Mechanical ventilation vs bag-valve ventilation during cardiopulmonary resuscitation: a randomized crossover noninferiority trial
In brief
During CPR, mechanical ventilation registered 146 mL less tidal volume than bag-valve ventilation
In 99 intubated adults during CPR, device-calculated inspired tidal volume averaged 368 mL with protocolized mechanical ventilation versus 514 mL with bag-valve ventilation, an adjusted difference of 146 mL. Mechanical ventilation was less variable, but the trial did not establish noninferiority; low-volume readings were more common with it. The analyzer was unvalidated, and clinical outcomes were not assessed.
- Journal
- Resuscitation plus (Q1)
- Published
- 13 October 2026
- Study design
- Randomized controlled trial
- Evidence level
- Level 1, High (CEBM 1b)
- Authors
- S Vinay, S Manu Ayyan, V T Amrithanand, B G Kowsthubha
- PMID
- 42985663
- DOI
- 10.1016/j.resplu.2026.101530
Why clinicians should know about it
- Picked for Emergency Medicine (top studies of the week, 8 November 2026): Ventilation technique during CPR directly informs resuscitation
Abstract
AIM: Evidence guiding mechanical ventilation (MV) during CPR is limited, and device-calculated inspired tidal volume (VTi) during ongoing chest compressions may differ from the set volume during compressions. We compared VTi with protocolized MV and bag-valve ventilation (BVV) through an endotracheal tube for noninferiority. METHODS: This single-center randomized crossover noninferiority trial enrolled 104 intubated adults meeting termination-of-resuscitation criteria without return of spontaneous circulation to receive BVV and MV in randomized order within a 5-min peri-termination window. Device-calculated VTi was recorded for 1 min per method during ongoing compressions. The primary outcome was the between-treatment difference in per-participant median VTi; the noninferiority margin was -150 mL (MV minus BVV). RESULTS: Among 99 participants completing both periods, mean per-participant median VTi was 514.4 mL with BVV and 368.0 mL with MV; coefficients of variation were 40.1% and 28.3%. The adjusted mean difference was -146.2 mL (95% CI, -187.0 to -105.5). Because the lower bound crossed -150 mL, noninferiority was not demonstrated. Median VTi exceeded 10 mL/kg predicted body weight in 31.3% with BVV and 0% with MV, whereas values below 6 mL/kg occurred in 26.3% and 45.5%. CONCLUSIONS: Protocolized MV produced lower and less variable device-calculated VTi than BVV but did not meet noninferiority. BVV had a wider upper tail, whereas MV more often produced values below 6 mL/kg. The unvalidated analyzer and absence of clinical outcomes limit interpretation. CLINICAL TRIAL REGISTRATION: Clinical Trials Registry - India; CTRI/2024/08/072738.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.