O15.4 Field Performance of the Alere Determine HIV Combo Assay in a Large Australian Multi-Centre Study in a Sexual Health Clinic Setting. (13th July 2013)

• Record Type: Journal Article
• Authors: Conway, D P; Holt, M; McNulty, A; Couldwell, D L; Smith, D E; Davies, S C; Cunningham, P; Keen, P; Guy, R
• Is Part Of: Sexually transmitted infections. Volume 89(2013)Supplement 1
• Page Start: A55
• Page End: A56
• Publication Date: 2013-07-13
• Subjects: Field performance -- Rapid HIV testing -- Sexual health clinic; Sexually transmitted diseases -- Periodicals; HIV infections -- Periodicals; 616.951005
• DOI: 10.1136/sextrans-2013-051184.0170
• Languages: English
• ISSNs: 1368-4973

Abstract

Abstract : Background: Rapid HIV testing has been available to gay and other men who have sex with men (MSM) overseas for many years. Alere's Determine HIV Combo 'fourth generation' rapid test containing antibody and antigen components is now available in Australia, but field performance data for this assay are limited. Methods: From September 2011, MSM attending four Sydney public sexual health clinics were offered rapid HIV testing using the Determine HIV Combo and also had sexually transmissible infection screening and conventional HIV serology. Rapid test sensitivity, specificity, and positive and negative predictive values (PPV, NPV) were calculated by comparing results to reference tests (Abbott Architect HIV Ag/Ab Combo, Biorad Genscreen HIV antigen and HIV Western blot). Results: In 15 months, 1716 men had 2043 rapid tests performed with four invalid rapid tests (0.2%) excluded from analysis. Of 34 men confirmed as HIV-positive by national HIV case definitions, 29 had reactive rapid tests (sensitivity = 85.3%, 95% CI: 68.2–94.5). With 29 true reactive rapid tests from a total of 44 reactive tests, PPV overall was 65.9% (50.0–79.1). Of five men with false non-reactive tests, four were seroconvertors. Rapid tests were non-reactive in 1990 out of 2005 cases where laboratory HIV testing was negative; hence, overall specificity was 99.3% (95% CI: 98.7–99.6) and NPV was 99.8% (99.4–99.9). Of 15 men with false reactive rapid tests, four had non-specific reactivity on HIV Western blot. Specificity and NPV for the antibody component was 99.5% (99.0–99.7) and 99.9% (99.6–100.0) and for the antigen component was 99.8% (99.4–99.9) and 99.6% (99.1–99.8), respectively. Conclusion: Antibody and antigen component specificity was consistent with the rapid test package insert; whereas sensitivity was lower, notably in those with recent infections. Hence, identifying patients at risk of recent infection is vital so that conventional laboratory serology is performed. A formal assessment of test performance in seroconvertors is warranted.