Abstract:
BACKGROUND AND OBJECTIVES:
Ensuring the safety of blood supply relies on robust screening and donor selection processes. This study assessed the technical performance of automated enzyme-linked immunosorbent assay (ELISA) (EVOLIS, Bio-Rad, France) and electrochemiluminescence immunoassay (ECLIA) (COBAS 6000 e601 and COBAS e411, Roche Diagnostics, Germany) in detecting hepatitis B virus (HBV), hepatitis C virus (HCV), and human immunodeficiency virus (HIV) infections.
MATERIALS AND METHODS
Retrospective data of transfusion transmissible infections (TTI) screening test results were collected from blood center records, encompassing two phases: January 2020 to May 2020 (phase 1) and March 2021 to April 2021 (phase 2). The sensitivity, specificity, positive predictive value, and negative predictive value of ELISA and ECLIA were calculated. Performance evaluation was conducted using the Youden’s index and the kappa coefficient.
RESULTS:
The study included a total of 824, 755, and 830 donor samples in phase 1 and 1237, 1238, and 1236 donor samples in phase 2 for HIV, HBV, and HCV testing, respectively. High agreement (kappa 0.88, 0.95) existed between ELISA and COBAS e601 for HIV and HBV, with substantial agreement (kappa 0.66) for HCV. Likewise, substantial agreement (kappa 0.66, 0.79, 0.79) was found between ELISA and COBAS e411 for all three parameters HIV, HBV and HCV, respectively. Notably, nearly all false-positive samples detected by ECLIA had low cutoff index values ranging from 1 to 6.5.
CONCLUSION:
The differences in sensitivity and specificity contribute to the discrepancy observed between various serological testing platforms, such as ELISA and ECLIA, for TTI testing. Specifically, low signal/cutoff ratios ≤6.5 obtained with ECLIA are highly indicative of false-positive results.