To derive a clinical prediction rule of termination of resuscitation (TOR) for out-of-hospital cardiac arrest (OHCA) with pre-hospital defibrillation given.

This was a retrospective multicenter cohort study performed in three emergency departments (EDs) of three regional hospitals from 1/1/2012 to 31/12/2018. Patients of OHCA aged ≥18 years old were included. Those with post-mortem changes, return of spontaneous circulation and receiving no resuscitation in EDs were excluded. A TOR rule was derived by logistic regression analysis based on demographics and end-tidal carbon dioxide level of included subjects with pre-hospital defibrillation given.

There were 447 included patients had received pre-hospital defibrillation, in which 148 had return of spontaneous circulation (ROSC), with 22 survived to discharge (STD). Independent predictors for death on or before ED arrival (no ROSC) included EMS call to ED time >20 min and ETCO2 level <20 mmHg from first three sets. A 2-criteria rule predicting death on or before ED arrival by fulfilling both of the independent predictors had a specificity of 0.95 (95% CI 0.90–0.98) and positive predictive value (PPV) of 0.95 (95% CI 0.90–0.98), if 2nd set of ETCO2 was used. The positive likelihood ratio was 10.04 (95% CI 4.83–20.89).

The 2-criteria TOR rule for OHCA patients with pre-hospital defibrillation had a high specificity and PPV for predicting death on or before ED arrival.