Trichomonas vaginalis is associated with HIV-1 in high-risk migrant men and women living in inner-city settlements in Johannesburg, South Africa.

Trichomonas vaginalis is associated with HIV-1 in high-risk migrant men and women living in inner-city settlements in Johannesburg, South Africa S. Delany-Moretlwe 1 , A. Abdolrasouli 2 , T. Clayton 2 , M. Oliff 1 , J. Vearey 1 , W. Moyo 1 , A. Andreasen 2 , P. Mayaud 2 1 Reproductive Health & HIV Research Unit, University of Witwatersrand, Johannesburg, South Africa 2 London School of Hygiene & Tropical Medicine, London, UK Background Little is known about the epidemiology of Trichomonas vaginalis (TV) in urban populations in Africa. We conducted a survey to determine the prevalence of TV and associations with HIV-1 and other risk factors in high-risk migrant men and women living in inner-city Johannesburg. Methods A cluster-sampling community-based survey was conducted in 2003-2004 among 1458 male hostel residents in Johannesburg and 1002 women in adjacent informal settlements. Participants aged >18 recruited from various locations provided socio-demographic, mobility, behavioural and health data. TV, Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) were determined by PCR in urine samples. Saliva was tested for HIV-1 antibodies using Orasure® ELISA. Predictors of TV were identified using multivariable logistic analysis adjusting for recruitment site. Results TV was detected in 81/1438 (5.6%) men and 188/991 (19.0%) women; 95% and 94% of infected men and women were asymptomatic. 345 (24.0%) men and 549 (55.5%) women were HIV-1 seropositive. Infection with CT and NG were 5.2% and 1.9% in men; and 11.1% and 2.0% in women. HIV-1 was associated with TV in both men (adjusted OR=1.77; 95% CI 1.08–2.91; p=0.023) and women (adjusted OR=1.53; 95% CI 1.07–2.18; p=0.020). Other risk factors for TV are indicated in the Table. These included smoking and an inverse relationship with perception of safety for both men and women, whilst other factors differed by sex. Conclusions TV was highly prevalent and was consistently associated with HIV-1 in this high-risk urban migrant population, with approximately 95% of infections being asymptomatic. Prevention and management of TV in these communities will require screening programmes. The extremely high HIV-1 prevalence rates observed call for urgent and innovative interventions in this high-risk inner-city environment. Table: Multivariable model of associations with T. vaginalis Risk factor TV/Total (%) Odds Ratio 95% CI p-value Men 81/1438 (5.6) Cigarette smoker 50/666 (7.5) 2.27 1.40–3.69 0.001 Alcohol drinker 32/686 (4.7) 0.51 0.31–0.83 0.006 Experienced or witnessed violence in last month 28/351 (8.0) 1.90 1.16–3.12 0.011 Perception of safety Very unsafe 22/434 (5.1) 1.36 1.02–1.80 0.033 Unsafe 32/676 (4.7) Safe 20/273 (7.3) Very safe 7/50 (14.0) Home language/ethnic group isiZulu 73/1376 (5.3) 1 Other 8/62 (12.9) 2.42 1.02–5.76 0.046 HIV-1 seropositive 27/345 (7.8) 1.77 1.08–2.91 0.023 Women 188/991 (19.0) Cigarette smoker 19/63 (30.2) 1.87 1.02–3.43 0.045 History of tuberculosis 25/86 (29.1) 2.13 1.25–3.62 0.006 Perception of safety Very unsafe 30/237 (12.7) 1.33 1.07–1.64 0.009 Unsafe 101/455 (22.2) Safe 43/243 (17.7) Very safe 12/54 (22.2) Reports genital discharge in last 6 months 49/317 (15.5) 0.63 0.43–0.92 0.018 Chlamydia trachomatis infection 40/110 (36.4) 3.05 1.92–4.85 <0.001 HIV-1 seropositive 121/549 (22.0) 1.53 1.07–2.18 0.020