Medical Director, ECMO Department Children’s Hospital Los Angeles
Disclosure information not submitted.
Introduction: Bleeding and thrombosis are significant complications in extra-corporeal membrane oxygenation (ECMO). In a large retrospective cohort study, bleeding events occurred in 70% of pediatric ECMO runs and thrombotic events occurred in 37.5%. Platelet transfusion in the absence of clinically significant bleeding may increase risks of transfusion. There is no current published literature evaluating different platelet transfusion thresholds and resultant complications. In this study, we aim to describe if a difference exists in the incidence of hemorrhagic and thrombotic events in critically ill pediatric patients who received one or more platelet transfusions while on ECMO after decreasing the platelet transfusion threshold from 100,000/uL to 80,000/uL in 2023.
Methods: We conducted a single-center, retrospective, IRB approved cohort study of all neonatal and pediatric patients that received ECMO from January to December 2021 compared to January to December 2023. Descriptive statistics were analyzed using SAS 9.4M9.
Results: 81 total patients were included: 40 patients from 2021 and 41 from 2023. The median age was 82 days (IQR 3, 387), median weight 4.4 kgs (IQR 3.3, 14), 53% were male. They received a median of 115 hours of ECMO (IQR 65, 178). The median platelet count for the 2021 cohort was 111 (IQR 100, 120) whereas the median platelet count for the 2023 cohort was 91 (IQR 82, 100). The 2021 cohort received a median 13 transfusions totaling 111 mL/kg per run. The 2023 cohort received 14 transfusions of 116 mL/kg per run. Both groups had the same survivability from ECMO, with the 2023 cohort having a lower survive to discharge, 73% compared to 60%, (p=0.7). The incidence in thrombotic and hemorrhagic complications identified after ECMO were the same between the groups.
Conclusions: Despite lowering the transfusion threshold for ECMO patients, our single center retrospective analysis did not demonstrate a difference in the utilization of platelet transfusions, or in thrombotic and hemorrhagic complications. This may be the result of the retrospective nature of this work and not considering the indications for transfusion. Further, as a descriptive cohort study, it is not powered to identify statistically significant differences and indicates the importance of doing a prospective randomized trial.
