BACKGROUND: Children supported by extracorporeal membrane oxygenation (ECMO) frequently receive platelet transfusions. However, little is known about how characteristics of platelet products affect transfusion effectiveness in ECMO patients. Our objective was to evaluate whether platelet product characteristics were associated with platelet count increment. STUDY DESIGN AND METHODS: Secondary analysis of the ECSTATIC randomized controlled trial. Children <18 years on ECMO were randomized to two prophylactic platelet transfusion threshold strategies. In this secondary analysis, the exposure was platelet transfusions characterized by storage duration, ABO compatibility, platelet additive solution, irradiation, and pathogen reduction status. The primary outcome was the platelet count increment after transfusion, dichotomized as higher versus lower based on the median value. Regression models were used to identify associations between product characteristics and platelet count increment, adjusting for clustering. RESULTS: cell/L (IQR 8; 36), and the median time from pre-transfusion to post-transfusion count was 5.9 h (IQR 4.8; 7.1). In univariable analysis, pathogen reduction was associated with a lower platelet count increment (odds ratio = 0.14, p = .01). In a multivariable model, higher pre-transfusion platelet count was independently associated with increased platelet count increment (adjusted odds ratio aOR = 1.03, p = .049), while pathogen reduction was independently associated with decreased count increment (aOR = 0.16, p = .048). CONCLUSIONS: Among critically ill children on ECMO, pathogen-reduction and pre-transfusion platelet count may influence the effectiveness of transfusion. These findings support the need for prospective trials that incorporate product-level characteristics into transfusion strategies for pediatric ECMO patients.
Karam et al. (Thu,) studied this question.