Effects of leukoreduction and storage duration on whole blood hemostatic function: a prospective ex vivo observational study

Background Whole blood (WB) resuscitation is a critical component of trauma care. Leukoreduction (LR) is commonly employed to mitigate transfusion-related risks. We aimed to understand the effects of LR on the functionality and metabolism of various components in refrigerated WB and provide new strategies for WB application in trauma treatment. Methods Ten bags of WB (400 mL ± 10%) were split into paired samples with and without LR and stored at 4°C ± 2 °C for 35 days. Storage-lesion tests were performed every 7 days from the day of blood collection. Descriptive heatmaps and weighted gene co-expression network analysis (WGCNA) were used to characterize global patterns of indicator change, while exploratory mediation models and penalized regression (LASSO) were applied to investigate pathways and rank variables associated with clot strength. Results Descriptive heatmaps and exploratory WGCNA analyses suggested two distinct temporal patterns: platelet-related and viscoelastic indicators decreased progressively, whereas metabolic, hemolytic, and electrolyte indicators increased over time. On Day 0 (immediately after leukoreduction), leukoreduced whole blood (LR-WB) had a markedly lower platelet count than paired nonleukoreduced whole blood group (NLR-WB) (23.40 ± 2.00 vs. 185.20 ± 2.93 *10 9 /L; p < 0.05), accompanied by reduced clot strength and impaired kinetics (MA 33.6 ± 9.82 vs. 61.09 ± 2.97 mm; K 5.20 ± 1.37 vs. 1.70 ± 0.35 min; Angle 45.83 ± 8.82 vs. 65.68° ± 4.33°; all p < 0.05). These between-group differences persisted across storage. Fibrinogen concentration remained comparable between two groups, but kinetic indicators (K, Angle) were consistently lower in LR-WB. In exploratory mediation models, platelet count, mean platelet volume, K, and Angle accounted for most of the model-based associations of LR and storage duration with MA. Penalized regression (LASSO) further highlighted blood cells’ (RBCs’) phosphatidylserine (PS) exposure and hemoglobin (FHb) as late-storage features associated with lower MA within this experimental dataset. Conclusion LR was associated with early platelet-driven hemostatic impairment, and storage time appeared to progressively shift injury mechanisms toward RBC membrane instability, hemolysis, and metabolic stress. RBCs’ PS exposure and FHb can serve as candidate indicators associated with clot strength.