Intraosseous vs. intravenous access during out-of-hospital cardiac arrest: a Bayesian secondary analysis of a randomised clinical trial

Abstract

Aim of the study: This study aimed to apply a Bayesian probabilistic framework to the Intravenous vs. Intraosseous Vascular Access for Out-of-Hospital Cardiac Arrest (IVIO) trial data to evaluate the likelihood of benefit for each vascular access method while incorporating various prior beliefs.

Methods: The IVIO trial was a randomised trial comparing intraosseous to intravenous access in 1,479 adults with non-traumatic out-of-hospital cardiac arrest. Bayesian analyses were pre-planned in the protocol and conducted using both non-informative and informative priors to calculate posterior probabilities for sustained return of spontaneous circulation, 30-day survival, and 30-day survival with a favourable neurologic outcome.

Results: Using non-informative priors for return of spontaneous circulation, the posterior probabilities that the effect of either vascular access exceeds the hypothesised difference were 1.2% (risk ratio > 1.27, favouring intraosseous access) and < 0.1% (risk ratio < 0.79 [1/1.27], favouring intravenous access). For 30-day survival and survival with a favourable neurologic outcome, the posterior probability that the risk ratio for intraosseous compared to intravenous access is between 0.83 (1/1.2) and 1.2 was 58% and 55%, respectively. For all analyses with informative priors, the results did not provide probabilities strongly favouring either intraosseous or intravenous access.

Conclusions: The probability of a clinically meaningful difference in return of spontaneous circulation between intraosseous and intravenous access for out-of-hospital cardiac arrest was very low, while results for 30-day outcomes were uncertain, with no strong evidence favouring either method.

Trial registration: EU Clinical Trials number: 2022-500744-38-00.

Clinicaltrials: gov number: NCT05205031.