]+/Western blot+)
and the last HIV-1– visit while acute infections (EIA–/polymerse
chain reaction [PCR]+) were
defined as occurring 15 days before the first PCR positive visit. We used the
Loess procedure to examine the overall trend of virologic responses.
Generalized estimating equations were used to evaluate the impact of covariates
on cervical HIV-1 RNA levels.
Results:
The mean
cervical HIV-1 load during early infection was 2.07 log10 HIV-1
copies/swab. Mean cervical loads were higher during acute infection (3.07 log10
copies/swab) than 1 to 2, 2 to 4, and 4 to 6 months post-infection (2.23,
1.99, 1.91 log10 HIV-1 copies/swab, p ≤0.01, respectively). A direct correlation was
observed between HIV-1 cervical and plasma viral RNA levels during early
infection (Spearman’s r = 0.51; p <0.0001) (see the figure). Breastfeeding
(0.37 log10 copies; p = 0.04)
and Depo-Provera use at time of infection (0.23 log10 copies; p = 0.08) were associated with higher HIV-1
cervical loads, while pregnancy (–0.39 log10 copies; p = 0.01) and longer time since
infection (–0.14 log10 copies per month; p <0.001) were associated with decreased viral loads. Site, age,
condom use, and participant and partner sexual risk behaviors were not
associated with cervical HIV-1 levels.
Conclusions:
Cervical HIV-1 RNA
levels are high during acute infection and then decline as long as 6 months
post-infection, in parallel with plasma viremia. Breastfeeding and Depo-Provera
use are associated with higher cervical HIV-1 loads while pregnancy is
associated with decreased cervical HIV-1 RNA loads.
|
