Laboratory Diagnosis of HIV in the Outpatient Setting: One Hospital’s Perspective

Abstract

Abstract Traditional methods for the diagnosis of human immunodeficiency virus types 1 and 2 (HIV-1, HIV-2) involve initial screening by an immunoassay followed by a specific method Western blot. However, Western blot is not sensitive compared to third-generation immunoassay, which detects both IgG and IgM antibodies against viral envelope proteins causing false-positive results. In addition, neither initial screening nor confirmatory Western blot is capable of detecting acute infection earlier in the disease process, when the virus is more adaptable and highly infectious. To detect both acute and established infection, Abbott Architect platform introduced a new fourth-generation antigen/antibody initial screening assay in 2013. Also, to reduce false-negative results from Western blot, an alternate method Bio-Rad multisport immunoassay was recommended by the Centers for Disease Control and Prevention (CDC) that has higher sensitivity and can differentiate HIV-1 and HIV-2 infections. In 2014, the CDC released updated recommendations for the laboratory diagnosis of HIV. The CDC diagnostic algorithm recommends an initial combination immunoassay that detects HIV-1 and HIV-2 antibodies, followed by supplemental testing with an immunoassay that differentiates HIV-1 and HIV-2 antibodies. Specimens reactive on the initial antigen/antibody immunoassay but nonreactive or indeterminate on the differentiation immunoassay should be followed by nucleic acid testing (NAT) for resolution of this discrepancy. The objective of our study was to review the effect of this updated testing algorithm on HIV testing results in the outpatient setting at Banner University Medical Center–Tucson in Tucson, Arizona. Our study utilized laboratory information system queries to retroactively review all outpatient HIV laboratory testing results obtained from 2013 to 2017. A total of 17,397 HIV-1/2 antigen/antibody immunoassays were performed during this time period. Of the initial antigen/antibody immunoassays, 1.1% were reactive (n = 183). Of these reactive tests, 85% were collected from individuals with established HIV infections (n = 155). HIV-1/HIV-2 antibody differentiation assays were performed on 175 patients, with 86% reactive (n = 150), 2.3% indeterminate (n = 4), and 12% nonreactive (n = 21). Acute HIV-1 infections (antigen/antibody immunoassay reactive, antibody differentiation immunoassay nonreactive or indeterminate, and NAT positive) accounted for 2.7% of patients with reactive initial antigen/antibody immunoassays and 0.03% of all patients screened for HIV infection with the initial antigen/antibody immunoassay (n = 5). Viral loads in the patients with acute HIV-1 infection ranged from 173,000 to >10,000,000 copies/mL. No HIV-2 infections were identified in our patient population. Given that the previous CDC testing algorithm for HIV-1 failed to identify acute HIV-1 infections, our data, which confirm the ability of the current CDC algorithm to detect HIV-1 infections, represent important information for infectious disease practitioners, public health officials, and laboratory directors.