Fasting prior to cardiac catheterization is a routine practice to minimize the risk of complications. Recent studies suggest that non-fasting protocols may be equally safe and increase patient satisfaction. We performed a meta-analysis of randomized controlled trials (RCTs) to examine the safety of fasting versus non-fasting prior to cardiac catheterization.

We searched for eligible RCTs comparing fasting versus non-fasting protocols prior to cardiac catheterization from inception through December 21, 2024. Studies were included if they reported at least one of the outcomes of interest– nausea/vomiting, aspiration event, new ventilation/oxygen requirements, hypotension, hypoglycemia, and acute kidney injury. The treatment effect of each outcome was measured using the logarithmic odds ratios (logOR) and estimated under the Bayesian paradigm. Under the hierarchical Bayesian random effect model, we elicited an informative prior for the logOR ∼ (0, 0.1), representing the null hypothesis of no treatment effect. Between-study heterogeneity was elicited with a weakly informative half-Cauchy prior of a 0.5 scale. All analyses were conducted using R version 4.0.

9 studies met the inclusion criteria with a total of 3567 patients (1805 in fasting and 1762 in non-fasting). The Bayesian meta-analysis yielded a posterior mean OR of 0.99 [95 % credible interval (CrI): 0.82–1.20] for nausea and vomiting, 0.99 (95 % CrI: 0.82–1.21) for aspiration event, 1.003 (95 % CrI: 0.83–1.22) for new ventilation and oxygen requirements, 1.04 (95 % CrI: 0.87–1.25) for hypotension, 1.02 (95 % CrI: 0.85–1.24) for hypoglycemia, and 0.97 (95 % CrI: 0.81–1.18) for acute kidney injury. All CrI include 1 and the point estimates are very close to 1, indicating a lack of evidence to drive away from the prior assumption of no average effect size. Sensitivity analyses using three distinct prior scenarios (non-informative, optimistic informative, and skeptical informative) and a subset of studies conducted in 2023–2024 yielded similar findings.