RESUMO
OBJECTIVES: Recommendations for extracorporeal cardiopulmonary resuscitation (ECPR) state that appropriate patient selection is important for the sake of efficacy and cost-effectiveness of ECPR. It is not known whether first documented rhythm plays a prominent role in economic outcomes of patients with cardiac arrest who received ECPR. METHODS AND RESULTS: We reviewed the medical records of 120 consecutive patients who received extracorporeal membrane oxygenation (ECMO) assisted CPR due to refractory circulatory collapse between 2008 and 2016 in Urasoe General Hospital. The patients presented with ventricular fibrillation or pulseless ventricular tachycardia (VF/VT; n = 59, 49.2%) or with asystole or pulseless electric activity (ASY/PEA; n = 61, 50.8%) as the first documented rhythm. Multivariate logistic regression analysis identified shorter duration from collapse to ECMO initiation (odds ratio, 1.95 per 10 min; 95% confidence interval, 1.32-2.89, p = 0.001), bystander CPR (odds ratio, 5.53; 95% confidence interval, 1.36-22.5, p = 0.017), and first documented rhythm of VF/VT (odds ratio, 3.93; 95% confidence interval, 1.30-11.8, p = 0.015) as clinical predictors for neurologically intact survival. Total hospital cost per life saved by ECPR for ASY/PEA was approximately twice that for VF/VT ($213,656 vs. $101,669). ECPR yielded Quality adjusted life years (QALYs) of 3.32 at a mean total cost of $39,634 for VF/VT and QALYs of 1.17 at a mean cost of $35,609 for ASY/PEA. The cost per QALYs was $11,081 for VF/VT and $29,447 for ASY/PEA. The incremental cost-effectiveness ratio of ECPR vs. conventional CPR was estimated to be $ 16,246 per QALY gained. CONCLUSION: ECPR for patients presenting with VF/VT was found to be highly cost-effective and ECPR for patients presenting with ASY/PEA was borderline cost-effective.