Résumé
Guidelines for optimized HCV screening are urgently required in Africa, especially for patients infected with HIV, who sometimes show false positive or false negative reactivity in anti-HCV antibody assays. Here, we assessed the usefulness of a fourth-generation HCV Ag-Ab ELISA for the identification of active HCV infection in HIV-positive patients.
This cross-sectional study was conducted between 03/2010 and 01/2013 and included 762 Gabonese HIV-positive adult patients. The results of ELISA (Monolisa HCV Ag-Ab ULTRA, Bio-Rad) were compared with those obtained by RT-PCR (gold standard). The optimal ELISA signal-to-cutoff (S/CO) ratio to identify patients with active hepatitis C (positive HCV RNA) was determined. Specimens were further tested by the INNO-LIA HCV Score assay (Innogenetics) and the Architect HCV Ag kit (Abbott) to define the best diagnostic strategy.
Sixty-seven patients tested positive for HCV (S/CO ratio ≥ 1) by ELISA. Of these, 47 (70.1%) tested positive for HCV RNA. The optimal S/CO associated with active HCV infection was 1.7. At this threshold, the sensitivity of ELISA was 97.9% (95% confidence interval (CI) 90.0-99.9%), its specificity was 91.3% (95% CI 85.0-95.5%), and HCV seroprevalence rate was 7.3% (56/762) (95% CI 5.6-9.4%). Among 57 HCV-seropositive patients with available INNO-LIA results, false reactivity was identified in 14 (24.6%), resolved HCV infection in two (3.5%), possible acute HCV infections in nine (15.8%) and likely chronic HCV infections in 32 (56.1%) patients. HCV core Ag was undetectable in 14/15 (93.3%) specimens that tested negative for HCV RNA whereas it was quantified in 34 (out of 39, 87.2%) samples that tested positive for HCV RNA.
Our study provides comprehensive guidance for HCV testing in Gabon, and will help greatly clinicians to improve case definitions for both the notification and surveillance of HCV in patients co-infected with HIV.