Christopher Parshuram, MD
Pediatric Critical Care Medicine Physician, PhD
The Hospital for Sick Children, United States
Disclosure information not submitted.
Patricia Fontela
MD MSc PhD
Montreal Children’s Hospital McGill, United States
Disclosure information not submitted.
Michael Sauthier, MD
MD PhD
Sainte Justine Hospital, United States
Disclosure information not submitted.
Marisa Tucci, BS, MD
MD MSc
Saint Justine Hospital, United States
Disclosure information not submitted.
Geneviève Du Pont-Thibodeau, MD, MSc
Pediatric-Intensivist
CHU Sainte-Justine, Quebec, Canada
Disclosure information not submitted.
Title: Transfusion-related adverse events in hospitalized children: an ancillary cohort study of EPOCH.
Introduction/Hypothesis: EPOCH(JAMA2018;317:1002-12), an international cluster randomized controlled trial (RCT), investigated the incidence of significant clinical deterioration events(SCDE) leading to PICU admission in children on general wards. Little is known about the epidemiology of transfusion-related adverse events in children hospitalized outside intensive care units(ICU). The purpose of our study was to determine (1) the epidemiology of transfusions in hospitalized children and (2) the characteristics of SCDEs in transfused versus non-transfused children. We hypothesized that transfusions were common and were associated with SCDEs.
Methods: EPOCH randomized hospitals to standard monitoring or to the use of the Pediatric Early-Warning System, which was created to predict SCDE. SCDE was defined as a composite outcome reflecting late ICU admission. In the Province of Quebec, all transfusions are recorded in a TracelineTM database. We merged the data collected in EPOCH and in TracelineTM of children enrolled in EPOCH in two tertiary pediatric hospitals. Only SCDE that happened after the first transfusion were considered as potentially associated with transfusion.
Results: 773 children were enrolled in our study. 254(32.8%) received at least one transfusion: red blood cell (RBCT) (234/254, 92.1%), plasma (81, 31.9%), platelets (134, 52.8%), or cryoprecipitates (37, 14.6%). 69 children(27.2%) received only one RBCT while 62 (24.4%) received >5 RBCTs. The mean volume of RBCT per patient per hospitalization was 1.36±2.5 L(median 0.616 L). The length of time between the last transfusion and the SCDE was 9.22±30.2 days (median 0.78). The proportion of participants who developed at least one SCDE was 18.1% (46/254) among patients who received at least one transfusion and 36.5% (282/773) among non-transfused patients. There was no difference in the type of SCDE in transfused and non-transfused patients. Transfused children with SCDE were more likely to have a high-risk admission diagnosis (p< 0.001).
Conclusions: A third of children on general wards received at least one transfusion; 18.5% of these children developed a SCDE after their first transfusion that lead to a PICU admission. Further studies are required to understand the risk of transfusions in this population.