Introduction: Despite recent advances in critical care, out-of-hospital cardiac arrest (OHCA) patients still have a low probability of survival. For the purpose of an objective assessment of a patient’s risk of death or neurological deficits, several clinical tools have recently been developed. Hypothesis: The first hypothesis is that OHCA and C-GRApH score could predict neurologic outcome after OHCA patients treated with targeted temperature management (TTM). The second hypothesis is that adding neurologic examination to those scores could further improve outcome prediction after OHCA. Methods: This retrospective study included OHCA patients treated with TTM from 2009 to 2017 in a tertiary teaching hospital in Seoul, Korea. We calculated two cardiac arrest specific risk scores (OHCA and C-GRApH) at the time of admission. Initial neurologic examination (Full Outline of UnResponsiveness Brainstem reflexes (FOUR_B) score and Glasgow Coma Scale motor score (GCS_M)) was also evaluated. The primary outcome was neurological outcome at 6 months after CA. Results: Of 311 subjects, 100 (32.2%) had a good neurologic outcome at 6 months after CA. OHCA score had AUC of 0.843 (95% CI 0.797 - 0.888) and C-GRApH score had AUC of 0.774 (95% CI 0.717 - 0.830). Addition of FOUR_B or GCS_M to OHCA score improved prediction of poor neurologic outcome (AUC 0.891, p = 0.001 for FOUR_B and AUC 0.878, p=0.004 for GCS_M). Results were similar for C-GRApH score to predict poor neurologic outcome. Conclusions: This study confirms the good prognostic performance of cardiac arrest specific scores to predict neurological outcomes in OHCA patients treated with TTM. Addition of neurologic examination variables further improved prognostication for neurologic outcome.

Introduction: Several studies have shown that the alpha-delta ratio (ADR) is associated with ischemic brain injury. Recently, an animal study has revealed that the ADR reaches the maximum point at 11 hours after ROSC, followed by downward curve (ADR peak) in the continuous EEG recordings of the rats with good outcome after cardiac arrest. We examined the existence of the early ADR peak after cardiac arrest in human. Methods: This is a prospective, observational study. Forty patients who survived out-of-hospital cardiac arrest underwent 33°C-targeted temperature management. We induced 33°C within 5 hours after ROSC and maintained for 24 hours, followed by 12 hours of rewarming period. All patients received sedative and paralytic agents until the restoration of normothermia. We started continuous EEG monitoring within 20 hours after ROSC until 72 hours after ROSC or mental recovery. We calculated alpha (8 - 13 Hz) and delta (0.5 - 4 Hz) frequency bands and computed ADR in the bifrontal channel (F3 - F4), bicentral channel (C3 - C4) and biparietal channel (P3 - P4). We did not remove artifact nor seizure events. Therefore, the whole continuous recordings were analyzed. Good neurologic outcome was defined as cerebral performance category 1 and 2 one month after ROSC. Results: Twenty-five patients showed poor outcome, while 15 patients showed good outcome. The figure is showing mean ADR ± SD change over time. The ADR reaches its highest amplitude 10 hours after ROSC and gradually decreases in the patients with good outcome regardless of the recording channel. However, the 10-hour ADR was not significantly different between the groups due to the large variances. Conclusions: Similar to the rats, the ADR peak appears early in the human patients with good outcome. However, due to the highly variable trend, application of ADR peak as a prognostic marker is limited. Regardless, the ADR implicates the crucial window of brain recovery time and remains an important subject requiring further study.