Carbon Footprint of General, Regional, and Combined Anesthesia for Total Knee Replacements

Forbes McGain, F.A.N.Z.C.A, F.C.I.C.M., Ph.D.; Nicole Sheridan, F.A.N.Z.C.A.; Kasun Wickramarachchi, B.Sc., M.P.H., M.D.; Simon Yates, M.D., Brandon Chan, M.B.B.S.; Scott McAlister, B.Sc., P.grad., Dip.Sci., M.Eng.Sci.

Disclosures

Anesthesiology. 2021;135(6):976-991. 

In This Article

Results

Between January 9, 2019, and June 10, 2019, 36 patients underwent total knee replacements in operating room 4 at Williamstown Hospital, Western Health, Melbourne. As planned for this convenience sample and dependent upon researcher availability, we obtained anesthesia data for 30 patients: 10 patients in each group of general anesthesia, spinal anesthesia, and general plus spinal (combination). We excluded 1 patient (from combined general and spinal group) as they received nitrous oxide, leaving 29 patients (discussed later). The average/mean knee replacement anesthesia duration times (and ranges) were as follows: general anesthesia, 161 (113 to 193) min, spinal, 200 (168 to 288) min, and combination, 189 (128 to 241) min. Eight general anesthesia patients received sevoflurane, one total intravenous anesthesia, and one sevoflurane/total intravenous anesthesia combination. Six general anesthesia patients were intubated, while four had a laryngeal mask placed. All 10 patients receiving spinal anesthesia had sedative propofol infusions. For the patients receiving combination anesthesia, six received sevoflurane, and three received total intravenous anesthesia, while eight were given laryngeal masks, and two were intubated.

Background Data: Masses and Types of Disposables, Gases, and Electricity Used for Reusable Equipment

Appendix 1 and Appendix 2 give background data and calculations about the masses and energy required to wash reusable equipment. Appendix 3 gives masses of pharmaceuticals, led by cefazolin, tranexamic acid, paracetamol, and propofol, which are given in larger quantities/masses than other drugs. Intravenous paracetamol was given to one or two patients per group. Note (from Materials and Methods) that propofol has a carbon footprint of only 21 g carbon dioxide equivalent emissions/g propofol,[31] so using 3-h total intravenous anesthesia propofol at 700 mg/h will have carbon dioxide equivalent emissions of less than 50 g carbon dioxide equivalent.

Table 1 gives the equipment types used including the mean, 25%, 75% (interquartile range), and minimum–maximum (range). The total masses of single-use equipment used were as follows: general anesthesia (mean, 996 g; interquartile range, 873 to 1,033 g; range, 725 to 1,392 g), spinal anesthesia (mean, 997 g; interquartile range, 934 to 1,076 g; range, 885 to 1,184 g), and combination anesthesia (mean, 1,237 g; interquartile range, 1,100 to 1,285 g; range, 1,009 to 1,687 g). For single-use equipment, the majority of waste was from total plastics: average for general anesthesia, 783/996 g (78%); spinal, 729/997 g (73%); and combination, 932/1,237 g (75%). Glass was the next most common discarded material. There were minor (less than 100 g total mass) masses of other materials discarded (copper, cotton, latex, neoprene, and steel).

Table 1 also indicates that delivered oxygen was much greater for spinal anesthesia (mean, 1,328 l; interquartile range, 1,080 to 1,545 l; range, 990 to 1,950 l) versus general anesthesia (mean, 197 l; interquartile range, 116 to 271 l; range, 74 to 320 l), or combination anesthesia (mean, 256 l; interquartile range, 131 to 332 l; range, 53 to 824 l). Seven patients having spinal anesthesia received oxygen flow rates of 6 l/min, and three of 8 to 10 l/min. For the nine general anesthesia patients who received sevoflurane, the range was 14 to 44 ml (range, 6 to 15 ml/h), and for the seven combined anesthesia patients, the range of sevoflurane use was 11 to 50 ml (5 to 16 ml/h). Using 6 ml/h of (liquid) sevoflurane for 3 h will have carbon dioxide equivalent emissions of approximately 6 ml × 3 h × 1.5 (density) × 130 global warming potential in carbon dioxide equivalent emissions for sevoflurane[13] = 3.5 kg carbon dioxide equivalent emissions.

Desflurane was unused, and nitrous oxide used for one patient. Both desflurane and nitrous oxide are known to have high global warming potential (2,540[11] and 265,[33] respectively), which could easily skew overall results for this 30-patient convenience sample. The one patient who received nitrous oxide had 111 l N2O over 3.25 h. The carbon dioxide equivalent emissions for the nitrous oxide alone are 111/24.5 = 4.5 moles = 4.5 × 44 g = 200 g (0.2 kg) N2O = 0.2 × 265[33] = 53 kg carbon dioxide equivalent emissions. Thus, compared with using sevoflurane alone, the carbon dioxide equivalent emissions from using nitrous oxide are more than 10-fold greater.

Table 2 indicates carbon dioxide equivalent emissions from anesthesia per patient anesthetic items as calculated from the types and masses of consumables used (Appendix 1 and Appendix 2), and the electricity requirements for washing/sterilizing reusable equipment, patient warming, anesthetic gas scavenging, and the anesthesia machine. Note in Table 2 the column heading "Carbon dioxide equivalent emissions per kg, item, ml, or l," which indicates the differing carbon intensities of materials for their entire life cycle. Cotton has high carbon dioxide equivalent emissions per kilogram due to decomposition emitting methane (vs. steel and plastic, which are nonbiodegradable).[22] Considerably more plastics were used than disposable cotton; thus, plastics contributed the majority of the carbon dioxide equivalent emissions for disposable equipment. The summary carbon dioxide equivalent emissions for each group in the last two lines of Table 2 indicate the directly measured averages, and the indirectly measured 95% CIs as calculated by Monte Carlo analysis. As noted in the Materials and Methods, the 95% CIs may not be reflective of the directly measured interquartile ranges and minima/maxima seen in Figure 2.

Figure 2.

Carbon dioxide equivalent emissions for general, spinal, and combined anesthesia: mean, interquartile range (25%–75%), and minimum/maximum.

Carbon Dioxide Equivalent Emissions: Effects of Anesthetic Duration

As Table 2 and Figure 3 indicate, the average/mean duration of spinal and combined anesthesia were approximately 40 and 30 min more (i.e., 20% longer) than general anesthesia. The increased duration for spinal/combined anesthesia is at least partly due to increased time to undertake the spinal anesthetic. The longer spinal and combined anesthetic duration increased the carbon footprint of electricity for the patient air warmer and scavenging by 0.8 and 0.6 kg carbon dioxide equivalent emissions, respectively. Further, because spinal anesthesia was 20% longer than general anesthesia, this added approximately 2.76 × 0.2 = 0.6 kg carbon dioxide equivalent emissions to oxygen use for the spinal anesthetic. A spinal anesthetic of 20% shorter duration would thus have approximately 1.4 kg carbon dioxide equivalent less emissions. The effects of anesthetic duration had a much lower magnitude of effect upon the carbon footprint of other anesthetic activities.

Figure 3.

Categorizations of carbon dioxide equivalent emissions: general, spinal, and combined anesthesia.

Carbon Dioxide Equivalent Emissions: Averages, Ranges, and Components

Using Monte Carlo modeling, we found that the carbon dioxide equivalent emission means/averages were similar for all three approaches, and that the 95% CIs overlapped considerably, resulting in difficulty in making group comparisons. For general anesthesia, the mean was 14.9 kg carbon dioxide equivalent emissions (95% CI, 9.7 to 22.5); spinal anesthesia, 16.9 kg carbon dioxide equivalent emissions (95% CI, 13.2 to 20.5); and combination anesthesia, 18.5 kg carbon dioxide equivalent emissions (95% CI, 12.5 to 27.3). Figure 2 provides graphical contextualization of the means, interquartile ranges, and minimum-maximum ranges of the carbon dioxide equivalent emissions for the three anesthesia approaches. Figure 2 indicates that the interquartile ranges are relatively close, but there are considerable intragroup outliers. The range for spinal anesthesia was less than for general or combination anesthesia as there was a more standard approach (spinal procedure, propofol infusion, no variability in [unused] anesthetic gas use, minor variation in oxygen delivery/hour).

Table 2 and Figure 3 indicate that electricity for the patient air warmer was responsible for at least 2.46 kg carbon dioxide equivalent (16%) emissions of all anesthesia approaches. Total single-use plastics, glass, and so forth were responsible for 3.5 (general anesthesia), 3.4 (spinal), and 4.3 (combination) kg carbon dioxide equivalent emissions, respectively (20 to 25% total, with the majority from single-use plastics). All pharmaceuticals beyond gases were responsible for 1.2 to 1.3 kg carbon dioxide equivalent emissions, 7 to 8% total for all three approaches. For general anesthesia, sevoflurane (global warming potential = 130 times carbon dioxide)[11] for 9/10 patients was the principal contributor; average 4.7 kg carbon dioxide equivalent emissions (32% total), range 2.7 to 8.6 kg carbon dioxide equivalent emissions. The patient who received total intravenous anesthesia represented the minimum 8.4 kg carbon dioxide equivalent emissions in the general anesthesia group. For the combination anesthesia group, sevoflurane contributed an average 3.1 kg carbon dioxide equivalent emissions (17% total), range 0.6 to 10.0 kg carbon dioxide equivalent emissions. For spinal and combination anesthesia, washing and sterilizing reusable gowns, plastic spinal trays, and so forth contributed 4.5 kg carbon dioxide equivalent and 4.0 kg carbon dioxide equivalent emissions, respectively (coal was 75% of electricity for Melbourne, with 1.1 kg carbon dioxide equivalent emissions/kilowatt-hour).[23,34] Oxygen use was also important to carbon dioxide equivalent emissions for spinal anesthesia (2.8 kg carbon dioxide equivalent emissions, 16% total) as O2 flow rates were 6 to 10 l/min, compared with 0.5 to 3 l/min for general and combination anesthesia approaches.

Environmental Impacts: International Comparisons

Figure 4 indicates the modeled results of our data with electricity sourced in three other countries/regions: China, the European Union, and the United States (source: Ecoinvent).[22] The carbon dioxide equivalent emissions per kilowatt-hour varies due to different energy sources. Australia and China have similar "carbon intensities" (carbon dioxide equivalent emissions per kilowatt-hour) due to their reliance on coal, while the European Union (and the United Kingdom) has large nuclear and hydro/wind/solar sources for electricity generation, and the United States is moving rapidly toward greater renewable electricity generation. Such modeling changed the carbon dioxide equivalent emissions for washing and sterilizing reusable equipment, and electricity for patient warming. We assumed that the carbon dioxide equivalent emissions due to the use of single-use equipment were identical between countries, i.e., produced in China, as this is the major source for single-use items in Australia and anecdotally elsewhere.

Figure 4.

Carbon dioxide equivalent emissions for general, spinal, and combination anesthesia (international comparisons).

From Figure 4, as expected, the carbon dioxide equivalent emissions for all three anesthesia approaches for Australia and China are close. For the European Union and the United States, the carbon dioxide equivalent emissions for spinal anesthesia are decreased compared to Australia due to the greater predominance of renewable electricity used to clean reusable equipment/gowns. In the European Union, spinal anesthesia has a carbon footprint of approximately 60% (9.9/16.9 kg carbon dioxide equivalent emissions) that in Australia. Comparing the results of Figure 2 (Australian data) with Figure 4 (international modeling), the minimum carbon dioxide equivalent emissions for general anesthesia in Australia (total intravenous anesthesia) is less than the European Union general anesthesia average (8.4 vs. 11.9 kg carbon dioxide equivalent emissions), but the minimum for spinal anesthesia for Australia (14.7 kg carbon dioxide equivalent emissions) is considerably higher than the European Union spinal average (9.9 kg carbon dioxide equivalent emissions) due to high carbon intensity Australian electricity required to clean reusable anesthesia equipment.

processing....