Reliability of rapid testing for hepatitis B in a region of high HIV endemicity

Hepatitis B (HBV) and HIV co-infection is common in resource-poor settings. A recent study from Malawi revealed poor correlation between hepatitis B surface antigen (HBsAg) point-of-care tests and reference tests in patients co-infected with HIV. We studied a cohort of 300 Malawian adults entering a treatment programme for HIV. Sera were tested for HBsAg first using the Determine rapid test and re-tested using a commercial enzyme immunoassay (EIA). All tests were done under optimal conditions in Liverpool, UK. Sera from all 25 patients positive for HBsAg using the rapid test and from 50 who were negative, were re-tested using the EIA, with complete concordance of results. The kappa correlation was 1, specificity 100% (93-100%) and sensitivity 100% (86-100%) compared to the reference test. Patients had advanced immune suppression (mean CD4=175 cells x 10(6)/l). In a non-field setting, the results of point-of-care Determine rapid hepatitis B tests appear reliable in patients with HIV-1 co-infection.