Anthropogenic climate change is evident, and its impacts are being felt around the globe. The World Health Organization (WHO) recently revealed that approximately one in four deaths can be attributed to preventable environmental causes. Reducing greenhouse gas (GHG) emissions across all sectors is critical to contain climate change.1 The Healthcare Sector is a significant contributor with approximately 4.6% of the global GHG emissions in 2017.2 Thus, there is an urge to reduce global greenhouse gas emissions in the Healthcare Sector. Anaesthesiologists took early action to reduce global GHG emissions within their profession. National and International Societies of Anaesthesiologists such as the World Federation of Anaesthesiologists, the European Society of Anaesthesiology and Intensive Care and the German Society of Anaesthesiology and Intensive Care provided recommendations on ecological sustainability.3–5 In addition to total intravenous anaesthesia, volatile anaesthetics (VAs) remain fundamental for delivering effective and reliable anaesthetic care. In Europe, sevoflurane, isoflurane and desflurane are currently the most commonly used VAs.6 Sevoflurane and desflurane are classified as fluorinated hydrocarbons, and isoflurane as fluorochlorinated hydrocarbon. Agreements and protocols established under the United Nations Framework Convention on Climate Change (UNFCCC) cover a range of greenhouse gases, not just carbon dioxide. Countries set targets for reducing emissions across a ‘basket’ of gases, including carbon dioxide (CO2), methane (CH4) and various fluorinated gases. To make these different gases comparable, a common metric is used — essentially an exchange rate — that converts emissions of non-CO2 gases into ‘CO2-equivalent’ amounts. This allows total emissions to be reported on a consistent scale. In the 1990s, the Kyoto Protocol adopted the use of ‘global warming potential’ (GWP), using a 100-year time frame (GWP-100) to calculate the CO2 equivalence of anthropogenic GHG emissions under the UNFCCC (http://unfccc.int/kyoto-protocol; accessed on 8 September 2025). The 2016 UNFCCC Paris Agreement is less prescriptive in its specification of which metric should be used, but nevertheless GWP-100 remains very widely used. As a result, recommendations from anaesthesiology specialty societies regarding ecological sustainability have often relied on the differences in the GWP-100 values of VAs.3–5 As Earth's climate is a rather complex and not necessarily linear system, there is an array of metrics available to evaluate the climate effects of VAs, including the GWP over different time horizons, atmospheric concentration, atmospheric lifetime, radiative efficiency, effective radiative forcing (ERF), global temperature-change potential, and, for some gases, ozone-depleting properties; the ultimately consequential parameter is, however, arguably the effect on global near-surface temperature.7–9 An overview of metrics evaluating the climate effect of VAs is given in Table 1. Table 1 - Metrics evaluating the climate effect of volatile anaesthetics and HFC-134a Name Formula Atmospheric concentration (parts per trillion) Atmospheric lifetime (years) Radiative efficiency (W m2 ppb−1) Effective radiative forcing (W m−2) GWP 20 GWP 100 GTP 50 GTP 100 Carbon dioxide CO2 422 800 000 100+ 1.33 × 10−5 2.16 1 1 1 1 Desflurane CHF2OCHFCF3 0.37 14.1 0.464 0.00017 7020 2590 1260 521 Isoflurane CHF2OCHCICF3 0.11 3.5 0.426 0.00003 1930 539 122 98.4 Sevoflurane (CF3)2CHOCH2F 0.16 1.9 0.308 0.00006 702 195 43.5 35.4 HFC-134a CH2FCF3 129.5a 14 0.167 0.018 4144 1526 733 306 GTP, global temperature-change potential; GWP, global warming potential.Carbon dioxide concentration as annual mean for 2024 reported by May 2025 according to the National Oceanic and Atmospheric Administration (https://www.climate.gov/news-features/understanding-climate/climate-change-atmospheric-carbon-dioxide; accessed on 8 September 2025). All other values as reported in the Intergovernmental Panel on Climate Change, Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change except for HFC-134a concentrations.aUpdated values for 2023 from the NOAA Annual Greenhouse Gas Index) (https://www.climate.gov/news-features/understanding-climate/climate-change-atmospheric-carbon-dioxide; accessed on 8 September 2025).20 Recognising the currently intensified debate and growing importance of this issue, we devote this editorial to a central question: is there an optimal metric for assessing the climatic impact of VAs? To address this inquiry rigorously, it is necessary to systematically evaluate advantages and limitations inherent to each of the aforementioned metrics. A crucial element of rational decision-making is, in addition to possessing domain knowledge, a quantitative understanding of uncertainty.10 This shifts decision-making from a deterministic to a probabilistic approach. In Bayesian terms, it is optimal to act if the probability of an anticipated event exceeds the ratio of the action's cost to the total of the action's cost plus the potential loss from inaction.11 In practice, when taking action is inexpensive and the potential loss from inaction is high, action is justified even at low probabilities. Conversely, when taking action is costly and the possible loss is low, action should be considered if the event is highly probable. As for the metrics, the atmospheric concentration of a GHG is a key input for climate calculations, but its impact on global temperature is mediated by factors such as each gas's absorption and emission properties, as well as complex atmospheric, biological and oceanic processes. These influences including the feedbacks among those processes lead to the climate sensitivity, which determines the change in global mean air temperature.12 In consequence, the global mean air temperature anomaly had not found widespread use as a direct climate change metric until the UNFCCC Paris Agreement. A similar line of argumentation can be made for the radiative efficiency, which describes the potential of a GHG to warm the atmosphere. It is very useful for a comparative assessment of the relative climate consequences of different gases prior to their widespread use – but it does not take into account their actual concentration or effects over time. The ERF takes this concept one step further: it combines radiative efficiency with atmospheric concentration. Therefore, it is one of the most important metrics when judging the current effect of past emissions of any GHG on the planetary radiation budget (i.e. the difference between incoming and outgoing radiative energy in the Earth's climate system). It has also been used widely to assess the past and possible future evolution of such effects. Beyond scientific use, it can be helpful for policymakers to prioritise GHG emission countermeasures. For example, emissions of HFC-134a (a GHG widely used to replace the ozone-depleting chlorofluorocarbons in air conditioning units) have increased by several orders of magnitude since the 1990s. In turn, the increasing ERF from this gas was influential in the process of adding hydrofluorocarbons (HFCs) to the Montreal Protocol in 2015. Now the phase-down of production and consumption of HFC-134a and other similar GHGs is internationally regulated by that treaty, although atmospheric concentrations of HFC-134a continue to rise. At the same time this example highlights a significant shortcoming of ERF, that is, its inability to capture future climate effects, especially those that accumulate over time. Had such considerations on future developments played a more prominent role in policymaking in the 1990s, it might have prevented much of the HFC-134a that has since been emitted into the atmosphere. This is where the atmospheric lifetime comes into play. It is defined as the time it takes for a trace gas to decay to 1/e (or to about 37%) of its initial concentration under the assumption that emissions have fully ceased. Therefore, it can give guidance on how long the emission of a GHG may alter atmospheric composition and therefore impact climate. As this climate effect is also dependent on the radiative efficiency of the gas in question, it is beneficial to have a metric that considers both. This is the case for the GWP, which integrates the two parameters over a user-defined time horizon, most commonly 100 years. Note that the choice of time horizon is arbitrary while at the same time very influential for the derived GWP (see Table 1) and opens windows of uncertainty and potential confusion. The choice of such a horizon can primarily be utilised as a way of focusing policymaker attention on the urgency of climate action, that is, whether to make decisions that may influence climate over shorter or longer timescales as is described in more detail in Kalmar 2024.13 The GWP is a relative measure of the climate effect of an emission – it provides a comparison of the time-integrated radiative forcing from a pulse emission of 1 kg of a given gas with a 1 kg emission of CO2 (which is why the CO2 metrics in Table 1 have a value of 1). Thus, it can be used to convert emissions of a gas into a CO2 equivalent emission, although that CO2-equivalence will depend on which metric (GWP or some alternative) and time horizon that is chosen. The GWP has its shortcomings, which have been widely discussed and debated e.g. Forster et al. (2021) and references therein.14 First and foremost, it does not consider current or future atmospheric emissions or concentrations, which are essential to prioritise urgent policy measures. So, although a gas may possess a high value of GWP, if little of that gas is emitted, or ever likely to be emitted, then it may have limited importance from a climate perspective. Nevertheless, it still offers a reliable metric for individuals and industries to assess the relative climate impact of gases and reduce their carbon footprint. Also, the GWP's core characteristic (‘time integrated radiative forcing’) does not, at first glance, appear to directly relate to surface temperature change. Alternative metrics have emerged in recent years. One example is the Global Temperature-change Potential (GTP), which translates the effect from a single GHG directly into a temperature change for a given year after its emission, again relative to that of CO2. An alternative is to use the absolute GTP (AGTP) which does not rely on the specification of a reference gas. Neither the GTP nor the AGTP are, however, considering the effects over longer time scales. Also, to compare the impact on future temperatures by emissions of gases with shorter lifetimes (say, less than 10 years) to those of longer lived ones such as CO2, extensions of the GWP (the so-called GWP* and its near relatives) have been introduced.14 Nevertheless, the GWP has retained its usefulness and remains the most commonly used metric to date. In conclusion, both the ERF and the GWP are important and credible metrics of climate impacts of GHGs including VAs. They incorporate many of the important parameters including radiative forcing, atmospheric concentrations and lifetimes. As a disclaimer, we note that – due to the very nature of an editorial – this presents a simplified view of the complex science behind the climate impacts of GHGs. More comprehensive assessments including uncertainty considerations can be found elsewhere.12–18 Finally, three further aspects merit mentioning. Firstly, the comparison of the small contribution of VAs to overall anthropogenic forcing from main GHGs such as CO2 as recently reemphasised.9 It should be noted that GHG effects on the climate are additive: for every GHG, every kilogram emitted adds to the total. Dividing these into seemingly insignificant small shares could, on the one hand, be seen as a way of marginalising the climatic impact of VA and encouraging inaction. On the other hand, mathematics teaches us that the summation (integration) of very small changes can lead to finite effects. This can become even more important when thresholds such as the 1.5 °C goal of the UNFCCC Paris Agreement come into play. In that case, even a relatively small contribution could be enough to allow the exceedance of the threshold in deterministic terms or, in a probabilistic approach, lead to decisions to start actions. These arguments are general in nature and do not apply only to anaesthetics: GHG emissions from Luxembourg, Bonn, or London may only make up a small contribution to global climate change, but they add up with other small contributions to global climate change and could lead to additional decisive contributions that shift the global average temperature deviation beyond the Paris Agreement thresholds of 1.5 or 2 °C. This means that climate mitigation measures avoiding even small emissions can help to fend off exceedances of thresholds. Secondly, the arguments about the desirability of restricting VA emissions, notably desflurane, can get polarised into two camps. One is that it has a high GWP-100 and therefore should be controlled. The other is that the current emissions of desflurane are so small that its contribution to climate change is likewise small. We think that both views have some merit but miss a much broader point, which can be illustrated by returning to the example of HFC-134a. It has a comparable lifetime (14 years), but significantly lower radiative efficiency (less than half that of desflurane). Consequently, its GWP-100 is considerably less (about 1500 versus desflurane's 2500).15 When the UNFCCC Kyoto Protocol was signed in 1997, it could have been argued (and perhaps was) that HFC-134a emissions were so low that they could be safely ignored. But the measured concentrations of HFC-134a are now ∼25 times higher than they were in 1997, and concentrations continue to increase at a few per cent per year.19 The present-day radiative forcing due to halogenated greenhouse gases is about 18% that of CO2. HFC-134a contributes 5% of that halogenated greenhouse gas forcing and it is almost certainly destined to play an even larger role, as the concentrations of ozone-depleting gases fall. Given that desflurane is a more powerful greenhouse gas than HFC-134a, arguments based on the size of its current contribution to climate change could be regarded as lacking foresight. Thirdly, there is a point to be made for low-hanging fruit: for example, desflurane is an obviously much more detrimental choice in terms of its climate impact per kilogram emitted (Table 1) than sevoflurane or isoflurane. If patient welfare and cost-effectiveness are not significantly impacted at the same time, replacing desflurane can be a straightforward way to move toward reduced climate impacts in the healthcare sector. In this sense, the low-hanging fruit argument is nothing else than the Bayes optimal decision described above. This is of course only one possible measure of many. As science and technology are constantly evolving, many more improvements can certainly be expected in the future. From the perspective of environmental ethics in conjunction with science theory, it always makes sense to regularly evaluate climate mitigation measures and, if new findings become available, to allow for modifications.20 Consequently, we recommend that such constructive approaches (e.g. via lifecycle assessments) should be prioritised over regressive ones, and viable emerging options should be reassessed in regular intervals.
Coburn et al. (Mon,) studied this question.