Abstract 14421: Comparative Effectiveness of Bag-Mask Ventilation, Supraglottic Airway, and Endotracheal Intubation in Out-of-Hospital Cardiac Arrest: A Systematic Review and Bayesian Network Meta-Analysis
Abstract
Background: Optimal prehospital airway management in out-of-hospital cardiac arrest (OHCA) remains unresolved. We conducted the first Bayesian three-node NMA incorporating all available evidence through May 2026. Methods: PubMed, Scopus, Cochrane CENTRAL, and Web of Science were searched from inception to May 2026. Eligible studies included RCTs and propensity-matched/IPTW observational studies in adult non-traumatic OHCA comparing bag-mask ventilation (BVM), supraglottic airway (SGA), and endotracheal intubation (ETI). Primary outcome was neurologically intact survival (CPC 1-2/mRS ≤3). A Bayesian hierarchical random-effects NMA was performed (gemtc/JAGS; 4 chains, 100,000 iterations). Results: Thirteen studies were included (4 RCTs, 9 propensity-matched observational studies; total n>700,000 patients across 5 countries). All Gelman-Rubin statistics were <1.01 confirming MCMC convergence. No network inconsistency was detected across any outcome (maximum DIC difference 0.02).For neurologically intact survival (primary outcome), BVM ranked highest (SUCRA 97.8%), with both ETI (OR 0.60, 95% CrI 0.38-0.96) and SGA (OR 0.58, 95% CrI 0.37-0.95) significantly inferior. ETI and SGA did not differ from each other (OR 1.03, 95% CrI 0.72-1.46). For survival to hospital discharge, BVM again ranked highest (SUCRA 97.0%), with SGA significantly inferior (OR 0.63, 95% CrI 0.40-0.97) and ETI showing a non-significant trend (OR 0.68, 95% CrI 0.44-1.04). ROSC showed no significant differences between strategies. Between-study heterogeneity was moderate across all outcomes (tau 0.316-0.491).In the paramedic-led EMS subgroup (9 studies), BVM superiority was strengthened: ETI OR 0.43 (95% CrI 0.25-0.71) and SGA OR 0.46 (95% CrI 0.27-0.74). In the shockable rhythm subgroup, BVM retained highest SUCRA (91.4%). Critically, in RCT-only (n=4) and post-2015 sensitivity analyses, no significant differences between strategies were detected (all CrIs crossing unity), indicating the BVM advantage is concentrated in older registry data under pre-2015 protocols. CINEMA confidence was Low for BVM superiority over ETI and SGA for neurological survival, and Very Low for all ROSC comparisons. Conclusions: This Bayesian NMA of >700,000 patients demonstrates BVM superiority for neurologically intact survival in OHCA, strongest in paramedic-led systems. The absence of significant differences in RCT-only analysis highlights an urgent need for contemporary randomized evidence. Findings directly address the 2025 AHA guideline evidence gap on optimal prehospital airway management.