732 - Pre-Transfusion Platelet Counts in Critically Ill Children with Extracorporeal Membrane Oxygenation

Poster Number: 732
Publication Number: 732.401

Oliver Karam, Yale, New Haven, CT, United States; Rebecca Birch, Westat, Rockville, MD, United States; Simone Glynn, Self, Reston, VA, United States; Marianne Nellis, Weill Cornell Medicine, Cornell University, New York, NY, United States

Background: Hemorrhagic complications are the most frequent complication for children supported by extracorporeal membrane oxygenation (ECMO), occurring in over two-thirds of the patients. To try to prevent or treat bleeding, children on ECMO receive a high volume of platelet transfusions. International guidelines suggest transfusing to keep the platelet count above 80 x 10⁹/L, although this recommendation is based solely on expert consensus. Recent interventional trials have suggested that in other populations of critically ill infants, platelet transfusions can be harmful, but it is unclear if this evidence has led to more restrictive transfusion practices for children on ECMO.

Objective:

Our objective was to describe the current platelet transfusion strategies in a large cohort of children supported by ECMO, using a contemporary database.

Design/Methods:

This is a retrospective multicenter study enrolling all children (0 to 18 years old) supported with ECMO and transfused platelets in four academic children's hospitals participating in the Recipient Epidemiology and Donor Evaluation Study-IV-Pediatric (REDS-IV-P), between April 2019, and July 2022. Transfusion doses were categorized as lower (≤ 75th percentile, i.e. ≤ 24 mL/kg) and higher doses ( > 75th percentile, i.e. > 24 mL/kg), as a surrogate marker for bleeding status.

Results: Three hundred ten children, median age 6.4 months (IQR 25 days-5.4 years) and median body weight 6.6 kg (IQR 3.7-18.0) on ECMO received 3437 platelet transfusions. Of these, 2578 were lower doses and 859 were higher doses. Among the former, the median pre-transfusion platelet count was 72 x 10⁹/L (IQR 56-94) in neonates (< 28 days) and 71 x 10⁹/L (IQR 48-102) in older infants and children. Among the latter, the median pre-transfusion platelet count was 113 x 10⁹/L (IQR 72-166) in neonates and 112 x 10⁹/L (IQR 66-177) in older infants and children (Figure 1). The median time between the platelet count and the platelet transfusion was 2.6 hours (IQR 1.3-4.7). There was no association between the patient’s age and pre-transfusion platelet count (p=0.86), but there was a significant difference between the four sites (p=0.01).

Conclusion(s): In a current cohort of children supported with ECMO, pre-transfusion platelet counts are heterogenous, and many are substantially higher than current expert-based recommendations. This variation in platelet transfusion strategies may in part reflect clinical severity or the bleeding status, but probably also reflects the current lack of strong evidence to guide platelet transfusion practices in this vulnerable pediatric population.