Operating theatres.
One operating theatre can emit as much as a small fleet of cars. The gases, the single-use steel, the ventilation running to an empty room at 2am — this is where a department audit pays back fastest.
Every action, graded.
Evidence grades travel with every claim. Click through to the study behind each number.
| ID | Action | Evidence | CO₂e / yr |
|---|---|---|---|
| ACT-011 | Run elective lists on sevoflurane, not desflurane Effects of departmental green anaesthesia interventions — British Journal of Anaesthesia 2025, DOI 10.1016/j.bja.2025.03.038 |
Strong RCT / dept data | 188 t/yr |
| ACT-012 | Model the volatile→TIVA transition before mandating it Cost-effectiveness of volatile→TIVA transition — Anaesthesia 2026, DOI 10.1111/anae.70111 |
Cohort / modelling | context |
| ACT-013 | Pressure-test the N₂O manifold; fix leaks at source Investigating N₂O leaks at St George — Anaesthesia & Intensive Care 2026, DOI 10.1177/0310057x251379095 |
Strong RCT / dept data | varies |
| ACT-014 | Count N₂O stock-rotation waste before you order again N₂O stock-rotation waste — Anaesthesia Reports 2026, DOI 10.1002/anr3.70086 |
Cohort / modelling | varies |
| ACT-015 | Join or copy a system-wide sustainable anaesthesia programme Sustainable Anesthesia Project — Healthcare (Basel) 2026, DOI 10.3390/healthcare14030300 |
Cohort / modelling | system |
Before you start.
Desflurane is the exit, not the rule
Clinical judgement first: some cases need what they need. The ledger grades the switch, not the anaesthetist.
Measure before you switch
Pull gas purchase logs by agent, 12 months back. Baseline first — that's what makes your number publishable.
The agent follows the airway, not the atmosphere.
Every action on this page passes through this gate.
Desflurane has a place — for some patients, some techniques, some anaesthetists' judgement. The 188-tonne saving applies to elective lists where an equally good alternative exists. The moment sevoflurane would compromise depth, recovery or safety for this patient, the carbon ledger closes: the ledger grades the formulary decision, never the anaesthetist's clinical choice in the room.
What the data shows.
Case: St George Hospital, 2021.
The team bought 1,121,400 litres of N₂O but could account for at most 801,866 litres of clinical use — 28% was leaking out of pipework that nobody had pressure-tested in years. Four-step method (procurement audit → clinical-need estimate → discrepancy → pressure test), fully replicable in any trust with a manifold. The gas bill hid the leak; the method found it.
Anaesthesia & Intensive Care 2026 · DOI 10.1177/0310057x251379095
Asked and answered.
Why is desflurane singled out?
Its global-warming potential is 2,640× CO₂ and it needs roughly 3× the fresh-gas flow of sevoflurane. Per modelled anaesthetic: desflurane 209.2 kg CO₂e vs sevoflurane 9.8 kg — a factor of 21 from the agent choice alone, before any workflow change.
Is TIVA the better target than sevoflurane?
Carbon-wise yes (2.0 vs 9.8 kg per anaesthetic), but the economics are honest work: the modelled cost per extra tonne saved is £416 vs the UK carbon price of £41.84 — 10× above market. Grade B, model it locally before mandating.
What does a department programme actually deliver?
A systematic review of 13 implemented programmes measured a mean 75.2% cut in department anaesthetic emissions. That is not a pledge; that is the post-intervention number.