Abstract
PURPOSE: We evaluated associations between outcomes and time to achieving temperature targets during targeted temperature management of out-of-hospital cardiac arrest. METHODS: Using Comprehensive Registry of Intensive Care for out-of-hospital cardiac arrest Survival (CRITICAL) study, we enrolled all patients transported to participating hospitals from 1 July 2012 through 31 December 2017 aged ≥ 18 years with out-of-hospital cardiac arrest of cardiac aetiology and who received targeted temperature management in Osaka, Japan. Primary outcome was Cerebral Performance Category scale of 1 or 2 one month after cardiac arrest, designated as "one-month favourable neurological outcome". Non-linear multivariable logistic regression analyses assessed the primary outcome based on time to reaching temperature targets. In patients subdivided into quintiles based on time to achieving temperature targets, multivariable logistic regression calculated adjusted odds ratios and 95% confidence intervals. RESULTS: We analysed 473 patients. In non-linear multivariable logistic regression analysis, p value for non-linearity was < 0.01. In the first quintile (< 26.7 minutes), second quintile (26.8-89.9 minutes), third quintile (90.0-175.1 minutes), fourth quintile (175.2-352.1 minutes), and fifth quintile (≥ 352.2 minutes), one-month favourable neurological outcome was 32.6% (31/95), 40.0% (36/90), 53.5% (53/99), 57.4% (54/94), and 37.9% (36/95), respectively. Adjusted odds ratios with 95% confidence intervals for one-month favourable neurological outcome in the first, second, third, and fifth quintiles compared with the fourth quintile were 0.38 (0.20 to 0.72), 0.43 (0.23 to 0.81), 0.77 (0.41 to 1.44), and 0.46 (0.25 to 0.87), respectively. CONCLUSION: Non-linear multivariable logistic regression analysis could clearly describe the association between neurological outcome in patients with out-of-hospital cardiac arrest and the time from the introduction of targeted temperature management to reaching the temperature targets.