Environmental and patient safety outcomes of a health-system Green Anesthesia Initiative (GAIA): a retrospective observational cohort study

Background

Inhaled anaesthetics are greenhouse gases. However, changes in the delivery of inhaled anaesthetics can mitigate environmental impact. We hypothesised that system-wide changes to the delivery of anaesthesia care would reduce environmental harm without compromising patient outcomes.

Methods

We launched the Green Anesthesia Initiative (GAIA) in March, 2022, with the aims of reducing the use of nitrous oxide, using less environmentally harmful inhaled fluorinated ethers, and increasing intravenous anaesthetic use. In this retrospective cohort study, we used electronic health record data from general anaesthetics performed on all patients older than 1 year between March 1, 2021, and Feb 28, 2023, at a single US academic medical centre across multiple sites, collecting data from before and after the introduction of GAIA. Patients with missing or invalid data recorded by the anaesthesia machine, patients given general anaesthetics for electroconvulsive therapy, and patients who met American Society of Anesthesiologists Physical Status Classification 6 were excluded. Using multivariable modelling, we compared estimated CO2, equivalents and, secondarily, anaesthetic dose, postoperative nausea and vomiting, pain scores on a 0–10 scale, and reports of intraoperative awareness with explicit recall.

Findings

We recorded 45 692 patients pre-intervention (23 193 [50·8%] female, 22 494 [49·2%] male, five [<0·1%] unknown) and 47 199 post-intervention (23 981 [50·8%] female, 23 209 [49·2%] male, nine [<0·1%] unknown). After the implementation of GAIA, CO2, equivalents were reduced by 14·38 kg per patient (95% CI –14·68 to –14·07; p<0·0001). There was no clinically meaningful difference in median anaesthetic delivered (minimum alveolar concentration –0·02 [95% CI –0·02 to –0·01]; p<0·0001) nor pain scores (–0·34 [–0·39 to –0·29]; p<0·0001). Postoperative nausea and vomiting was unchanged (odds ratio 0·98 [95% CI 0·94–1·02]; p=0·26). A small number of definite intraoperative awareness events were reported in both periods (one pre-intervention and two post-intervention).

Interpretation

A health-system wide intervention reduces greenhouse gas emissions attributable to anaesthesia care without detriment to patient outcomes.