Linking Stored Red Blood Cell Metabolism to Transfusion Recipient Iron Homeostasis Pathophysiology in Critically-Ill Children

Background: Transfusion of packed red blood cells (pRBCs) is a common, life-saving therapy for many critically-ill children. However, routine blood bank storage results in the progressive accumulation of a series of RBC modifications, termed the "storage lesion," that reduce RBC recovery in vivo, may affect transfusion recipient physiology, and may be implicated in transfusion-related adverse events. In particular, circulating non-transferrin-bound iron (NTBI) in recipients can be increased after pRBC transfusion. NTBI promotes bacterial growth, induces oxidative injury, and inversely correlates with post-transfusion recovery, all of which may increase the risk of sepsis, multiorgan failure, and death. We aimed to determine whether the metabolic state of donor pRBCs at the time of transfusion correlates with recipient NTBI levels, which serves as a marker of hemolysis and iron homeostasis in critically-ill pediatric patients.

Study Design/Methods: Critically-ill pediatric patients, aged < 21-years-old with a minimum weight of 5kg in the pediatric intensive care unit at a single-site tertiary care hospital, who required pRBC transfusion, were prospectively enrolled after parental consent (n = 49 patients). Patient blood samples were obtained pre-transfusion and 4 hours following the patient's first qualifying pRBC transfusion, and analyzed for markers of iron and heme homeostasis, hemolysis, liver function, and inflammation utilizing standard laboratory methods. pRBC transfusate (1 mL) was frozen at -80°C immediately at the end of transfusion for each transfused patient. Hydrophilic and lipophilic compounds were extracted separately using methanol:acetonitrile:water or methanol, respectively, before analysis using high-throughput LC-MS based metabolomics and lipidomics platforms to determine the relative abundance of small molecule compounds. Metabolite levels in the pRBC transfusates were correlated with the differences in pre- and post-transfusion clinical laboratory values and selected patient parameters (e.g. blood type, patient age, and storage age of the pRBC unit at the time of transfusion), using Pearson correlation coefficients and associated p-values.

Results: The relative levels of 221 unique metabolites and lipids in the pRBC transfusates were correlated with the corresponding transfusion recipient's clinical laboratory values and selected patient parameters. Analysis of Pearson's correlations between the levels of these compounds in the donor pRBC transfusates and the pre- and post-transfusion differences in the recipients' NTBI levels (ΔNTBI) demonstrated statistically significant correlations for 24 metabolites (R = 0.28-0.43, p = 0.0022-0.0476, see Table). Analysis of recipient clinical laboratory biomarkers associated with recipient ΔNTBI revealed a positive correlation with increases in serum iron (R = 0.41, p = 0.0034) but not with other changes in recipient clinical laboratory biomarkers. The age of the stored pRBCs at the time of transfusion did not correlate with ΔNTBI levels (R = 0.24, p = 0.0912).

Conclusions: Statistically significant correlations of donor pRBC transfusate metabolites with transfusion recipient ΔNTBI are primarily seen with hypoxanthine, lactate, metabolites of glutathione homeostasis, amino acids, unsaturated free fatty acids, acyl-carnitines, and metabolites of indole and tryptophan metabolism. These data suggest that higher levels of oxidative stress, proteolysis, membrane remodeling in stored donor pRBCs, and, possibly, microbiome metabolism in the pRBC donor, are associated with significant increases in recipient plasma NTBI and serum iron levels, likely from hemolysis of storage-damaged RBCs. The levels of these metabolic biomarkers of the RBC storage lesion correlate better with ΔNTBI than the storage age of the donor unit, supporting the notion that pRBC metabolic age is not equivalent to pRBC chronologic age, and that the former may be more relevant to transfusion recipient pathophysiology in critically-ill children.

Table.

View largeDownload slide

Table.

View largeDownload slide

Close modal