Comparison of four commercial screening assays for the diagnosis of human T-cell lymphotropic virus types 1 and 2.

Serological assays for human T-cell lymphotropic virus types 1 and 2 (HTLV-1/2) are widely used in routine screening of blood donors. The aim of this study was to compare the performance of four commercial screening assays for HTLV-1/2 infection frequently used in South America. A total of 142 HTLV-1 and HTLV-2 seropositive and 336 seronegative samples were analyzed by using four commercial tests (BioKit, Vironostika, Murex and Fujirebio). These tests are commonly used for HTLV-1/2 detection in blood banks in Argentina. A nested-PCR was used as the reference standard. The most sensitive tests for HTLV-1/2 were Fujirebio and Biokit (98.6%) followed by Murex (97.2%) and Vironostika (96.5%). The most specific test was Murex (99.7%), followed by Biokit (97.0%), Fujirebio (95.8%), and Vironostika (92.9%). The kappa index of agreement was higher for Murex (kappa=0.97), followed by BioKit (kappa=0.94), Fujirebio (kappa=0.92), and Vironostika (kappa=0.86). The highest index of agreement was shown by Murex test while Vironostika had the lowest performance. Of the four tests evaluated, only the Vironostika assay is approved by the Food and Drug Administration. These results should be considered for choosing the most accurate serological screening assays in order to obtain an optimal efficiency of the current algorithm for HTLV-1/2 diagnosis.