Comparing the climate impact of inhaled anesthetics and clinical strategies to cut operating-room emissions
Every general anesthetic delivered by inhalation begins at the vaporizer, where a liquid volatile agent — desflurane, sevoflurane, or isoflurane — is metered into the fresh gas flow and carried to the patient through the breathing circuit. Nitrous oxide, a gas rather than a volatile liquid, is blended in directly from a pipeline or cylinder. Unlike most drugs, inhaled anesthetics are eliminated almost entirely unchanged by exhalation — desflurane is metabolized at under 0.02%, sevoflurane at roughly 5% — so the overwhelming majority of every dose leaves the lungs chemically intact. The circuit's scavenging system, built to protect staff from occupational exposure, captures this exhaled gas and vents it — typically untreated — straight outdoors. A routine clinical tool thus becomes, breath by breath, a continuous release of potent greenhouse gases.
A modern anesthesia workstation combines a fresh gas flow meter, an agent-specific vaporizer, a circle breathing system with a CO2 absorber (soda lime canister), and a scavenging interface. Fresh gas — a blend of oxygen, air, and sometimes nitrous oxide — passes through the vaporizer, which adds a precisely calibrated concentration of anesthetic vapor before the mixture reaches the patient via endotracheal tube, supraglottic airway, or face mask.
The circle system is designed for partial rebreathing: exhaled gas passes through the CO2 absorber, which chemically removes carbon dioxide, allowing the remaining anesthetic-laden gas to be recirculated and topped up with fresh gas rather than discarded. The proportion rebreathed versus vented depends directly on the fresh gas flow rate the clinician sets — the single largest controllable determinant of waste-gas volume.
Excess gas exits through a pressure-relief valve into the scavenging system, tubing that carries waste anesthetic gas away to protect staff from chronic exposure. Scavenging is a workplace-safety intervention, not an environmental one: at most hospitals worldwide, scavenged gas is exhausted directly outdoors with no capture, filtration, or destruction.
The healthcare sector is not an innocent bystander in climate change — it is responsible for an estimated 4.4% of global net greenhouse gas emissions, and if it were a country it would rank as the fifth-largest emitter on Earth. Within a hospital, the operating room is disproportionately resource-intensive, consuming three to six times more energy per square meter than the rest of the building, driven by high air-exchange rates, sterile processing, and specialized equipment.
Inhaled anesthetic gases and nitrous oxide are consistently identified as major contributors to OR-specific emissions, with some single-center studies attributing roughly half of an operating room's total greenhouse gas footprint to anesthetic gas use alone. Because anesthesia providers directly select the agent and flow rate for every case, anesthesiology is unusual in having a large share of its environmental footprint under the direct, real-time control of the clinician at the bedside.
One hour of desflurane anesthesia at a conventional 2 L/min fresh gas flow generates a carbon footprint comparable to driving a typical passenger car roughly 300–400 km — from a single operating room, in a single hour.
Desflurane, sevoflurane, and nitrous oxide are pharmacologically similar — all modulate GABA-A and NMDA receptor activity to produce reversible unconsciousness — which is why they are often interchangeable clinically. What differs enormously is their molecular fate after exhalation. Desflurane's very low blood-gas solubility gives rapid, precisely titratable control over anesthetic depth, a genuine clinical advantage, but its fully fluorinated structure makes it extraordinarily stable — and radiatively potent — in the atmosphere. Sevoflurane offers similar clinical flexibility at a fraction of the atmospheric cost. Nitrous oxide sits in between on a per-molecule basis, but its 114-year atmospheric lifetime and the large volumes in which it is administered give it an outsized cumulative footprint.
Desflurane, sevoflurane, isoflurane, and nitrous oxide all produce general anesthesia through overlapping mechanisms — potentiating inhibitory GABA-A chloride channels and depressing excitatory NMDA glutamate receptors — which is why they can often be substituted for one another with only modest adjustments to technique. What differs is their molecular structure downstream of the operating room.
Desflurane and sevoflurane are both fluorinated methyl ethyl ethers, but desflurane's structure — fluorine substituted at nearly every available position — gives it exceptional chemical stability. The same stability that makes it resistant to breakdown in the body (hence its rapid, predictable onset and offset) also makes it resistant to breakdown by atmospheric hydroxyl radicals, the main tropospheric "cleaning" mechanism for trace gases. Sevoflurane, lacking full fluorination, degrades far more readily both in vivo and in the atmosphere.
Nitrous oxide is chemically unrelated — a simple triatomic molecule (N2O) — but its inertness in the lower atmosphere lets it survive long enough to reach the stratosphere, where it also participates in ozone-destroying reactions.
Global warming potential (GWP) expresses how much heat a given mass of gas traps over a defined horizon (typically 100 years) relative to the same mass of CO2. On this scale, desflurane's GWP100 of approximately 2,540 dwarfs sevoflurane's 130 — desflurane is roughly 19–20 times more radiatively potent per kilogram. Isoflurane, an older and cheaper agent still used widely in lower-resource settings, sits at an intermediate GWP100 of approximately 510.
Nitrous oxide's GWP100 of 265 looks moderate next to desflurane, but two factors inflate its real-world impact: its 114-year atmospheric lifetime is by far the longest of the group, and it is typically administered as a high-volume carrier gas at 2–3 L/min for an entire case, rather than as a low-volume potent vapor. The combination of long persistence and high clinical volume can give nitrous oxide an absolute climate impact that rivals a volatile agent used at low flow, despite lower per-molecule potency than desflurane.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Desflurane | |||
| Sevoflurane | |||
| Isoflurane | |||
| Nitrous Oxide |
Once vented, anesthetic gas molecules do not simply vanish — they persist in the troposphere and lower stratosphere for years to over a century, absorbing outgoing infrared radiation in the same atmospheric window exploited by CO2 and methane. Desflurane's carbon-fluorine bonds make it exceptionally resistant to breakdown by hydroxyl radicals, giving it a 14-year atmospheric lifetime and a GWP100 of 2,540 — one kilogram traps as much heat over a century as 2,540 kilograms of CO2. Sevoflurane degrades far faster, lasting roughly 1.1 years, so its radiative legacy is dramatically smaller despite similar clinical use. Nitrous oxide is the outlier: it survives 114 years and, through its N-O bond chemistry, is also the single largest ozone-depleting substance still being emitted today.
Radiative forcing is the physical mechanism by which greenhouse gases warm the Earth's surface: certain molecules absorb outgoing infrared radiation in specific wavelength bands and re-emit part of it back toward the surface. Halogenated anesthetic agents are particularly effective absorbers because their carbon-fluorine and carbon-chlorine bonds vibrate at frequencies that fall within the "atmospheric window" — the 8–12 micrometer band where the atmosphere is otherwise relatively transparent to infrared, and where CO2 and water vapor absorb only weakly.
This is why halogenated anesthetics, like other fluorinated compounds such as HFCs and SF6, carry GWP values thousands of times greater than an equivalent mass of CO2 despite existing at only trace atmospheric concentrations. Global inhaled anesthetic use is estimated to contribute on the order of a few million tonnes of CO2-equivalent emissions annually — a tiny fraction of total global forcing, but within healthcare's own footprint, and given that far lower-impact alternatives exist for nearly every case, the marginal cost of continued high-GWP agent use is considered readily avoidable.
Nitrous oxide occupies a unique and troubling position in atmospheric chemistry: it is simultaneously a potent, long-lived greenhouse gas and, since the phase-out of chlorofluorocarbons under the Montreal Protocol, now the single largest ozone-depleting substance emitted by human activity — a finding highlighted in a landmark 2009 Science paper by Ravishankara and colleagues.
In the stratosphere, N2O reacts with excited oxygen atoms to form nitric oxide (NO), which catalytically destroys ozone in a chain reaction related to the one behind the Antarctic ozone hole, though driven by a different precursor than the CFCs that dominated the 20th-century ozone crisis. Because medical nitrous oxide use is a genuinely controllable point source — unlike agricultural N2O emissions from fertilized soils, which dominate the global budget — anesthesia has a disproportionate opportunity, relative to its emission share, to demonstrate this source can be reduced through clinical choice alone.
Global inhaled anesthetic and nitrous oxide use is estimated to generate on the order of 3 million tonnes of CO2-equivalent emissions each year — comparable to a mid-sized coal power plant's annual output, arising entirely from a single category of medical device.
Two clinical levers overwhelmingly determine an inhaled anesthetic's carbon footprint: fresh gas flow rate and agent selection. Because waste anesthetic gas volume scales directly with fresh gas flow, cutting FGF from a conventional 2 L/min toward minimal-flow settings of 0.5 L/min can reduce volatile agent consumption — and emissions — by 75–80% for an equivalent case, simply by letting more exhaled agent be rebreathed after CO2 absorption. Switching agents compounds the effect: moving a single case from desflurane to sevoflurane at the same flow cuts CO2-equivalent emissions roughly twenty-fold. Total intravenous anesthesia (TIVA) with propofol eliminates direct atmospheric forcing entirely, trading it for a smaller footprint elsewhere. Emerging gas-capture systems can recover or destroy over 99% of vented agent before it ever reaches the atmosphere.
Waste anesthetic gas volume is, to first approximation, directly proportional to fresh gas flow: at high flow (4–6 L/min) the circuit behaves almost like an open system, with most vaporized agent flushed straight to the scavenger rather than rebreathed. As fresh gas flow decreases, an increasing fraction of the patient's own exhaled agent is retained in the circuit, scrubbed of CO2, and recirculated — so the vaporizer needs to add proportionally less fresh agent to maintain the same alveolar concentration.
Low-flow anesthesia (roughly 1 L/min) and minimal-flow anesthesia (0.5 L/min or below) exploit this relationship deliberately. Published estimates suggest reducing fresh gas flow from a conventional 2 L/min to a minimal-flow target of 0.5 L/min can cut volatile agent consumption — and therefore both cost and emissions — by 75–80% for an equivalent case, with the added benefits of better heat and humidity conservation in the airway. The main tradeoffs are a slower response to vaporizer changes and a requirement for reliable gas monitoring, both well addressed by modern workstations.
Because desflurane's GWP100 is roughly twenty times sevoflurane's, switching the default institutional agent is one of the highest-leverage interventions available to an anesthesia department — often larger than any achievable flow-rate reduction on desflurane alone. Many professional societies now explicitly recommend sevoflurane over desflurane as first-line, specifically on environmental grounds, reserving desflurane for the narrow set of cases where its faster offset gives a genuine, patient-specific advantage.
Total intravenous anesthesia (TIVA), typically propofol by target-controlled infusion, eliminates volatile agent use — and direct atmospheric forcing — entirely. TIVA is not impact-free: propofol manufacturing and packaging carry their own footprint, and unmetabolized propofol entering wastewater has documented aquatic toxicity, so it trades one category of impact for a substantially smaller one rather than reaching true zero. For nitrous oxide, simply removing it as a routine carrier gas in favor of air-oxygen mixtures avoids both its greenhouse and ozone-depleting effects with minimal clinical tradeoff in most cases.
Modeling suggests combining minimal-flow technique with a desflurane-to-sevoflurane switch can reduce a single case's inhaled-anesthetic carbon footprint by well over 90% relative to high-flow desflurane — with no change beyond the vaporizer and flow-meter settings.
No single case's savings move the needle on climate change, but multiplied across a hospital's annual surgical volume, these clinical choices become measurable institutional outcomes. A large hospital performing roughly 12,000 anesthetic cases per year can convert modest per-case reductions — lower fresh gas flow, a sevoflurane-first formulary, TIVA where appropriate — into thousands of tonnes of avoided CO2-equivalent emissions annually. The UK's National Health Service demonstrated this at national scale, coordinating a near-complete phase-out of routine desflurane use across English trusts by 2024. This dashboard aggregates the same per-case dynamics across a simulated annual case volume, translating decisions at the anesthesia machine into a hospital-wide emissions trajectory.
In 2024, NHS England completed a coordinated near-total phase-out of routine desflurane use across English hospital trusts, following several years of trust-level restrictions, formulary changes, and clinician education driven by the NHS's broader Net Zero commitment to reach carbon neutrality for directly controlled emissions by 2040. The initiative relied on straightforward levers: removing desflurane vaporizers from routine stock, requiring documented clinical justification for exceptional use, and defaulting to sevoflurane or TIVA.
Professional bodies have reinforced this shift with formal guidance: the Association of Anaesthetists (UK) and the American Society of Anesthesiologists have each issued sustainability recommendations favoring low-flow technique and lower-GWP agent selection, and the World Federation of Societies of Anaesthesiologists has published environmental sustainability guidance for global adoption. Institutional defaults matter because individual clinician behavior change is difficult to sustain without them — removing desflurane from the shelf is more durable than relying on case-by-case discretion.
A single case's emissions reduction is measured in kilograms; a hospital's is measured in tonnes. A large hospital performing on the order of 12,000 anesthetic cases annually, shifting its default practice from historical high-flow desflurane toward low-flow sevoflurane or TIVA, can plausibly avoid on the order of several hundred to over a thousand tonnes of CO2-equivalent emissions per year — figures broadly consistent with the NHS's own reporting of roughly 7,000 tonnes of annual CO2-equivalent savings from its national desflurane restriction.
Expressed as a relatable equivalence, avoiding roughly 1,000 tonnes of CO2e per year is comparable to taking approximately 215 average passenger cars off the road for a full year — a scale that makes an otherwise abstract clinical decision (which vaporizer to fill, what flow rate to dial in) tangible as climate action, without new capital equipment, added cost, or any compromise in patient safety.
None of these interventions require new technology: the entire mechanism is a clinician turning a flow-meter dial and choosing which vaporizer to fill — yet aggregated across a hospital's case volume, the resulting emissions reduction rivals interventions that took years of capital investment to achieve in other sectors.