Serum S100 Protein Is a Reliable Predictor of Brain Injury After Out-of-Hospital Cardiac Arrest: A Cohort Study
Abstract
Purpose: To evaluate serum S100 protein at hospital admission and after 48 h in early neuroprognostication of comatose survivors of out-of-hospital cardiac arrest (OHCA). Methods: The study included 48 consecutive patients after OHCA, who survived for at least 72 h after the event.
The patients were divided based on their best cerebral performance category (CPC) achieved over a 30 day follow-up period: favorable neurological outcome (CPC 1-2) vs. unfavorable neurological outcome (CPC 3-4). Predictors of an unfavorable neurological outcome were identified by multivariable regression analysis.
Analysis of the receiver operating characteristic curve (ROC) was used to determine the cut-off value for S100, having a 0% false-positive prediction rate. Results: Of the 48 patients, 30 (63%) had a favorable and 18 (38%) had an unfavorable neurological outcome.
Eleven patients (23%) died over the 30 day follow-up. Increased S100 levels at 48 h after OHCA, but not the baseline S100 levels, were independently associated with unfavorable neurological outcome, with an area under the ROC curve of 0.85 (confidence interval 0.74-0.96).
A 48 h S100 value >=0.37 μg/L had a specificity of 100% and sensitivity of 39% in predicting an unfavorable 30 day neurological outcome. Conclusion: This study showed that S100 values assessed 48 h after an OHCA could independently predict an unfavorable neurological outcome at 30 days.