Retrovirus infections in a sample of injecting drug users in Rio de Janeiro City, Brazil: prevalence of HIV-1 subtypes, and co-infection with HTLV-I/II

Abstract

Retrovirus infections among injecting drug users (IDUs), a core at-risk population for both HIV-1 and HTLV-I/II infections in Brazil, were assessed within an ongoing cooperative research. The study assessed the seroprevalences of HIV-1 and HTLV-I/II infections, as well as the prevalence of HIV-1 subtypes in a sample of IDUs from Rio de Janeiro, Brazil. An attempt to evaluate HIV incidence was carried out using a dual 'sensitive/less sensitive' testing strategy. Cross-sectional evaluation of 175 IDUs. Serostatus for HIV-1 and HTLV-I/II were established by enzyme-linked immunosorbent assays, and confirmed by western blot. The dual testing strategy aimed to estimate HIV-1 incidence rates. Differentiation between HTLV-I and -II was performed by western blot. DNA samples were polymerase chain reaction amplified by a nested protocol, and HIV-1 subtyping was determined by heteroduplex mobility assay. Forty-six and 29 samples were found to be, respectively, positive for HIV-1 and HTLV-I/II, 15 of them co-infected by both viruses. Among HTLV-I/II-infected patients, 75.9% were infected by HTLV-I. Thirty-one HIV samples were identified as B subtype, with seven of them showing the typical "Brazilian B" pattern in the gp120 V3 loop, and ten were identified as F subtype. The use of less sensitive assays for HIV infection wrongly identified a deeply immunocompromised patient as an incident case. Moderately high seroprevalences were found for both HIV-1 and HTLV-I/II infections, HIV-1/HTLV-I co-infections being of special concern. A non-statistically significant higher prevalence of F subtype was observed, when compared with the distribution of F/B subtypes among Brazilian patients from other exposure categories. No recent HIV-1 infections were detected, but a limitation of the "sensitive/less-sensitive" testing strategy was made evident.