Characterization of Packed Red Blood Cells Prepared with Experimental Multifunctional Leukocyte and Antibody-Reduction Filters

Abstract 2320

Transfusion-related acute lung injury (TRALI) remains a leading cause of transfusion-related death. With a decrease in antibody-mediated, plasma-induced TRALI, packed red blood cells (PRBCs) have become implicated more often. We have developed 3 experimental filters that concomitantly remove leukocytes, platelets, antibodies (abs), and lipids from PRBCs. We hypothesize that these filters remove antibodies against HLA class I and class II antigens and lipids without affecting the quality of stored PRBCs. Methods: 31 female donors whose plasma was historically positive for HLA by both Luminex™ beads (flow cytometry) and ELISA (GTI) testing were recruited and 4 male controls per filter were also drawn. Whole blood was drawn, PRBCs separated, then filtered with one of the experimental filters (B (n=11), P (n=10), or M (n=10)) or the FDA-approved Pall BPF4. Samples were drawn prior to filtration (Pre) or immediately after (Post) and at days 21 and 42 of storage. IgG, IgM, and HLA abs (by both flow cytometry and ELISA) were measured, plus a CBC was performed, pre-& post-filtration. Adenosine triphosphate (ATP), 2,3-diphoshoglycerate (2,3-DPG), pH, and neutrophil (PMN) priming activity were performed on all samples with the latter completed as the augmentation of the maximal rate of fMLP-activated PMNs measured as the superoxide dismutase inhibitable reduction of cytochrome c (nmol O₂⁻/min). Statistical differences were determined by independent or paired (priming) ANOVA, and the data is expressed as the mean±SEM with *=p<.05 vs. controls or day 0 pre-plasma (Table). Results: All filters demonstrated similar reduction of leukocytes and platelets identical to the BPF4 control. At recruitment, only 10/31 females still had plasma antibodies to HLA antigens, with 4 having abs to both, and all were removed with the experimental filters: The ATP and pH levels were all within acceptable limits. The 2,3-DPG levels in the BPF4 filter were not significantly different from the experimental filters at day 42. As compared to the BPF4 units, the priming activity in all of the experimental filters on day 21 and day 42 was significantly reduced (^†=p<.05). We conclude that the experimental filters significantly reduced leukocyte and platelet contamination, depleted IgG and HLA antibodies, and decreased the amount of lipid priming activity. The use of these filters for RBC manufacture may decrease both immune and non-immune TRALI.