A Comparison of Two HIV Antibody ELISAs with a Combined Antigen-Antibody ELISA for the Occurrence of False Results in a Hospital-Based Blood Bank In north-Central Nigeria: Implications on Blood Safety and Availability

Abstract

Background: Global targets of providing safe blood available universally by 2020 as well as that seeking to eliminate HIV transmission by 2030 has compelled many hospital-based blood banks to employ relevant strategies in their transfusion practice towards realizing these targets. Aim: To assess two HIV antibody ELISA tests (Determine™ HIV- 1/2 and Dia Pro HIV 1/2/0 ELISA) deployed for blood safety and availability amongst blood donors at a hospital-based blood bank in north-central Nigeria for the occurrence of false test results. Materials and Methods: This cross-sectional study was carried out at the hospital-based blood bank of Jos University Teaching Hospital (JUTH) between May and August 2008. The sera of four hundred and forty blood donors were serially tested with two HIV antibody ELISAs Determine™ HIV- 1/2 and Dia Pro HIV 1/2/0 ELISA and then comparatively tested with Genscreen ®PLUS HIV Ag- Ab ELISA (a combined HIV antigen-antibody ELISA. The proportion of false negative (FN), false positive (FP), negative predictive value (NPV) and positive predictive value (PPV) of test results were determined using prism pad statistical package version 5 with p<0.05 taken as the level of statistical significance. Results: False negative (FN) tests results were recorded as 60.98% and 46.34% for Determine TM HIV 1/2/0 and Dia Pro HIV 1/2/0 respectively related to the sensitivities (39.02% and 53.65%) and specificities (100% and 99.50%) of Determine TM HIV 1/2/0 and Dia Pro HIV 1/2/0 respectively. The proportion of FP test results were found to be 0 and 0.50 percent for Determine TM HIV 1/2/0 and Dia Pro HIV 1/2/0. The positive predictive value (PPV) was 100% and 91.67% while the negative predictive value (NPV) was 94.1% and 95.4% for Determine TM HIV 1/2/0 and Dia Pro HIV 1/2/0 respectively. Conclusion: The high FN and low NPV obtained with the two HIV antibody ELISAs suggest their unsafety if deployed for TTI screening in our blood bank. The low FP and relatively higher PPV, on the other hand, suggest low deferral rate for Determine TM HIV 1/2/0 but not necessarily so for Dia Pro HIV 1/2/0. The contentious deployment of these antibody tests would compromise initiatives necessary in developing blood safety and availability in order to achieve the global targets. Recommendation: Hospital-based blood banks in our setting should be supported to deploy only the combined HIV antigen-antibody ELISA for all blood donors and donations in order to ensure blood safety and availability pending the implementation of nucleic acid amplification testing (NAAT).