Sexual Violence, Genital Cytokines, and Colposcopy Findings: A Cross-Sectional Study of Women Engaged in Sex Work in Mombasa, Kenya.

Abstract

Background: Sexual violence (SV) increases human immunodeficiency virus (HIV) susceptibility in a sustained manner. This study evaluated genital cytokines and colposcopy findings in women reporting both recent and more remote SV.

Methods: A cross-sectional study of HIV-1 negative Kenyan women who engage in sex work was performed. Cervicovaginal fluid was collected by menstrual cup and cytokines (IFNγ, TNFα, IL-1β, IL-6, IL-10, MIP-1α, MIP-1β, and CXCL10) measured using chemiluminescence. Cervical injury was assessed by colposcopy. Associations between recent (≤30 days prior), more remote (>30 days prior), and no (reference category) SV exposure and cytokine concentrations were evaluated using linear regression.

Results: Among 282 participants, 25 (8.9%) reported recent SV and 123 (43.6%) reported more remote SV. Only two cytokines (IL-10 and CXCL10) were associated with the 3-category SV variable in bivariable modeling at the prespecified cutoff ( P < 0.2) and carried forward. In multivariable analyses, more remote SV (β = 0.72; 95% confidence interval [CI], 0.06-1.38; P = 0.03), but not recent SV (β = 0.20; 95% CI, -0.99 to 1.39; P = 0.74) was associated with cervicovaginal IL-10 compared with no SV. Recent (β = 0.36; 95% CI, -0.94 to 1.67; P = 0.58) and more remote (β = 0.51; 95% CI, -0.21 to 1.24; P = 0.16) SV were not associated with CXCL10 compared with no SV. Cervical epithelial friability (χ 2 = 1.3, P = 0.51), erythema (χ 2 = 2.9, P = 0.24), vascular disruption (χ 2 = 1.4; P = 0.50), epithelial disruption (χ 2 = 2.6, P = 0.27), or any colposcopy finding (χ 2 = 1.2, P = 0.54) were not associated with SV category by χ 2 test.

Conclusions: The mechanism linking SV to sustained increases in HIV susceptibility may not be related to persistent genital inflammation or injury.