Abstract
BACKGROUND: Geenius HIV 1/2 Supplemental Assay (Geenius; Bio-Rad Laboratories) is the only Food and Drug Administration-approved HIV-1/HIV-2 antibody differentiation test for the second step in the HIV laboratory testing algorithm. We characterized the occurrence of true HIV-1 and HIV-2 infections as well as false results in 6 US clinical laboratories using Geenius. METHODS: We examined routine HIV testing outcome data from the time the laboratories began using the algorithm with Geenius until September 30, 2017. We calculated the positive predictive value for Geenius HIV-1 and HIV-2 reactivity separately. RESULTS: Of 5,046,684 specimens tested, 41,791 had reactive antigen/antibody test results. Most specimens with reactive antigen/antibody results were HIV-1 antibody-positive established infections (n = 32,421), 1,865 of which also had indeterminate HIV-2 bands present. Ninety-three specimens were HIV-2 antibody positive or untypable for HIV-1/HIV-2 antibody. Acute HIV-1 infections were found in 528 specimens; 881 specimens lacked the nucleic acid test to determine the possibility of acute HIV-1 infection. False-positive antigen/antibody test results were present in 7505 specimens. Few specimens (n = 363) had false-positive antigen/antibody results with indeterminate Geenius and negative HIV-1 nucleic acid test results. The positive predictive values of Geenius reactivity were 99.4% for HIV-1 and 4.3% for HIV-2. CONCLUSIONS: Routine testing using the laboratory testing algorithm with Geenius resulted in most specimens resolving as HIV negative or HIV-1 positive. The occurrence of indeterminate HIV-2 bands with a Geenius final assay interpretation of HIV-1 positive was more common than true HIV-2 infections. Reporting indeterminate HIV-2 results in this situation may cause confusion with interpreting HIV infection status.