S S Myers1, R Heimer, D Liu, D Henrard. 1. Department of Internal Medicine, Yale University School of Medicine, New Haven, CT 06510.
Abstract
OBJECTIVE: Direct HIV testing of individual injecting drug users is not always feasible. As an alternative, we have evaluated the sensitivity and specificity of several techniques for detecting HIV-1-specific products in used syringes. DESIGN: Polymerase chain reaction (PCR) and antibody-capture assays were compared using syringes prepared with blood from HIV-1-positive and -negative individuals. METHODS: PCR sensitivity was maximized, enabling detection of single copies of HIV-1-specific proviral DNA. The limits of detection from used syringes were determined for PCR by diluting extracts and correlated to CD4+ cell counts. Similarly, limits of detection were determined for enzyme immunoassays (EIA) and Western blot. RESULTS: All techniques were highly specific, although with PCR false-positives were detected occasionally. EIA proved more sensitive than Western blot in detecting needles containing HIV-1-infected individuals' blood. Even after prolonged storage of syringes at room temperature, EIA was equal to or better than PCR as an HIV-1 detection technique. The most sensitive method for detecting HIV-1 was the viral-based EIA when the recommended predilution step was omitted. CONCLUSIONS: EIA proved preferable to PCR because of their higher sensitivity, absence of false-positives and easier sample preparation and analysis.
OBJECTIVE: Direct HIV testing of individual injecting drug users is not always feasible. As an alternative, we have evaluated the sensitivity and specificity of several techniques for detecting HIV-1-specific products in used syringes. DESIGN: Polymerase chain reaction (PCR) and antibody-capture assays were compared using syringes prepared with blood from HIV-1-positive and -negative individuals. METHODS: PCR sensitivity was maximized, enabling detection of single copies of HIV-1-specific proviral DNA. The limits of detection from used syringes were determined for PCR by diluting extracts and correlated to CD4+ cell counts. Similarly, limits of detection were determined for enzyme immunoassays (EIA) and Western blot. RESULTS: All techniques were highly specific, although with PCR false-positives were detected occasionally. EIA proved more sensitive than Western blot in detecting needles containing HIV-1-infected individuals' blood. Even after prolonged storage of syringes at room temperature, EIA was equal to or better than PCR as an HIV-1 detection technique. The most sensitive method for detecting HIV-1 was the viral-based EIA when the recommended predilution step was omitted. CONCLUSIONS: EIA proved preferable to PCR because of their higher sensitivity, absence of false-positives and easier sample preparation and analysis.