Performance evaluation of eight rapid tests to detect HIV infection: A comparative study from Brazil.

Abstract

Rapid tests (RTs), also known as point-of-care tests, usually release results within 30 minutes with no need for a qualified staff, equipment, or laboratory structure. The Brazilian Ministry of Health published a resolution in 2013, recommending the use of RTs for the diagnosis of HIV infection, where one positive RT must be followed by another different RT. This was meant to increase the chance of proper diagnosis in specific settings and special populations. However, data comparing and validating the different HIV RTs available in Brazil are scarce. Therefore, the present study seeks to evaluate eight anti-HIV RTs available in the Brazilian market regarding their analytical performance: sensitivity, specificity, positive and negative predictive values, positive and negative likelihood ratios, and accuracy. We also evaluated the agreement between kits (Kappa index) and the quality of the reading pattern of the tests. This was an observational, analytical, and concordance study, in which previously defined positive and negative samples, based on their serological pattern for anti-HIV antibodies (chemiluminescent immunoassay-ECLIA-used as screening and Western Blot used as the confirmatory test) were tested. Analytical performance and Kappa index were calculated, considering a 95% CI and p<0.05. This study identified differences in the performances of the eight tested kits. Six out of eight RTs showed good performance and can be used in the routine laboratory and health care units as screening tests. Regarding the quality of the RT band reading pattern, two brands had several samples showing quite faint bands, thus compromising its use in clinical and laboratory settings.