Parallel enzyme-linked immunosorbent assay screening for human immunodeficiency virus among blood donors in five Chinese blood centres: a retrospective analysis.

Abstract

To evaluate the strategy of parallel screening with different enzyme-linked immunosorbent assays (ELISA) among Chinese blood donors. Parallel screening with ELISA has been the main strategy to detect human immunodeficiency virus (HIV) in blood donations in China for more than a decade. The performance of the strategy should be analysed. A total of 821,927 donations collected from five Chinese blood centres in 2008-2010 were tested using two third-generation ELISAs by different manufacturers licenced and confirmed by the Western blot (WB) in this study. The confirmatory positive predictive values (PPV), false positive rates (FPR), false negative rates (FNR) and potential risks for transfusion resulting from single or sequential ELISA screening were evaluated. A total of 5318 (0·647%) of donations screened HIV reactive and were discarded. WB confirmatory results on 1668 available samples suggested that PPVs for dual ELISA, one round ELISA reactive and grey zone samples were 75·1, 0·7 and 0·5%, respectively. Eight out of 1124 one round ELISA reactive and 1 out of 195 grey zone samples were WB confirmed positive. All but one ELISA assay displayed comparable PPVs but variable FPRs and FNRs that differed by blood centre. In the absence of nucleic acid testing (NAT), parallel ELISA screening prevented a substantial number of HIV infected donations from entering the Chinese blood supply. However, the loss of false positive donors should be re-evaluated especially given the frequently reported blood supply shortage in China.