Abstract 11845: How Early Should We Initiate Targeted Temperature Management in Comatose Cardiac Arrest Survivors?

Abstract

Introduction: The optimal time for initiation of targeted temperature management (TTM) in comatose cardiac arrest survivors has not been elucidated. We analyzed the association of time from return of spontaneous circulation (ROSC) to initiation of TTM and induction time with outcomes in cardiac arrest survivors. Methods: A retrospective analysis of data for adult comatose cardiac arrest survivors who were treated with TTM from 2008 to 2015 was conducted. Time from ROSC to initiation of TTM was defined as pre-induction time. Pre-induction time was divided into four groups as follows: ≤120, 121-240, 241-360, and ≥361 min. Additionally, it was divided by different binary cutoffs of 120, 180, 240, 300, 360, or 420 min to identify a threshold association. The primary endpoint was neurologic outcome at discharge. Results: The study included 478 patients; of these, 121 (25.3%) died, and 317 (66.3%) were discharged with poor neurologic outcomes. The median pre-induction time was 225 min (interquartile range [IQR], 155-300 min), and it differed according to good (202 min; IQR, 150-300 min) and poor (230 min; IQR, 160-310 min) neurologic outcomes ( p =0.022). A multivariate logistic regression analysis revealed that delayed pre-induction (odds ratio [OR], 1.003; 95% confidence interval [CI], 1.001-1.005) and ≥361 min of pre-induction time (OR, 6.238; 95% CI, 1.928-19.628) compared to <120 min of pre-induction time were associated with poor neurologic outcome; however, 121-240 min (OR, 2.263; 95% CI, 0.969-5.284) and 241-360 min (OR, 2.111; 95% CI, 0.866-5.147) of pre-induction time were not associated with neurologic outcome. When the cohort was divided by cutoffs of 120 min (OR, 2.288; 95% CI, 1.044-5.012) and 360 min (OR, 2.679; 95% CI, 1.133-6.332), pre-induction time remained a significant predictor of poor neurologic outcome. Good neurologic outcome groups had a longer induction time. However, induction time was not an independent predictor of neurologic outcome. Conclusions: Delayed cooling was associated with increased poor neurologic outcomes at discharge among comatose cardiac arrest survivors treated with TTM. Patients with pre-induction time over 120 or 360 min had a higher probability of poor neurologic outcome.