Study objective: Mechanical chest compression devices have been developed to facilitate continuous delivery of high-quality cardiopulmonary resuscitation (CPR). Despite promising hemodynamic data, evidence on clinical outcomes remains inconclusive. With the completion of 3 randomized controlled trials, we conduct a meta-analysis on the effect of in-field mechanical versus manual CPR on clinical outcomes after out-of-hospital cardiac arrest.
Methods: With a systematic search (PubMed, Web of Science, EMBASE, and the Cochrane Libraries), we identified all eligible studies (randomized controlled trials and nonrandomized studies) that compared a CPR strategy including an automated mechanical chest compression device with a strategy of manual CPR only. Outcome variables were survival to hospital admission, survival to discharge, and favorable neurologic outcome.
Results: Twenty studies (n=21,363) were analyzed: 5 randomized controlled trials and 15 nonrandomized studies, pooled separately. For survival to admission, the pooled estimate of the randomized controlled trials did not indicate a difference (odds ratio 0.94; 95% confidence interval 0.84 to 1.05; P=.24) between mechanical and manual CPR. In contrast, meta-analysis of nonrandomized studies demonstrated a benefit in favor of mechanical CPR (odds ratio 1.42; 95% confidence interval 1.21 to 1.67; P<.001). No interaction was found between the endorsed CPR guidelines (2000 versus 2005) and the CPR strategy (P=.27). Survival to discharge and neurologic outcome did not differ between strategies.
Conclusion: Although there are lower-quality, observational data that suggest that mechanical CPR used at the rescuer's discretion could improve survival to hospital admission, the cumulative high-quality randomized evidence does not support a routine strategy of mechanical CPR to improve survival or neurologic outcome. These findings are irrespective of the endorsed CPR guidelines during the study periods.
Copyright © 2015 American College of Emergency Physicians. Published by Elsevier Inc. All rights reserved.