Cardiac Intensivist Washington University in St Louis CLAYTON, Missouri, United States
Abstract:
Introduction: Bleeding and thrombosis continue to be life-threatening risks for children supported on extracorporeal membrane oxygenation (ECMO), complicated by the lack of consensus regarding optimal anticoagulation management and the poor reliability of common laboratory measures of coagulation. We analyzed the courses of children supported with ECMO after undergoing cardiopulmonary bypass (CPB) to identify patient, ECMO, operative and laboratory variables that may be associated with bleeding or clotting events.
Methods: Retrospective chart review of children (0-18yrs) in a high ECMO volume cardiac center from 2018-2021. We collected patient variables (including demographics, underlying pathology, end-organ injury), ECMO variables (including mode, flow, extracorporeal cardiopulmonary resuscitation - ECPR, anticoagulant strategy), operative variables (including CPB, cross clamp, circulatory arrest and hypothermia data) and laboratory variables (including serial simultaneous labs of hemostasis). We also identified bleeding and thrombotic events as defined by Annals of Thoracic Surgery consensus statement on hemostasis on ECMO. Chi square, Mann-Whitney U test, and linear and logistic regression analyses were used to identify associations between above variables and adverse events.
Results: We identified 28 patients (32% female, 71% Caucasian) with median age of 55 days (IQR 0days–5.2years) and median ECMO duration of 123 hours (IQR 62-185). Bivalirudin was used in 75% of cases and heparin in the remaining patients. Clinical or circuit clotting events were identified in 46% of patients. Bleeding events were identified in 75% of patients in the first 48 hours of ECMO following CPB. Bleeding events were associated with intraoperative hypothermia duration (p=0.03). Otherwise, clotting or bleeding events were not associated with patient variables (including weight, age or presence of AKI), ECMO variables (including flow rates or ECPR occurrence), operative variables (including CPB, cross clamp or circulatory arrest occurrence or duration). Anticoagulant dosing and laboratory parameters of hemostasis (including PTT, TEG, aXa, fibrinogen, platelets) were distributed randomly in relationship both to one another, and to bleeding and clotting events.
Conclusion: Hemorrhage and thrombosis are prevalent as adverse events post-CPB for children supported on ECMO. Early post-CPB bleeding is associated with duration of operative hypothermia. Other factors associated with adverse hematologic events could not be identified in this granular single center review. This suggests that other factors, not easily accounted for (e.g. inflammatory response, genetic factors), may contribute to propensity of bleeding and clotting in the post-CPB period. A high degree of vigilance is required when caring for children in this high-risk time period and larger multi-center studies are needed to better identify bleeding or clotting risk profile in this cohort.
