RESUMEN
Delirium is a frequent, serious, and preventable complication in critically ill children. Inflammation has been implicated as a mechanism for the development of delirium. Platelet transfusions may potentiate the body's pro-inflammatory responses. We hypothesized that receipt of platelets would be associated with delirium development in a pediatric intensive care unit (PICU). We performed a single-center retrospective cohort analysis including children admitted to the PICU between 2014 and 2018 who were transfused platelets within the first 14 days of admission. Data obtained included severity of illness, level of respiratory support, exposure to medications and blood products, as well as daily cognitive status. To account for time-dependent confounding, a marginal structural model (MSM) was constructed to delineate the relationship between platelet transfusion and next-day delirium. MSM demonstrated a 75% increase in the development of next-day delirium after transfusion of platelets (aOR 1.75, 95% CI 1.03-2.97). For every 1 cc/kg of platelet transfused, odds of next-day delirium increased by 9% (odds ratio 1.09, 95% CI 1.03-1.51). We reported an independent association between platelet transfusion and next-day delirium/coma after accounting for time-dependent confounders, with a dose-response effect. Minimizing platelet transfusions as much as clinically feasible may decrease delirium risk in critically ill children.
RESUMEN
Importance: Diagnostic genetic testing can lead to changes in management in the pediatric intensive care unit. Genetic risk in children with critical illness but nondiagnostic exome sequencing (ES) has not been explored. Objective: To assess the association between loss-of-function (LOF) variants and pediatric critical illness. Design, Setting, and Participants: This genetic association study examined ES first screened for causative variants among 267 children at the Morgan Stanley Children's Hospital of NewYork-Presbyterian, of whom 22 were otherwise healthy with viral respiratory failure; 18 deceased children with bronchiolitis from the Office of the Chief Medical Examiner of New York City, of whom 14 were previously healthy; and 9990 controls from the Institute for Genomic Medicine at Columbia University Irving Medical Center. The ES data were generated between January 1, 2015, and December 31, 2020, and analyzed between January 1, 2017, and September 2, 2022. Exposure: Critical illness. Main Outcomes and Measures: Odds ratios and P values for genes and gene-sets enriched for rare LOF variants and the loss-of-function observed/expected upper bound fraction (LOEUF) score at which cases have a significant enrichment. Results: This study included 285 children with critical illness (median [range] age, 4.1 [0-18.9] years; 148 [52%] male) and 9990 controls. A total of 228 children (80%) did not receive a genetic diagnosis. After quality control (QC), 231 children harbored excess rare LOF variants in genes with a LOEUF score of 0.680 or less (intolerant genes) (P = 1.0 × 10-5). After QC, 176 children without a diagnosis harbored excess ultrarare LOF variants in intolerant genes but only in those without a known disease association (odds ratio, 1.8; 95% CI, 1.3-2.5). After QC, 25 children with viral respiratory failure harbored excess ultrarare LOF variants in intolerant genes but only in those without a known disease association (odds ratio, 2.8; 95% CI, 1.1-6.6). A total of 114 undiagnosed children were enriched for de novo LOF variants in genes without a known disease association (observed, 14; expected, 6.8; enrichment, 2.05). Conclusions and Relevance: In this genetic association study, excess LOF variants were observed among critically ill children despite nondiagnostic ES. Variants lay in genes without a known disease association, suggesting future investigation may connect phenotypes to causative genes.