Immediate Defibrillation or Defibrillation After Cardiopulmonary Resuscitation

OBJECTIVES: This study aimed to determine whether short cardiopulmonary resuscitation (CPR) by emergency medical services before defibrillation (CPR first) has a better outcome than immediate defibrillation followed by CPR (shock first) in patients with ventricular fibrillation/pulseless ventricular tachycardia (VF/pulseless VT) out-of-hospital cardiac arrest. METHODS: We analyzed a national database between 2006 and 2008, and included patients aged 18 years or more who had witnessed cardiac arrests and whose first recorded rhythm was VF/pulseless VT. Those study subjects were divided into five groups in accordance with the CPR/defibrillation intervention sequence. Each group was subdivided into call-to-response intervals of <5 minutes and ≥ 5 minutes. We identified 267 patients in the shock-first group and 6,407 patients in the CPR-first group. One-month survival and neurologically favorable one-month survival rates were used for outcome measures. The association of intervention type on outcomes (one-month survival or neurologically favorable one-month survival) was analyzed using multivariate logistic regression analyses by adjusting potential confounding factors such as survey year, gender, age (years), bystander CPR, intubation, and call-to-response interval (min). RESULTS: The overall one-month survival rate was 26.2% (3,125/11,941) and the neurologically favorable one-month survival rate was 16.6% (1,983/11,934). The CPR-first group had a one-month survival rate of 27.8% (1,780/6,407) and a neurologically favorable one-month survival rate of 17.8% (1,140/6,404), and the shock-first group had survival rates of 24.7% (66/267) and 18.4% (49/267), respectively. There were no significant differences in one-month survival and neurologically favorable one-month survival in these two primary comparison groups (odds ratio [95% confidence interval], 0.85 [0.64-1.13] and 1.04 [0.76-1.42], respectively). Logistic regression analysis showed that neither CPR first nor shock first was associated with the rate of one-month survival or neurologically favorable one-month survival, after adjusting for potential confounders. CONCLUSIONS: In our study, CPR prior to attempted defibrillation did not present a better outcome compared with shock first as measured by either one-month survival or neurologically favorable one-month survival, after adjusting for potential confounders. Further studies are required to determine whether CPR first has an advantage over shock first.