Materials and Methods
This prospective, nonrandomized, single center life cycle assessment was performed and follows the observational study Strengthening the Reporting of OBservational studies in Epidemiology checklist (www.strobe-statement.org.). The hospital ethics committee gave study approval (HREC/2018/Western Health/64), deeming that patient consent was not required (observational study not requiring patient data). We considered that 10 patients to each group (general anesthesia, spinal, and combined [general and spinal] anesthesia) provided an adequate convenience sample. We enrolled patients who were having elective total knee replacements consecutively, only excluding patients due to researcher unavailability. Life cycle assessment is a scientific method used to quantify the environmental footprint of a product or service throughout an entire life cycle. Previous studies have examined the carbon footprint of anesthetic equipment, which we have incorporated.[15–17] Our study focused on the carbon footprint of anesthesia as climate change is becoming increasingly important. Appendix 1 and previous reviews[13,18] contain further information about life cycle assessment methods.
Using the International Organization for Standardization (Geneva, Switzerland) ISO-14040 standards, we defined our study's functional unit as all anesthesia for a total knee replacement in a hospital in Victoria, Australia. The ISO-14040 standards life cycle assessment system boundary defines inclusions/exclusions (Figure 1). We did not include data for heating/ventilation/air conditioning, or any surgical equipment. Electricity consumption for anesthesia devices was estimated (not measured) from manufacturer data or from previous publications.[20,21]
We obtained patient anesthetic start and stop times. General anesthetics could be either volatile gas anesthetics or total intravenous anesthesia, with all cases requiring an airway device (laryngeal mask/endotracheal tube). Spinal anesthetics were delivered with sedation and by definition required no airway device. We present carbon dioxide equivalent emissions as total data, not per hour. For many items (drugs, plastic syringes, spinal anesthetic trays and gowns, inhalational induction), considerably more were used during the first hour of anesthesia than subsequently.
We examined the composition and weights of reusable and disposable consumables: gloves, gowns, syringes, airway devices, patient warming blankets, temperature probes, intravenous fluids, drugs, and gases, and associated immediate packaging. Volumes of oxygen, medical air, volatiles, and nitrous oxide use were obtained from the anesthetic machine (Aisys CS, GE Healthcare, USA) computer at the end of each case. Oxygen flows for sedated patients were manually recorded. We used the Andersen et al. study's global warming potential data for anesthetic gases. We used two life cycle inventories (Ecoinvent, Switzerland, and the Australian Life Cycle Inventory) to obtain carbon dioxide equivalent emissions associated with devices and processes.
For reusable items, previous data were used to estimate the environmental impacts of cleaning (sterile gowns, face masks, anesthetic breathing circuits, laryngoscope blades, and drug trays). We thus attributed the energy costs of reusable anesthetic equipment, i.e., kilowatt-hour/size of item as a proportion of washer load,[25,26] and 1.9 kilowatt-hours/kg items sterilized (Appendix 1). The reusable anesthetic breathing circuits were changed weekly unless contaminated, so their contribution to total carbon dioxide equivalent emissions was small (conservatively 25 operations per operating theater per week). Also included were the carbon dioxide equivalent emissions from carbon dioxide absorbent use (0.13 kg carbon dioxide equivalent emissions/h from Zhong et al.). Energy requirements for liquid oxygen were 0.001 kilowatt-hours/l for oxygen gas and 0.0003 kilowatt-hours/l for compressed medical air (Ecoinvent for electricity data, Australian Life Cycle Inventory for carbon dioxide equivalent emissions per kilowatt-hour).
Since we knew equipment mass, we used average production data about carbon dioxide equivalent emissions/kilogram waste from the Ecoinvent and Australian life cycle inventories as appropriate. We assumed general waste for all disposables except for some polyvinyl chloride recycling (face masks, oxygen tubing, and intravenous fluid bags), and polypropylene (spinal tray sterile wrap). Contaminated items (e.g., suction tubing) were assumed infectious/clinical waste (higher carbon dioxide equivalent emissions/kilogram, Ecoinvent), and pharmaceutical waste was assumed to undergo high-temperature incineration.
No public life cycle inventory data exist for most pharmaceuticals. We used the Parvatker et al. study's carbon dioxide equivalent emissions data approximations for 20 common anesthetic pharmaceuticals. From Parvatker et al., the average/mean g carbon dioxide equivalent emissions/g drug across the 20 drugs was 340 g carbon dioxide equivalent emissions/g drug, with, for example, propofol at 21 g carbon dioxide equivalent emissions/g propofol, and midazolam 444 g carbon dioxide equivalent emissions/g midazolam. Cefazolin, paracetamol, or tranexamic acid were unstudied, but we used this average 340 g carbon dioxide equivalent emissions/g drug to calculate carbon dioxide equivalent emissions. We estimated carbon dioxide equivalent emissions associated with intravenous fluid manufacture from our previous morphine life cycle assessment study (including production and sterilization of 0.9% NaCl bags). Some recycling was already occurring in the operating room (plastics/paper/cardboard).[7,32]
Data were modeled in SimaPro-9 life cycle assessment software (PRé Consultants, The Netherlands). We developed an inventory that quantified materials and energy used, and modeled this using the Ecoinvent (version 3.5) and Australian Life Cycle Inventory databases. We used Monte Carlo software algorithms (SimaPro) to obtain results and 95% CIs. We modeled our results with those for identical anesthetics being provided in China, the European Union, and the United States. We give the 95% CIs (from Monte Carlo analysis) only for the means/averages, and only for group aggregates (rather than individual components, e.g., plastics or electricity use), as the same assumptions are inherent in modeling the components that make up the aggregates (producing CIs for each component is lengthy and the numbers small). The 95% CI of the mean (indirectly obtained by Monte Carlo) indicates what the variability of the results could be if the study was performed many times, and may not closely reflect the directly obtained minima/maxima results. Further details about life cycle assessment methods are contained within Appendix 1.
Anesthesiology. 2021;135(6):976-991. © 2021 American Society of Anesthesiologists | Lippincott Williams & Wilkins