Syphilis serology and HIV infection in Harare, Zimbabwe

• Published in: Sexually transmitted infections 1999-12, Vol.75 (6), p.426-430
• Main Authors: Gwanzura, L, Latif, A, Bassett, M, Machekano, R, Katzenstein, D A, Mason, P R
• Format: Article
• Language: English
• Subjects: AIDS/HIV; Biological and medical sciences; Cohort Studies; Enzyme-Linked Immunosorbent Assay; Female; HIV Infections - complications; HIV Infections - diagnosis; HIV Infections - epidemiology; Human viral diseases; Humans; Infectious diseases; Male; Medical sciences; Original; Prevalence; Reproducibility of Results; Syphilis - epidemiology; Syphilis Serodiagnosis - methods; Tropical medicine; Viral diseases; Viral diseases of the lymphoid tissue and the blood. Aids; Zimbabwe - epidemiology
• ISSN: 1368-4973; 1472-3263
• DOI: 10.1136/sti.75.6.426

OBJECTIVE: To determine the reliability of serological tests in detecting syphilis in a factory worker cohort and examine the impact of concurrent HIV infection on serological tests for syphilis. METHOD: Reactions to non-treponemal and treponemal antigens were tested using sera from a cohort of 3401 factory workers in Harare, Zimbabwe. The participants consented to regular testing for syphilis, by VDRL, and HIV using two ELISAs. All sera from men who were VDRL positive, and a random sample of VDRL negative sera, were tested by RPR, TPHA, and where appropriate FTA-Abs. From the results, men were defined as having no syphilis, active syphilis, incident syphilis, historic syphilis, or giving biological false positive reactions. RESULTS: 709 sera were examined from 580 men. There were 78 cases of active syphilis in the cohort, giving a prevalence of 2.3%, and the seroincidence was 0.25 per 100 person years of follow up. The prevalence of HIV in the cohort was 19.8%. There was a strong association between syphilis, whether active, incident or historic, and HIV seropositivity. With both HIV positive and negative sera the negative predictive values of VDRL and RPR were > 99.9% while the positive predictive value for VDRL (30%) was lower than for RPR (39%). Biological false positive reactions were detected in 0.5% of the cohort, with in most cases a transient rise in VDRL titres up to < 1/16. Higher false positive titres occurred in five men, each of whom was HIV positive. CONCLUSIONS: The VDRL is reliable in detecting possible cases of syphilis even in a community with a high prevalence of heterosexually transmitted HIV. There is need, however, for confirmatory tests. The prevalence of syphilis in this cohort is very low in comparison with other countries in southern Africa, but is consistent with recent data from Harare. Despite a strong association between syphilis and HIV it was clear that syphilis could not be counted as a major factor fueling the HIV epidemic in Zimbabwe.