Transfusion of blood products during extracorporeal membrane oxygenation: a narrative review of rationale, indications, impact on immune function and outcome

The use of extracorporeal membrane oxygenation (ECMO) support poses several risks, particularly thrombosis and bleeding. As a result, transfusion of blood components is frequent during extracorporeal support. In this review we aim to describe the rationale and indications of blood products transfusions, and their impact on the immune function and outcome. The red blood cells (RBC) transfusion threshold is very debated, because of awareness of transfusion-associated adverse events due to liberal strategies. To date, no specific recommendations exist but a comprehensive physiologic approach appears feasible to evaluate the need for RBC transfusion. For patients without bleeding, the guidelines of the Extracorporeal Life Support Organization (ELSO) suggest fresh frozen plasma (FFP) administration if the prothrombin time (PT) ratio is higher than 1.5-2.0 and/or there is significant bleeding. Conversely, for bleeding patient indications often refer to trauma guidelines, where it is recommended to use a 1:1 ratio of RBC and FFP in massive transfusion situations. The indications for antithrombin supplementation are unknown and large inhomogeneity exists between different ECMO centers and between pediatric and adult patients. Supplementation of fibrinogen is considered only for bleeding patients and/or with fibrinogen level below 100 or 150 mg/dL. ELSO guidelines suggest 25-50 IU/kg of prothrombin complex concentrate as an alternative to FFP for patients with active bleeding and a prolonged PT. Recombinant activated factor VII might be a potential therapeutic option for intractable bleeding despite conventional treatment but may cause life-threatening thrombotic complications. Platelet transfusions might be limited to cases of severe thrombocytopenia accompanied by bleeding. ELSO guidelines recommend a target of at least 80x109/L platelets. Liberal platelets transfusion thresholds may be reasonable in case of intracranial hemorrhage. Albeit rare, multiple adverse events of blood products transfusion are described. There is no evidence of transfusion-related acute lung injury during ECMO support, likely because of the difficulty to distinguish the cause of clinical worsening in patients with severe respiratory failure. Infections represent a major contributor on morbidity and mortality in ECMO patients. However, as of today, no literature has explored the impact of transfusions on immune function of ECMO patients. Currently, there are no specific guidelines for transfusions in ECMO patients and the management is highly variable among centers. Further research is warranted on this topic.