Accumulating evidence suggests that transfusion of red blood cells (RBCs) stored for longer durations is associated with increased rates of infection, morbidity, and mortality in hospitalized patients. During refrigerated storage in vitro, RBCs undergo cumulative biochemical and biomechanical changes that reduce their survival in vivo. After transfusion, storage-damaged RBCs are rapidly cleared from the circulation by reticuloendothelial macrophages, usually within the first hour. The associated hemoglobin iron is then rapidly catabolized and returned to the plasma at a pace that can exceed the rate of uptake by the physiologic iron transporter, transferrin, thereby producing plasma non-transferrin-bound iron. Our overarching hypothesis is that the rapid clearance of storage-damaged RBCs is responsible for some of the adverse effects of transfusion, specifically by delivering a large bolus of toxic hemoglobin-derived iron to macrophages. In particular, plasma non-transferrin-bound iron can produce oxidative damage to lipids, proteins, and DNA, thereby activating endothelial cells and platelets to increase thrombotic risk. Furthermore, increased macrophage iron and plasma non-transferrin-bound iron can enhance infectious risk by providing sources of iron to microbial pathogens both within macrophages and within the systemic circulation. Our studies in healthy human volunteers demonstrate that transfusions of “older” RBCs (stored for 40–42 days), but not “fresh” RBCs (stored for less than seven days), significantly increase circulating non-transferrin-bound iron levels in vivo (1) and enhance bacterial growth in vitro in posttransfusion serum samples. Animal studies (2) confirm these findings and provide additional evidence that transfusion exacerbates infection in vivo with Salmonella typhimurium, Escherichia coli, and Staphylococcus aureus. In addition, murine studies suggest that rapid transfusion of the equivalent of two units of older, stored RBCs produces a proinflammatory cytokine response, which was not seen after slow transfusion of one unit of older, stored RBCs to healthy human volunteers. Studies in dogs are currently under way to test whether the RBC transfusion rate and dose affects the proinflammatory response. In conclusion, transfusion of older, stored RBCs results in delivery of a potentially toxic dose of iron to macrophages that both increases macrophage intracellular iron and produces plasma non-transferrin-bound iron. This iron toxicity may increase infectious and thrombotic risks and potentially exacerbate an existing proinflammatory response.
No relevant conflicts of interest to declare.