Neurological outcome of extracorporeal cardiopulmonary resuscitation for out of hospital cardiac arrests

Purpose: To evaluate the efficacy of extracorporeal cardiopulmonary resuscitation (ECPR) using clinical neurological outcome. Methods: Patients were included in this single center retrospective analysis between 2003 and 2012, if they met following criteria: 1. out of hospital cardiac arrest who were still in cardiac arrest in any rhythm on arrival to the hospital, 2. underwent emergent extracorporeal cardiopulmonary bypass using veno-arterial extracorporeal membrane oxygenation (V-A ECMO) as an adjunct to conventional CPR. Trauma patients were excluded. Total of 64 patients were identified and evaluated on three clinical neurological outcome categories: neurologically intact, comatose consisting of severely disabled, vegetative state or brain death, and death. The time interval from collapse to induction of V-A ECMO on each category were compared. Analyses were made on neurological outcomes based on initial rhythm and types of disease. Comparison was made on median lengths of hospital stay in each outcome. Results: Of 64 patients, 18 patients (28%) were neurologically intact, 14 patients (22%) were comatose, and 32 patients (50%) were dead on discharge. Median time intervals from collapse to induction of V-A ECMO were 37 minutes for neurologically intact patients (p<0.05), 60 minutes for comatose patients, 60 minutes for patients who died. Neurological outcomes of 42 patients with ventricular fibrillation/ventricular tachycardia as an initial rhythm were as follows: 13 (31%) neurologically intact, 10 (24%) comatose, 19 (45%) dead. For 13 patients with pulseless electric activity (PEA) and 8 patients with asystole, outcomes were as follows in the same order shown above: PEA 4 (31%), 1 (8%), 8 (62%), asystole 0 (0%), 3 (38%), 5 (63%). Out of 51 patients with cardiac etiology, 11 (21.6%) were neurologically intact, 12 (23.5%) were comatose, and 28 (54.9%) were dead. In contrast, of 13 patients with non-cardiac etiology, 7 (53.8%) were neurologically intact, 2 (15.4%) were comatose, and 4 (30.8%) were dead. Median length of hospital stay for patients with neurologically intact, comatose, death were 11.5 days, 24.5 days, 2 days, respectively. Conclusion: ECPR yields 28% survival with good neurological outcome to otherwise futile patient population.