Evaluation of a modified 2-step diagnostic algorithm for human immunodeficiency virus type 1 utilizing an HIV-1/2...

A multi-test algorithm is routinely used for HIV-1 diagnosis. Guidelines recommend an HIV-1/2 antigen/antibody (Ag/Ab) immunoassay screen that reflexes reactive samples to an HIV-1/2 antibody differentiation immunoassay for confirmation. If negative or indeterminate antibody differentiation results are observed, an HIV nucleic acid amplification test (NAAT) is used to determine HIV status. An alternative algorithm in which screen reactive samples reflex to HIV NAAT has been proposed. The objective of this study was to compare the accuracy of the standard versus alternative algorithm for final HIV-1 interpretation.

A total of 392 remnant HIV-1 samples from routine testing at two Quest laboratories were tested using the cobas® HIV-1/HIV-2 Qualitative NAAT. Laboratory algorithm interpretations and TAT were evaluated between the two algorithms.

The positive percent agreement (PPA) between the two algorithms was 95.3% (181/190) and negative percent agreement (NPA) was 100% (202/202). Discordant results were identified as HIV-1 positive by the current algorithm but negative by the alternative algorithm. Previously diagnosed positive patients undergoing treatment or re-testing could not be ruled out. Workflow analysis showed that the there is less hands-on time per sample for the HIV NAAT compared to the differentiation immunoassay, which may result in faster TAT depending on testing volumes.

The data show the alternative algorithm is acceptable for HIV laboratory diagnosis in terms of analytical performance with minimal impact to TAT at our institution.