Quantifying the Effect of Cardiopulmonary Resuscitation Quality on Cardiac Arrest Outcome A Systematic Review and Meta-Analysis

Background-Evidence has accrued that cardiopulmonary resuscitation quality affects cardiac arrest outcome. However, the relative contributions of chest compression components (such as rate and depth) to successful resuscitation remain unclear. Methods and Results-We sought to measure the effect of cardiopulmonary resuscitation quality on cardiac arrest outcome through systematic review and meta-analysis. We searched for any clinical study assessing cardiopulmonary resuscitation performance on adult cardiac arrest patients in which survival was a reported outcome, either return of spontaneous circulation or survival to admission or discharge. Of 603 identified abstracts, 10 studies met inclusion criteria. Effect sizes were reported as mean differences. Missing data were resolved by author contact. Estimates were segregated by cardiopulmonary resuscitation metric (chest compression rate, depth, no-flow fraction, and ventilation rate), and a random-effects model was applied to estimate an overall pooled effect. Arrest survivors were significantly more likely to have received deeper chest compressions than nonsurvivors (mean difference, 2.44 mm; 95% confidence interval, 1.19-3.69 [P<0.001]; n=6 studies; I-2=0.0%; P for heterogeneity=0.9). Likewise, survivors were significantly more likely to have received chest compression rates closer to 85 to 100 compressions per minute (cpm) than nonsurvivors (absolute mean difference from 85 cpm, -4.81 cpm; 95% confidence interval, -8.19 to -1.43 [P=0.005]; from 100 cpm, -5.04 cpm; 95% confidence interval, -8.44 to -1.65 [P=0.004]; n=6 studies; I-2<49%; P for heterogeneity >0.2). No significant difference in no-flow fraction (n=7 studies) or ventilation rate (n=4 studies) was detected between survivors and nonsurvivors. Conclusions-Deeper chest compressions and rates closer to 85 to 100 cpm are significantly associated with improved survival from cardiac arrest.