Multiorgan failure after return of spontaneous circulation in pediatric in-hospital cardiac arrest

Abstract

Purpose of the study: Previous studies found that the gray to white matter ratio (GWR) on brain computed tomography (CT) could be used to predict poor outcomes in cardiac arrest survivors. However, these studies did not consider the etiology of cardiac arrest. We sought to evaluate if the GWR on brain CT can help to predict poor outcomes after out-of-hospital cardiac arrest (OHCA) of cardiac etiology. Materials andmethods:Using amulticenter retrospective registry of adult cardiac arrest survivors treated with therapeutic hypothermia, we identified survivors of OHCA of cardiac etiology who underwent brain CTwithin 24h after successful resuscitation. Gray andwhitematter attenuationsweremeasured, and the GWRs were calculated as in previous studies. The prognostic values of the GWRs were analyzed, and a logistic regression analysis was performed to determine the contribution of the GWR in predicting poor outcomes (Cerebral Performance Category 3–5). Results: Of 283 included patients, 140 had good outcomes and 143 had poor outcomes. Although the GWRs could predict poor outcomes with statistical significance, the sensitivities were remarkably low (3.5–5.6%) at cut-off values with 100% specificity. No significant difference in predictive performance was found between the primary predictive model, containing independent poor outcome predictors, and the primary predictive model combined with the GWR. Conclusions: In a cohort of comatose adults after OHCA of cardiac etiology, the GWR demonstrated poor predictive performance and was not helpful in predicting poor outcomes.