Extracorporeal membrane oxygenation (ECMO) is a potentially life-saving intervention for children and adults with severe cardiac or respiratory failure. [1][2][3] The use of veno-venous (VV) and veno-arterial (VA) ECMO as a bridge to cardiopulmonary recovery or to heart or lung transplantation has dramatically increased in the last decade. 4 In this issue of TRANSFUSION, Bilodeau et al. seek to characterize and quantify the various causes of red blood cell (RBC) loss in pediatric patients on ECMO. 5 The authors propose the use of the RBC loss index (RLI), a novel metric based upon changes in hematocrit and the volume of transfused RBCs, calculated as mL of RBCs lost per liter of total (patient + circuit) volume per hour. They also propose using RBC extracellular vesicle (REV) generation as a marker of sublethal injury to RBCs during ECMO. The investigators suggest that sublethal RBC damage may stimulate extravascular clearance and eventual RBC loss. We applaud the authors on their effort to provide objective metrics for blood loss as well as RBC injury in children on ECMO, and believe that this article should be of great interest to the readers of TRANSFUSION.Transfusion of blood products, such as RBCs, plasma, platelets, and cryoprecipitate, while on ECMO occurs for a variety of reasons. Anemia is common with ECMO and typically attributed to a combination of circuit-related hemolysis, frequent blood sampling for laboratory monitoring, and an increased risk of bleeding given the use of anticoagulants, consumptive coagulopathy, and thrombocytopenia/platelet dysfunction. 6,7 While bleeding remains a principal cause of anemia in patients on ECMO, the authors attempt to quantify the contribution of other etiologies. They note that in the "non-bleeding" ECMO patients, diagnostic phlebotomy accounted for almost a quarter of blood loss, followed by intravascular hemolysis at 16%. This calculation suggested that a majority (up to 60%) of RBC loss in patients without evidence of bleeding was due to other etiologies. Studies have shown that patients supported with ECMO may inadvertently be exposed to RBC transfusion for the treatment of otherwise mild anemia resulting from blood loss, particularly phlebotomy. Frequent phlebotomy and laboratory sampling can contribute to transfusion requirements in more than 50% of patients and have been the sole reason for at least one transfusion in 40% of patients. 7