Prediction of short- and long-term outcomes by electrocardiography in survivors of out-of-hospital cardiac arrest

Background: Programs focusing on early defibrillation have improved both short- and long-term survival of patients with VF out-of-hospital cardiac arrest (OHCA). Subsequent long-term management of survivors would be facilitated by a straight-forward, non-invasive method of identifying those at highest risk for recurrence. Therefore, we assessed the predictive value of the standard ECG to determine both short- and long-term outcomes in survivors of VF OHCA to assist in risk stratification of those patients at highest risk of sudden death. Methods: All patients with an OHCA between November 1990 and December 2000 who received early defibrillation for VF in Olmsted County Minnesota (MN) were included. Cox proportional hazards modeling was used to examine ECG variables and subsequent ICD deployment and death. Results: Two hundred patients presented in VF OHCA; of these 138 (69%) survived to hospital admission (seven died in the emergency department prior to admission) and 79 (40%) were discharged. The QRS duration (141 ± 41 ms in nonsurvivors, 123 ± 35 in survivors, P = 0.004) was predictive of short-term mortality in patients who did not survive to hospital discharge. The ventricular rate, PR interval, presence of right or left bundle branch block, QTc, ST elevation myocardial infarction, and atrial fibrillation/flutter were nonpredictive. The average length of follow up for hospital dismissal survivors was 4.8 ± 3.0 years. In univariate analysis, each 30 ms interval increase in the QRS width and PR interval was associated with increased mortality and ICD deployment hazard ratio of 1.6 (CI 1.1-2.5, P = 0.02) and 1.12 (CI 1.0-1.2, P = 0.05), respectively. In multivariate analysis accounting for admission ejection fraction, a PR > 200 ms [HR 4.5 (CI 1.7-11.8, P = 0.022)], QRS width increase greater than 30 ms [HR 1.9 (CI 1.3-2.8, P < 0.001)], and a QRS > 120 ms [HR 2.4 (CI 1.1-5.4, P = 0.032)] were predictive of long-term mortality and ICD shocks. Conclusion: Careful evaluation of the admitting and discharge ECG provides prognostic information for in-hospital and long-term outcomes, respectively in this cohort of out-of-hospital cardiac arrest survivors. The QRS duration on the dismissal ECG following VF OHCA provides prognostic information which might be useful to identify those at highest risk long-term, and who would benefit from more aggressive antiarrhythmic therapy and cardiac stabilization.