Environmental Impact of Minimally Invasive Surgery in the United States: An Estimate of the Carbon Dioxide Footprint

Abstract

Purpose

To attempt to quantitate the carbon footprint of minimally invasive surgery (MIS) through approximated scope 1 to 3 CO₂ emissions to identify its potential role in global warming.

Patients and Methods

To estimate national usage, we determined the number of inpatient and outpatient MIS procedures using International Classification of Diseases, ninth revision-clinical modification codes for all MIS procedures in a 2009 sample collected in national databases. Need for surgery was considered essential, and therefore traditional open surgery was used as the comparator. Scope 1 (direct) CO₂ emissions resulting from CO₂ gas used for insufflation were based on both escaping procedural CO₂ and metabolic CO₂ eliminated via respiration. Scopes 2 and 3 (indirect) emissions related to capture, compression, and transportation of CO₂ to hospitals and the disposal of single-use equipment not used in open surgery were calculated.

Results

The total CO₂ emissions were calculated to be 355,924 tonnes/year. For perspective, if MIS in the United States was considered a country, it would rank 189th on the United Nations 2008 list of countries' carbon emissions per year. Limitations include the inability to account for uncertainty using the various models and tools for approximating CO₂ emissions.

Conclusion

CO₂ emission of MIS in the United States may have a significant environmental impact. This is the first attempt to quantify CO₂ emissions related to MIS in the United States. Strategies for reduction, while maintaining high quality medical care, should be considered.

Introduction

Laparoscopic surgical techniques and indications have expanded dramatically over the past 30 years since the inception of laparoscopy in medical practice. Because of recent advances in robot-assisted surgery, the number of laparoscopic robot-assisted procedures is exponentially rising as well. For example, in urology, more than half of radical prostatectomies are currently performed robotically. In 2009, an estimated 65% to 85% of all prostatectomies were completed using a robot-assisted laparoscopic approach.^1,2 This is remarkable considering the technology only received approval from the U.S. Food and Drug Administration in 2000. The collateral effects of minimally invasive technology are controversial and are currently being debated.³ The environmental collateral effects of minimally invasive surgery (MIS) have not been considered.

Carbon dioxide is the principle gas used in MIS for insufflation. CO₂ contributes 9% to 26% of the greenhouse effect, mostly from fossil fuel use, implicating it in the current global warming trend since the industrial revolution of the 20^th century.⁴ The burning of fossil fuel has produced three quarters of CO₂ emissions globally with the remaining amount secondary to deforestation, land utilization, and other factors.⁴ The levels of CO₂ emission are projected to be 90% to 250% increased in the year 2100 compared with baseline levels in 1750 if current trends continue unmitigated.⁴ This has prompted urgent warnings from the scientific community regarding the dire consequences of the resulting global warming. It follows that major consumers of fossil fuel have started to consider alternatives in an attempt to abate this undesired trend.

The environmental impact of healthcare in the United States, the second highest producer of CO₂ emission in the world and the 19.91% overall global contributor, has only recently been estimated in a research letter in the Journal of the American Medical Association by Chung and colleagues.⁵ They estimated that the healthcare sector contributes 7% of the entire U.S. CO₂ emission. There has been no published literature known to the authors assessing the impact that MIS use of CO₂ has on this figure in the United States.

The aim of this analysis is an attempt to quantitate the carbon footprint of MIS through approximated scope 1 to 3 greenhouse gas (GHG) emissions, as defined by the GHG protocol,⁶ to identify its potential role in global warming.

Patients and Methods

The need for surgery was considered as essential and, therefore, the analysis used traditional open surgery as the comparator. Other components of the overall carbon footprint common to surgery in general (ie, operating theater, electricity use, patient travel, paper products used) were considered equivalent. Only additional aspects unique to MIS were considered in the analysis. Other GHGs, as inventoried in the Kyoto protocol,⁷ were not considered, but the authors recognize that a complete accounting of the environmental impact of MIS would include this. CO₂ was considered alone because it is used for insufflation in MIS and it is the only GHG additionally unique to the procedure. Essentially, our calculation addresses the additional carbon emissions of performing procedures via MIS rather than open surgery.

To determine an estimate of CO₂ emission related to MIS, the first step was to separate the contributing components into scopes of emission. Scope 1, as defined by the GHG protocol, involves direct GHG emissions from sources that are owned or controlled by the entity. Scope 2 involves indirect GHG emissions resulting from the generation of electricity, heating and cooling, or steam generated off site but purchased by the entity, and the transmission and distribution. Scope 3 involves indirect GHG emissions from sources not owned or directly controlled by the entity but related to the entity's activities.⁶

Scope 1 CO₂ emissions were considered as gas that was used during MIS for insufflation. CO₂ escapes into the atmosphere via two processes during MIS. Directly, CO₂ escapes via leaks at port sites, decompression of insufflation at the end of surgery, or inadvertently because of the CO₂ tubing valve open. Indirectly, patients will absorb CO₂ across intra-abdominal viscera, eventually diffusing into the bloodstream. This absorbed CO₂ is ultimately eliminated via respiration into the atmosphere. The amount may be calculated using the equation proposed by Christopher and Wolf and depends on end-tidal CO₂, tidal volume, respiratory rate, atmospheric pressure, partial pressure of water vapor, and the weight of the patient.⁸ This absorbed CO₂ is such a minute amount after calculation that it was not included in the analysis.

A typical CO₂ cylinder used in our institution's operating room contains 65 liters of compressed USP grade gas. Using the Ideal Gas Law, 1 mole of any gas occupies 22.4 liters at 1 atmosphere of pressure.⁹ Because 1 mole of CO₂ weighs 44 g, calculations reveal there are 0.00015 metric tonnes of CO₂ in one cylinder. To estimate the operative time/cylinder, we used our institution's procedure numbers, operative times, and overall CO₂ use for the year 2009. The calculated time/cylinder was 1.6 L/hour of laparoscopy based on 2387 procedures. The above calculations were introduced in an initial general surgery analysis by Gilliam and associates¹⁰ at the University of Liverpool, United Kingdom.

Data regarding the number of MIS procedures performed both for inpatient and outpatient settings in the United States were identified in national databases. Inpatient common MIS procedures (Table 1) were identified using International Classification of Diseases, ninth revision-clinical modification codes in the Nationwide Inpatient Sample collected by the Healthcare Cost and Utilization Project.^11–13 These data were cross referenced with inpatient and ambulatory statistics compiled by the U.S. Department of Health and Human Services and the Centers for Disease Control and Prevention.¹⁴ Outpatient MIS procedures were extracted from this dataset. Intuitive Surgical's robot-assisted procedure data were obtained from a publicly available online investor presentation from their website.¹⁵ Average operative times for each procedure were estimated using data from our institution, and we supplemented procedure data not currently performed at our institution with average procedure times from published series.^16–21 The number of cylinders and CO₂ emissions were calculated from these data.

Table 1.

Total Scope 1 Carbon Emissions for Minimally Invasive Surgery

	Total number of procedures
	Inpatient	Outpatient	Total hours
Gastrointestinal
Cholecystectomy	374,485	348,000	722,485
Appendectomy	218,558	227,000	277,850
Bariatric surgery	126,850	151,000	445,558
Colon	77,108
Gynecologic
Hysterectomy	91,835	84,000	527,505
Salpingo-oophorectomy/tubal ligation	389,288	91,000	240,144
Urology
Prostatectomy	90,000		360,000
Nephrectomy (partial/radical/nephro-U)	34,022		102,066
Miscellaneous
Laparoscopy NOS	64,569	59,000	123,569
Robot-assisted procedures NOS	93,508	280,524	374,032
Total hours		3,233,917	Hours
Total # CO2 cylinders		2,021,298	Cylinders
Total CO2 emission		303.0	Tonnes

NOS=natural orifice surgery.

Contributions to scope 2 and 3 CO₂ emissions were identified as all other processes involved before and after the actual MIS procedure. Calculable processes before surgery were broadly categorized as CO₂ capture/compression and transportation (delivery) of CO₂ to hospitals. Postprocedure CO₂ emissions were calculated relating to single use equipment unique to MIS and their requirement of incineration as biomedical waste. All other indirect emissions were considered to be equivalent to open surgery for purposes of comparison.

The Environment Input-Output Life-Cycle Assessment (EIOLCA) model developed by Carnegie Mellon University Design Green Institute (US 2002 version) was used to estimate CO₂ emission involved in CO₂ capture/compression.²² The theory was originally conceived by Wassily Leontief, and his work on input-output life-cycles won him the Nobel Prize in economics in 1973. Industrial CO₂ is produced in numerous ways, mostly as a by-product of other processes, such as hydrogen energy production plants converting methane to CO₂. Medical or USP grade CO₂ requires high standards of purity and therefore is often mined from natural CO₂ springs, where it is produced by acidified water acting on dolomite or limestone.⁴ Our institution's supplier confirmed our CO₂ is mined from a natural source in Delaware City, DE. The Carnegie Mellon EIOLCA tool was used by inputting the estimate of economic sector activity for the largest medical CO₂ supplier in the United States, specifically for industrial gas manufacturing. This was estimated using the supplier's 2009 annual corporate report and their published breakdown of sales by economic sector.²³ To specifically target the MIS procedural use of CO₂, only the United States (52% of all sales), medical (8%) and packaged gas (31%) portions of 2009 annual sales were used for the EIOLCA model. Not all packaged medical CO₂ delivered is used for MIS, however, and therefore we attempted to correct for this by using our institution's 2009 data as an index case: 6102 L of CO₂ were delivered, but only 2604 L (43%) of CO₂ were directly used for MIS procedures. In a similar attempt to focus on MIS in the model, only industrial gas categories of manufacturing, power generation/supply for mining, and gas extraction output CO₂ emissions were included in the final total.

Transportation of CO₂ to healthcare facilities was estimated also by using our institution as an index case. The number of miles/CO₂ cylinder was calculated for CO₂ emission estimation. We used a standard CO₂ semitruck transport with an approximate 6 miles/gallon fuel efficiency and estimated based on a 16 tonne payload (weight of CO₂ gas alone) and an 18 tonne base freight weight in the model.²⁴ The total distance from the source mine in Delaware City, DE, to our institution in New York City is 140 miles. A carbon footprint calculator based on U.S. Department of Transportation (US DOT) fuel efficiency data and Greenhouse Gas Protocol Initiative (GHGPI) mobile guides were used to estimate the carbon emissions for transportation.²⁵ Calculations revealed that every CO₂ cylinder used requires 2 miles of semitruck transport time.

Data for the number of disposable instruments, specifically laparoscopic trocars, were obtained from U.S. market engineering research as of 2004,²⁶ because this was the only published data available nationally. The average weight of a laparoscopic trocar was approximated. Robot-assisted procedures, which usually need three to four disposable instruments, were estimated based on Intuitive Surgical's procedure numbers, instrument catalogue unloaded weights,²⁷ and using a general rule of 10 uses before disposal. The incineration of the instrument biomedical waste was estimated based on a common carbon footprint calculation with the assumption that incinerating 1 kg of plastic produces approximately 6 kg of CO₂.²⁸

Results

There were 2,520,223 MIS procedures included for 2009. The total operative time was estimated at 3,233,917 hours that translated into 2,021,198 CO₂ cylinders. The total CO₂ scope 1 emissions were 303 tonnes (Table 1).

Scope 2 and 3 CO₂ emissions from CO₂ capture/compression were calculated using the EIOLCA model and inputting $69 million of economic sector activity specific to the United States, medical, packaged gas (with Memorial Sloan-Kettering Cancer Center index case correction factor). The subtotal of CO₂ emissions for industrial gas manufacturing, power generation and supply, and gas extraction were calculated as 351,400 tonnes. For CO₂ transportation, the US DOT/GHPI calculation for 4,042,396 miles traveled to deliver CO₂ revealed 2970 tonnes of CO₂ emissions. Finally, to incinerate 208,441 kg of plastic biomedical waste from disposable trocar and robotic instrument use, 1251 tonnes of CO₂ emissions were generated. The total indirect CO₂ emissions were 355,621 tonnes (Table 2).

Table 2.

Total Scope 2 and 3 Carbon Emissions for Minimally Invasive Surgery

CO2 capture/compression		$USD (millions)
U.S. CO2 supplier	Total global sales (adjusted for inflation) 2009	9102
	U.S. sales (52%)	4733
	Medical sector (11%)	521
	Packaged gas (31%)	161
	MSKCC MIS correction (43%)	69
		CO2 emissions
EIOLCA calculation	Industrial gas manufacturing	251 000
	Power generation and supply	83 700
	Gas extraction	16 700
	Subtotal CO2 emissions	351,400	Tonnes
CO2 transportation	Number of CO2 cylinders	2,021,198	Cylinders
	Total miles/all cylinders	4,042,396	Miles
US DOT/GHGPI calculation	Subtotal CO2 emissions	2970	Tonnes
Incineration of biomedical waste
U.S. laparoscopic trocar data 2004	Number of disposable laparoscopic trocars	6,200,000	Trocars
	Average weight of trocar	30	Grams
	Total weight plastic	186,000	kg
U.S. robotic instrument data 2009	0.8 kg/unloaded instruments/10 uses
	Total weight plastic	22,441	kg
	Subtotal CO2 emissions	1251	Tonnes
	Total scope 2–3 CO2 emissions	355,621	tonnes

USD=United States dollars; MSKCC=Memorial Sloan-Kettering Cancer Center; MIS=minimally invasive surgery; EIOLCA=Environment Input-Output Life-Cycle Assessment; US DOT=United States Department of Transportation; GHGPI=Greenhouse Gas Protocol Initiative.

The overall CO₂ emissions from MIS were estimated at 355,924 tonnes/year.

Discussion

Sustainable healthcare has only recently entered into the medical lexicon. This first and only attempt by researchers in the United States to quantify the environmental impact of healthcare was published this past year by Chung and colleagues.⁵ Their report, using the same EIOLCA tool as the present study, estimated that healthcare contributes to 7% of total U.S. CO₂ emissions per year. They suggested that measuring and then reducing healthcare environmental impact should be considered as an extension of improving healthcare quality overall. Indeed, the National Health Service in England has organized and initiated a Carbon Emission Reduction program as part of their overall sustainable healthcare agenda.²⁹ Our analysis is an attempt to quantify the carbon footprint of MIS and identify major sources of CO₂ emission to propose mitigating factors in the United States.

A previous small study from the University of Liverpool in 2008 attempted to calculate the carbon footprint of general surgery from 2005 to 2007 at their center.¹⁰ They concluded that laparoscopy contributes a negligible total amount of CO₂ emission to global warming. This claim was unfortunately shortsighted. Their analysis only included direct CO₂ emissions. Similar to our study, the scope 1 CO₂ emissions of MIS are exceedingly small when comparing it on a national and global scale. This narrow thinking is similar to measuring the CO₂ released while drinking an artificially carbonated beverage and concluding that it has no environmental impact because the number is so minute. All processes involved in manufacturing and delivering the beverage, as well as disposal, need to be considered. Thus, the GHG Protocol initiative requires all three scopes to be calculated when performing industry carbon accounting.³⁰

There have been a number of unsuccessful attempts to replace CO₂ use in laparoscopy. These efforts have centered on mitigating the potential adverse physiologic, oncologic, and immunologic consequences studied during capnoperitoneum.³¹ Ranging from the use of other gaseous mediums such as helium or argon for insufflation, to completely gasless systems, non-CO₂ MIS has not gained widespread acceptance. Consideration of strategies such as these may be a potential approach in reducing the carbon footprint of MIS, although the practicality of this suggestion is at this time limited.

The total estimated CO₂ emission from MIS is 355,924 tonnes of CO₂/year in this study. To put this in perspective, it amounts to just 0.1% of the entire calculated U.S. healthcare carbon emission as evidenced by Chung and coworkers.⁵ Another way to put this, however, is that it amounts to driving a medium sized car 80,000 times around the earth at the equator or 645,000 flights from New York City to London.³² Still another way, MIS in the United States amounts to more CO₂ emission/year than yearly CO₂ emissions of 27 entire countries as listed by the United Nations from 2008 data.³³ It would rank 189th overall.

The monumental task of attempting to empirically quantitate CO₂ emissions according to a specific activity, such as MIS, needs to be emphasized. If measuring CO₂ emissions is the first crucial step in the process of eventually reducing a carbon footprint, more transparency and more statistics are needed by all players to identify mitigating factors.

The overwhelming majority of CO₂ emissions in this study were indirect. Therefore, it is incumbent on healthcare, as a consumer in the industrial market, to work with their suppliers to attempt to reduce the overall carbon footprint. Individuals performing MIS can do their part by reducing the amount of inadvertent CO₂ released during an operation by using the ALARA (as low as reasonably achievable) principle. When factoring in all of the indirect aspects, a small amount of CO₂ conserved can translate quickly into a meaningful impact. Also, using nondisposable items will significantly decrease the carbon footprint. This is of particular concern considering the rise of robotic surgery where there are appreciably more disposable items used than either traditional laparoscopic or open surgery.²

Study limitations include the inability to account for uncertainty, particularly using the EIOLCA tool, which has been previously outlined.⁵ The scope 2 and 3 carbon emissions relating to the initial CO₂ capture/compression far outweigh other factors calculated; therefore, the final number will reflect any inadequacies and errors intrinsic to the model itself. It is, however, the best estimation method we found for this type of analysis currently available. Interestingly, the major CO₂ supplier used as an index case in our model publishes a sustainability report and their company's complete carbon disclosure.³⁴ They have been recognized as a leader in corporate responsibility and sustainability, likely because of their commitment to such processes. Our estimate is surprisingly validated considering so many factors. For 2009, they report, in a statement of GHG emissions, that their scope 1 emissions categorized according to CO₂ as the source as 320,000 tonnes CO₂ emissions. Their scope 2 emissions were 9,317,000 tonnes, and scope 3 emissions were 226,000 tonnes. Again, using their annual report percentages for 2009 (52% U.S. sales, 11% medical sector, 31% packaged gas, 43% MIS correction factor), the total emissions are 75,203 tonnes of CO₂ emissions. This represents 21% of our calculated total; however, there are a number of other companies that supply medical gas to hospitals in the United States. Unfortunately, no other company has as rigorous carbon accounting as the index company and therefore external validation of our estimate is not possible at this time.

We did, however, perform a number of analyses in an attempt to demonstrate an association between the increased use of MIS, increased CO₂ use, and, consequently, increased carbon footprint. Using our institution's operating room CO₂ use since 1999, a sharp increase is noted in 2005 when robot-assisted surgery use rose dramatically (Fig, 1). By combining our CO₂ supplier's yearly sales in medical packaged gas to the United States from1997 to 2008 (adjusted for inflation and just before the U.S. financial market crisis), total number of laparoscopic cases performed in the United States (from 1997–2009 in the Nationwide Inpatient Sample collected by the Healthcare Cost and Utilization Project dataset), and Intuitive Surgical's reported U.S. robot-assisted procedures (2006–2011), an apparent association becomes evident (Fig. 2). All indications from this analysis point to increasing use of CO₂ in operating theaters over time. Furthermore, if our CO₂ emissions estimate is correct and this trend continues, the carbon footprint of MIS may become a significant issue for sustainable healthcare in the future.

FIG. 1.

Our institution's volume of carbon dioxide operating room use for minimally invasive surgery from 1999 to 2010.

FIG. 2.

Trends of United States yearly inpatient laparoscopic procedures (---), yearly robot- assisted procedures (...), and carbon dioxide supplier's yearly U.S. medical packaged gas sales adjusted for inflation (- - - -).

Conclusion

The CO₂ emissions of MIS in the United States, when considering both direct and indirect factors, have a significant environmental impact. This should be considered in larger strategies to reduce healthcare's carbon footprint while maximizing healthcare quality.

Abbreviations Used

CO₂

carbon dioxide

EIOLCA

Environment Input-Output Life-Cycle Assessment

GHG

greenhouse gas

GHGPI

Greenhouse Gas Protocol Initiative

MIS

minimally invasive surgery

U.S.

United States

US DOT

United States Department of Transportation

Acknowledgments

Supported by The Sidney Kimmel Center for Prostate and Urologic Cancers and by Award Number U54CA137788/U54CA132378 from the National Cancer Institute.

Michael McGregor, M.A. (Memorial Sloan-Kettering Cancer Center, Editorial Office, Urology Division) provided review and editing assistance. Lystra Swift and Melvin McLean provided invaluable assistance in data collection.

Disclosure Statement

No competing financial interests exist.