Background: Serological diagnosis of HIV infection is traditionally accomplished with a screening Enzyme Immunoassay (EIA) followed by confirmatory Western blot testing. A recent national shortage of FDA approved Western blots demonstrated the clear need for alternative confirmatory testing strategies.

Methods: We screened 577,618 samples drawn during the period of 1 January 2001 through 31 December 2001, from Department of Defense (D.O.D.) healthcare beneficiaries for HIV-1 antibody by EIA. Of these samples, 1,371 were initially reactive by screening EIA and 477 were repeatedly reactive by a second EIA platform necessitating confirmatory Western blot testing. Out of the initial 477, 467, and 425 samples had sufficient volume to be evaluated with Orasure saliva Western Blot (utilizing a serum modification) and the Procleix nucleic acid test (NAT), respectively. All NAT negative, Western blot reactive specimens were subjected to testing with the Roche Amplicor HIV-1 Monitor v1.0 assay to compare NAT results.

Results: Analysis of the oral fluid western blot for year 2001 U.S. D.O.D serum samples showed a 95.7% concordance with the reference FDA approved serum western blot. All 20 discordant samples involved non-reactive/indeterminate result pairs. This yielded a sensitivity and specificity of 100% for the oral fluid blot. Further studies with the oral fluid western blot involving commercial serocoversion panels, HIV-1 vaccine recipients and a commercial HIV-1 subtype panel, which contained 30 geographically diverse samples, yielded a sensitivity and specificity of 98.5% (single false negative on the earliest serum from a seroconversion panel) and 100%, respectively. Comparative analysis of NAT with traditional serological HIV-1 diagnostic tests for 425 samples showed a 90.4% sensitivity and specificity of 90.4% and 98.2%, respectively. All discordant samples (NAT non-reactive, WB reactive samples) were subjected to testing with the Roche Amplicor HIV Monitor v1.0 assay. The HIV Monitor assay was 100% concordant with the NAT assay confirming the absence of detectable nucleic acid.

Conclusions: 1) The Orasure oral fluid Western blot with serum modification is an acceptable confirmatory test for diagnosis of HIV infection; 2) NAT is an unacceptable confirmatory test for HIV infection in this study. NAT insensitivity may have been influenced by the presence of samples from individuals with known HIV infection and concurrent antiretroviral therapy.