|
|
 |
Abstract
Trichomonas vaginalis as a risk factor for HIV acquisition
B. van der Pol1, C. Kwok2, A. Rinaldi3, R. Salata4, F. Mmiro5, R. Mugerwa5, T. Chipato6, C. Morrison3
1Indiana University School of Medicine, Infectious Diseases, Indianapolis, United States, 2Family Health International, Biostatistics Department, Research Triangle Park, United States, 3Family Health International, Clinical Research Department, Research Triangle Park, United States, 4Case Western Reserve University, Department of Medicine, Cleveland, United States, 5Makerere University, Faculty of Medicine, Kampala, Uganda, 6University of Zimbabwe, Department of Obstetrics and Gynecology, Harare, Zimbabwe
Background: Trichomonas vaginalis (TV) is the most common non-viral STI worldwide with particularly high prevalence in HIV-endemic regions. We used stored samples collected from women enrolled in a longitudinal study to evaluate the impact of TV infection on risk of HIV-1 acquisition.
Methods: HIV-negative women were enrolled in a longitudinal study in Uganda and Zimbabwe from 1999-2004. Quarterly samples were collected for C. trachomatis and N. gonorrhoeae PCR and for microscopy. The 213 women that HIV seroconverted during the study (cases) were matched (on study site, age and time in study) with 2 HIV-uninfected women (controls). Samples from the seroconversion (SC) visit and the previous visit for cases, and equivalent visits for controls, were tested using a TV PCR assay. Logistic regression was performed to test for differences between cases and controls in TV prevalence while controlling for known HIV risk factors.
Results: TV PCR identified 80 infections while wet mount identified only 35 (44%) of these infections. TV prevalence (by PCR) at either visit was 17.4% and 6.4% in cases and controls, respectively (p<.001). In multivariate analysis controlling for other STIs, behavioral, and demographic factors, the adjusted odds ratio for HIV acquisition was 4.36 (95% CI 1.55, 12.30), 4.27 (95% CI 1.11, 16.44), and 1.82 (95% CI 0.57, 5.78), for TV identified at the prior visit only, the SC visit only, or at both visits, respectively.
Conclusions: TV is strongly associated with increased risk for HIV infection in this population of African women. Few longitudinal studies have examined the effect of TV infection on HIV acquisition using PCR rather than wet mount. In this study, sensitivity of wet mount was only 44%, resulting in many untreated infections. Given the high prevalence of TV in HIV-endemic areas, TV treatment may have a substantial impact on preventing HIV acquisition among women.
Back
|