Abstract
Background:
Inflammation in the female genital tract is a key risk factor for HIV acquisition, but it remains unclear whether clinical or immunological measures best predict risk. We aimed to compare HIV acquisition among women with clinically and/or immunologically defined inflammation.
Setting:
HIV-uninfected women enrolled in the CAPRISA 004 tenofovir gel randomized controlled trial in South Africa were followed for up to 34 months.
Methods:
We analyzed data from 889 women, with cytokine measurements available for 774 participants. Clinical genital abnormalities were assessed at scheduled visits, and 9 cytokines were measured in cervicovaginal lavage samples. HIV incidence was compared across categories of clinical and immunological inflammation using time varying Cox proportional hazards models, adjusting for relevant covariates.
Results:
Immunological inflammation, defined as ≥9 elevated cytokines, was present in 18% (140/774) of women. Among specific clinical signs, abnormal genital discharge (adjusted hazard ratio: 2.67, 95% confidence interval [CI]: 1.14 to 6.23,
P
= 0.024) and cervicitis (adjusted hazard ratio: 10.34, 95% CI: 2.46 to 43.65,
P
= 0.001) were significantly associated with increased HIV acquisition. Women with both clinical and immunological inflammation had the highest risk of HIV acquisition, with adjusted hazard ratios of 2.08 (95% CI: 1.10 to 3.91,
P
= 0.022) and 2.46 (95% CI: 1.21 to 5.03,
P
= 0.013), respectively.
Conclusions:
Clinical and immunological definitions of inflammation were each independently associated with increased HIV acquisition risk and combined, they reflected greater susceptibility. These findings highlight the important role of genital inflammation in women's HIV susceptibility, suggesting that clinical signs may provide practical early indicators of risk even as cytokine profiles provide more sensitive measures of underlying inflammation.