Current evidence of survival benefit between chest-compression only versus standard cardiopulmonary resuscitation in out-of-hospital cardiac arrest : Updated systematic review and meta-analysis of randomized controlled trials with trial sequential analysis.

Abstract

Evidence to support abetter cardiopulmonary resuscitation method between standard vs. continuous chest compression (STD-CPR vs. CCC-CPR) is lacking. Our systematic review followed PRISMA guidelines. We searched PubMed, ScienceDirect, EBSCOhost, and ProQuest database from 1985 to 26September 2019 restricted to randomized controlled trial, human study, and English articles. Quality assessment of between-study heterogeneity and atrial sequential analysis (TSA) were conducted. We estimated overall significance with 80% power and adjusted Zvalues thresholds using O'Brien-Fleming α‑spending function. Required information size with 21% relative risk using the estimation between-group incidences provided from the median rate across trials was determined. Inconclusive TSA result will lead to size estimation of future RCT. Quality of evidence was analyzed using Grading of Recommendations Assessment, Development and Evaluation (GRADE) Handbook and TSA. Based on three trials in OHCA with dispatcher-guided and bystander-initiated CPR, our meta-analysis favors CCC-CPR for survival to hospital discharge, compared to STD-CPR (RR [Risk Ratio] = 1.21[1.01-1.46], 95% CI, p = 0.68, I2 = 0). However, current meta-analyses with 3031patients appeared to be inconclusive. There is asignificant risk of type1 error and therefore, results are potentially false positive. It is estimated that aminimal of 4331patients needed to deem aconclusive result and atotal of 5894patients with similar risk profile required to stabilize statistic results in future trials. Quality of evidence is downgraded to moderate due to serious imprecision based on TSA. Based on these analyses, evidence is inadequate to conclude the superiority of one CPR method over the other. Further trials with larger numbers of patients are needed to deem aconclusive and stable meta-analysis.