Met het ondertekenen van de ‘Green Deal Duurzame Zorg’ dient de gezondheidszorg rekening te houden met de invloed van behandelingen op het milieu (Green Deal, 2022). Opereren gaat gepaard met een relatief grote hoeveelheid afval en mogelijk varieert de milieu-impact tussen verschillende operatietechnieken. In het algemeen zijn er drie operatietechnieken beschikbaar (robot-geassisteerde-, laparoscopische- en open chirurgie) voor dezelfde indicatie. De robot-geassisteerde operatietechniek wordt steeds vaker uitgevoerd in de Nederlandse praktijk (Tummers, 2021). De meest recente studies tonen geen grote verschillen in effectiviteit aan tussen robot-geassisteerde en laparoscopische chirurgie (Aiolfi, 2021; Muaddi, 2021). Wanneer de uitkomsten van operatietechnieken voor de patiënt vergelijkbaar zijn, kan er overwogen worden om te kiezen voor de meest duurzame operatietechniek. Het is momenteel nog onduidelijk welke invloed de betreffende operatietechniek heeft op duurzaamheid. In deze module worden de duurzaamheidsuitkomsten van de drie verschillende operatietechnieken met elkaar vergeleken.

1. Climate change (critical)

2. Waste (critical)

3. Acidification (important)

4. Eutrophication (important)

5. Human toxicity (important)

6. Ecotoxicity (important)

7. Ozone depletion (important)

Description of studies

Power (2012) describes an LCA to quantify the CO₂ footprint of minimally invasive surgery (MIS) compared to open surgery in the United States. MIS included robot-assisted laparoscopic surgery and laparoscopic surgery. A total of 2,520,223 MIS procedures from the year 2009 were included. The analysis compared MIS procedures to traditional open surgery, using an Economic Input-Output-LCA (EIO-LCA). Therefore, prior to the analysis, varying aspects between MIS and open surgery were determined and only these factors were included in the analysis to quantify the CO₂ footprint. Other fixed components of the overall CO₂ footprint common to surgery in general (e.g. operating theatre, electricity use, patient travel, paper products used) were considered equivalent and were not taken into account. Different scopes of emissions were identified:

• Scope 1 CO₂ emissions: The CO₂ emission was defined as the amount of gas that was used for insufflation during MIS.
• Scope 2 and 3 CO₂ emissions: The CO₂ emissions were defined as all other processes before and after the MIS procedure. Furthermore, manufacturing and transportation of CO₂ cylinders, used to transport the CO₂ for insufflation, were taken into account.

Emissions related to the manufacturing process of the disposable instruments and to the postoperative stay were not included in the analysis. Thereby, not all disposable instruments were taken into account. Next to that, electricity use was considered equivalent, although it is expected to differ between the operating techniques, in particular between MIS and open surgery. MIS uses electricity driven instruments and cameras and in addition the robot uses robotic arms which require electricity. The relevant outcome measure was climate change (defined as CO₂ footprint).

Woods (2015) retrospectively reviewed 150 procedures of the consecutive and most recent patients to have undergone a staging procedure for endometrial cancer for three surgical modalities (50 per arm): robot-assisted laparoscopy (RA-LSC), conventional laparoscopy (LSC) and laparotomy. The functional unit was one endometrial staging procedure. Collected clinical data from the Albert Einstein College of Medicine, NY, USA, spanning the years 2008-2011, consisted of: age, body mass index (BMI), procedure type, operating time, history of prior abdominal surgery, length of stay, uterine weight, and surgical instruments used. The relevant outcome measures included climate change (defined as CO₂ footprint) and waste. CO₂ footprint was calculated by using an attributional LCA method (PAS 2050, GHG Protocol). Only energy and waste were included in this analysis. Data on energy and waste were obtained from previous studies, the National Energy Foundation and the US Energy Information Administration. Waste production (in kg) was determined based on operating room instrument data, specific to each modality. Disposal of 1 kg waste in a municipal landfill was considered equivalent to 1 kg CO₂ equivalents. Energy expenditure (in kWh) was calculated based on independent source data. Energy consumption was categorized in environmental, instrumental, robotic system and equipment. The total energy consumed per category was multiplied by the operative time, to establish a unique energy consumption value for each surgery. Emissions for transport and manufacturing of instruments and goods and for postoperative stay were not considered.

Thiel (2015) used a hybrid LCA framework (ISO 14040-44) to quantify the environmental emissions of four types of hysterectomy (vaginal, abdominal, laparoscopic, robotic). The functional unit of the study was one hysterectomy. This research used a hybrid LCA framework, by incorporating process LCA data and Economic Input Output LCA (EIO-LCA) data. Monte Carlo simulations were used to quantify the variability and uncertainty in emissions for each component of a hysterectomy. Waste (in kg) was deduced from the surgeries of patients that underwent an abdominal, a laparoscopic, or a robotic hysterectomy for noncancer related reasons. For one year, waste from 62 cases of each type of hysterectomy were quantified and characterized (17 laparoscopic and 15 abdominal, vaginal, and robotic). A life cycle inventory was conducted of the data collected through waste audits and site assessments, followed by the life cycle impact assessment, where environmental impacts for the inputs and outputs of the four types of hysterectomy were calculated for both process LCA and EIO-LCA. Relevant outcome measures included climate change (defined as CO₂ footprint), waste, acidification, eutrophication, human toxicity, ecotoxicity, and ozone depletion. The environmental impact of postoperative stay was not considered.

Results

1. Climate Change

All three studies (Power, 2012; Woods, 2015; Thiel, 2015) reported on CO₂ footprint.

Power (2012) reported that MIS had more CO₂ emissions (355,924 tonnes/year) in comparison to open surgery. Scope 1, gas that was used during MIS for insufflation, resulted in 303 tonnes of CO₂-emissions. Scope 2-3, CO₂ emissions (CO₂ capture/compression, transportation of CO₂ cylinders and incineration of disposable instruments) resulted in 355,621 tonnes of CO₂-emissions. This LCA of MIS, showed that the biggest contributor in CO₂ emissions is the capture/compression of CO₂ for insufflation. This is mainly caused by industrial gas manufacturing, followed by the actual use of the CO₂ for insufflation, the transportation of the gas, and the incineration of the disposable instruments.

Woods (2015) reported that the CO₂ footprint (based on waste and consumed energy) was 4,498 CO₂ equivalents for all 150 endometrial staging procedures. This corresponds with 30 kg CO₂ equivalents/patient. A robotic-assisted laparoscopy (RA-LSC) procedure resulted in a CO₂ footprint of 40.3 kg CO₂ equivalents/patient, a laparoscopy (LSC) procedure in 29.2 kg CO₂ equivalents/patient, and a laparotomy (LAP) in 22.7 kg CO₂ equivalents/patient. The CO₂ footprint of the RA-LSC represents a 38% and 77% increase over LSC and LAP, respectively. The emission of 40.3 kg CO₂ equivalents is comparable to a car ride from Amsterdam to Antwerp (±160 km, petrol car, 1:14). Comparing the three techniques, the RA-LSC has the biggest CO₂ footprint. Both energy use and waste contribute most to this CO₂ footprint. Due to the Da Vinci robot, the energy consumption of RA-LSC is the highest. The energy use for equipment is highest in the LSC group. Energy use adds more to the CO₂ footprint than waste.

Thiel (2015) reported a difference in CO₂ footprint between four types of hysterectomy. For all outcomes in this LCA, the surgical modality with the highest impact is defined as 100% and the other modalities are relatively compared to the modality with the highest impact. Robotic hysterectomies have the highest impact (100%), followed by the laparoscopic (65-70%), abdominal (35-40%) and vaginal modality (30-35%). In other words, the laparoscopic modality resulted in 30-35% less contribution to CO₂ footprint than a robotic modality. This LCA showed the biggest environmental impact is attributable to the robotic and laparoscopic surgical approaches for a hysterectomy. Within these modalities, the use of single-use instruments, single-use materials (e.g. gowns, gloves), and anesthetic gases are the biggest contributors.

2. Waste

All three studies (Power, 2012; Woods, 2015; Thiel, 2015) reported on waste. Power (2012) only reported a total of 208,441 kg/year from disposable trocar and laparoscopic instrument use for MIS. No other data regarding waste was provided in this study.

Woods (2015) reported that the LAP group produced 8.3 kg CO₂ equivalents of solid waste, the LSC group 11.2 kg CO₂ equivalents, and the RA-LSC group 11.2 kg CO₂ equivalents, respectively. Thus, RA-LSC represented a 74% and 36% increase over LAP and LSC, respectively. The solid waste consisted of four categories: infection control waste, single-use devices, consumable waste, and sterile wraps (see Table 1). The consumable waste was the largest contributor to solid waste for all surgical techniques. The largest amount of single-use device waste was generated by the LSC, followed by RA-LSC and LAP. The largest amount of waste for infection control was generated by the RA-LSC, followed by LAP and LSC. Sterile wraps contributed the least on waste for all surgical techniques.

Table 1. Types of waste described in Woods (2015)

Thiel (2015) reported that the robotic approach for a hysterectomy produced the highest amount of waste, following the laparoscopic, abdominal, and vaginal approaches. Robotic hysterectomies produced 13.7 kg municipal solid waste (MSW) per patient, which resulted in a total of 30% more waste compared to the other approaches.

This consisted of 50% gloves and other plastics, 22% by weight by gowns, drapes and bluewraps, 18% paper, and 5% cotton.

Abdominal hysterectomies had an average of 9.2 kg of MSW production and produced the largest amount of cotton waste per surgery. Across all four surgeries, the gowns, drapes and bluewraps were the majority of the MSW (robotic: 22%, laparoscopic: 35%). Other types of plastics, from thin film packaging wrappers to hard plastic trays, ranged from 36% MSW for vaginal hysterectomies to 46% for robotic hysterectomies.

3. Acidification

Only Thiel (2015) reported on acidification for the four surgical hysterectomies. The robotic approach had the highest impact on acidification (100%), followed by laparoscopic (70-75%), abdominal (25%), and vaginal approach (20%).

4. Eutrophication

Only Thiel (2015) reported a difference in eutrophication for the four surgical hysterectomies. The robotic approach had the highest impact on eutrophication (100%), followed by the laparoscopic (70-75%), abdominal (60-65%), and vaginal approach (55%).

5. Human toxicity

Only Thiel (2015) reported a difference in human toxicity for the four surgical hysterectomies. Human toxicity was divided into two subcategories: carcinogenic and non-carcinogenic potential. The robotic approach had the highest impact in both carcinogenic and non-carcinogenic potential categories (100%). Regarding the carcinogenic impact, the robotic approach was followed by the abdominal (90-100%), laparoscopic (80%), and vaginal approach (80%). Regarding the non-carcinogenic impact, the robotic approach was followed by the laparoscopic (85-90%), abdominal (80-90%), and vaginal approach (70-75%).

6.Ecotoxicity

Only Thiel (2015) reported a difference in ecotoxicity for the four surgical hysterectomies. The robotic approach had the highest impact on ecotoxicity (100%), followed by the laparoscopic (90-95%), abdominal (90%), and vaginal approach (75%). 

7. Ozone depletion

Only Thiel (2015) reported a difference in ozone depletion for the four surgical hysterectomies. The robotic approach had the highest impact on ozone depletion (100%), followed by the laparoscopic (60-65%), abdominal, and vaginal approach (0-5%).

Level of evidence of the literature

There are currently no widely recognized guidelines for designing, conducting, or reporting systematic reviews in LCA (Zumsteg, 2012). The Standardized Technique for Assessing and Reporting Reviews of LCA (STARR-LCA) checklist was developed based on the PRISMA statement (which was designed to help systematic reviewers transparently report why the review was done, what the authors did, and what they found) (Page, 2021; Zumsteg, 2012; Costa, 2019; Drew, 2021). The STARR-LCA proposed a critical appraisal tool to assess the level of evidence of LCAs (Drew, 2021) which is used in this module to provide an indication of the study quality (see Appendix 2 ‘Critical appraisal of LCAs’). This tool consists of 16 assessment criteria for the different phases of an LCA, covering a set of quality indicators (such as internal validity, external validity, consistency, transparency, bias). The score gives an indication of the current study quality (a higher score indicates a higher study quality).

The use of the GRADE approach in environmental and occupational health is relatively new, and likely to grow in coming years, as systematic reviews become more common in the field of LCAs and the limitations of expert-based narrative review methods are increasingly recognized (Aiassa, 2015; EFSA, 2010; Mandrioli, 2015; Morgan, 2016; Woodruff, 2014). Further research is warranted in this field (Morgan, 2019), which is acknowledged by the working group. Although standards for using GRADE for LCAs are lacking, the working group decided that the level of evidence starts for LCAs starts at grade high. After all, the working group is mainly interested in which interventions have the greatest impact on environmental sustainability and LCAs are the best method to assess this (see also the report: ‘Leidraad Duurzaamheid in richtlijnen’ (NVvH, 2023). Furthermore, the GRADE standards are followed as much as possible.

As the three included studies contained LCAs (Power, 2012; Woods, 2015; Thiel, 2015), the level of evidence started at grade high.

1. Climate Change

Three studies (Power, 2012; Woods, 2015; Thiel, 2015) reported on ‘climate change’. The level of evidence of this outcome measure was downgraded with 2 levels to low due to: risk of bias (-2; limitations on functional unit, unclear system boundaries or stages, missing system coverage, limited representativeness of data, limited transparency on characterization, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised).

2. Waste

Three studies (Power, 2012; Woods, 2015; Thiel, 2015) reported on ‘waste’. The level of evidence of this outcome measure was downgraded with 2 levels to low due to: risk of bias (-2; limitations on functional unit, unclear system boundaries or stages, missing system coverage, limited representativeness of data, limited transparency on characterization, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised).

3. Acidification

One study (Thiel, 2015) reported on ‘acidification’. The level of evidence of this outcome measure was downgraded with 3 levels to very low due to: risk of bias (-2; unclear system boundaries or stages, limited transparency on characterization, unclear reported results, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised) and indirectness (-1; limited representativeness as data were only collected from one study including 62 hysterectomy cases).

4. Eutrophication

One study (Thiel, 2015) reported on ‘eutrophication’. The level of evidence of this outcome measure was downgraded with 3 levels to very low due to: risk of bias (-2; unclear system boundaries or stages, limited transparency on characterization, unclear reported results, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised) and indirectness (-1; limited representativeness as data were only collected from one study including 62 hysterectomy cases).

5. Human toxicity

One study (Thiel, 2015) reported on ‘human toxicity’. The level of evidence of this outcome measure was downgraded with 3 levels to very low due to risk of bias (-2; unclear system boundaries or stages, limited transparency on characterization, unclear reported results, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised) and indirectness (-1; limited representativeness as data were only collected from one study including 62 hysterectomy cases).

6. Ecotoxicity

One study (Thiel, 2015) reported on ‘ecotoxicity’. The level of evidence of this outcome measure was downgraded with 3 levels to very low due to risk of bias (-2; unclear system boundaries or stages, limited transparency on characterization, unclear reported results, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised) and indirectness (-1; limited representativeness as data were only collected from one study including 62 hysterectomy cases).

7. Ozone depletion

One study (Thiel, 2015) reported on ‘ozone depletion’. The level of evidence of this outcome measure was downgraded with 3 levels to very low due to risk of bias (-2; unclear system boundaries or stages, limited transparency on characterization, unclear reported results, sensitivity/uncertainty analyses were lacking, limitations inadequately critically appraised) and indirectness (-1; limited representativeness as data were only collected from one study including 62 hysterectomy cases).

A systematic review of the literature was performed to answer the following question: What is the role of environmental sustainability of robot-assisted laparoscopic surgery compared with conventional laparoscopic surgery or open surgery?

P: Patients who underwent surgery

I: Robot-assisted surgery

C: Conventional laparoscopic surgery or open surgery

O: Climate change (CO₂ footprint/Global Warming Potential (GWP)), waste, acidification, eutrophication, human toxicity, ecotoxicity, ozone depletion

Relevant outcome measures

Life cycle assessment (LCA) is a methodological tool used to quantitatively analyse the life cycle of products/activities within the context of environmental impact. The assessment comprises all stages needed to produce and use a product, from the initial development to the treatment of waste (the total life cycle). An LCA is mainly based on four phases: 1) goal and scope definition, 2) inventory analysis, 3) impact assessment, and 4) interpretation. The third phase is the life cycle impact assessment (LCIA), in which emissions and resource extractions are translated into a limited number of environmental impact scores by means of so-called characterisation factors. The ReCiPe model is a method for the impact assessment in an LCA (Huijbregts, 2016, Huijbregts, 2017). To determine the outcome measures regarding environmental impact, the ReCiPe model of the National Institute for Public Health and the Environment (in Dutch: Rijksinstituut voor Volksgezondheid en Milieu, RIVM) was used.

The outcomes determined by the working group are based on the ReCiPe model. The working group considered climate change (CO₂ footprint/Global Warming Potential) and waste as critical outcome measures for decision making; and acidification, eutrophication, human toxicity, ecotoxicity and ozone depletion as important outcome measures for decision making.

A priori, the working group did not define the outcome measures listed above but used the definitions used in the studies.

Outcomes focused on environmental life cycle assessment (LCA) impact categories are relatively new in healthcare. Given the variety in scopes and methods of performing and reporting LCAs, the working group did not define a priori the minimal important difference. Differences between the techniques were evaluated by the working group after data extraction.

Glossary

• Acidification: Reduction of the pH due to acidifying effects of emissions. Acid deposition of acidifying contaminants on soil, groundwater, surface waters, biological organisms, ecosystems, and substances. Expressed in SO₂ (sulfur dioxide) equivalents (Acero, 2015).
• Climate change: In the outcome category “climate change” we include two types of outcome measures:
  • CO₂ footprint: The total greenhouse gas (GHG) emissions caused by an individual, organization, event, or product given a period of time. Expressed in CO₂ equivalents adopting the GWP (The Carbon Trust, 2018).
  • Global Warming Potential (GWP): A measure of how much energy 1 kg of a specific GHG will absorb over a given period of time, relative to the emission of 1 kg of carbon dioxide (CO₂). The GWP is 1 for CO₂. GWP was developed to allow comparisons of the global warming impacts of different gases (EPA, 2021). The larger the GWP, the more that a given gas contributes to climate change (the greater the global warming effect of the gas).
• Economic input-output life cycle assessment (EIO-LCA): EIO-LCA uses annual input–output economy models and links monetary values of the industry sector (such as building sector) to their environmental inputs/outputs.
• Ecotoxicity: Toxic effects of chemicals on ecosystems. Environmental toxicity is measured as three separate impact categories which examine freshwater (e.g. lakes and rivers), marine (e.g. estuaries and the ocean) and land. Ecotoxicity describes exposure and the effects of toxic substances on the environment (Acero, 2015). Common environmental toxicants are diethyl phthalate (enters environment through industries manufacturing cosmetics, plastic and many other commercial products), bisphenol A (found in many mass-produced products such as, medical devices, food packaging, cosmetics, computers), pesticides, and oil. It is expressed in 1,4-dichlorobenzene (DB) equivalents.
• Energy mix: All sources of energy used from which energy is produced that can be directly used, e.g. in the form of electricity. Sources can be coal, oil, natural gas, hydro, nuclear and other renewables (e.g. wind, solar, thermal).
• Eutrophication: Accumulation of nutrients in water. Eutrophication includes the effects on ecosystems found in land or water due to over-fertilization or excess supply of nutrients (e.g. algal bloom), which can lead to changes in ecosystems and diversity of species. Expressed in PO₄ (phosphate) equivalents (Acero, 2015). This leads to ecosystem damage and has a negative effect on the climate.
• Hotspot: In a life cycle assessment (LCA) an "environmental hotspot" refers to a specific stage where a significant environmental impact is detected, presenting a significant opportunity for improvement.
• Human toxicity: Toxic effect of chemicals on humans. Human toxicity reflects the potential harm of a unit of chemical released into the environment, and it is based on the inherent toxicity of a compound and its potential dose (Acero, 2015). It is differentiated as non-cancer and cancer toxicity and affects human health. It is expressed in 1,4-dichlorobenzene (DB) equivalents.
• Hybrid instruments: Instruments which are partly disposable and partly reusable.
• Life Cycle Assessment (LCA): This is a methodology for assessing environmental impacts associated with all the stages of the life cycle of a commercial product, process, or service. An LCA mainly consists of four steps: 1) goal and scope, 2) inventory analysis, 3) impact assessment, and 4) interpretation.
• Life Cycle Impact Assessment (LCIA): This is the third phase of an LCA, which aims to evaluate the potential environmental and human health impacts resulting from the elementary flows determined in the LCI.
• Ozone depletion: Reduction of the stratospheric ozone layer due to emissions of ozone depleting substances. Damage to the ozone layer reduces its ability to prevent ultraviolet (UV) light entering the earth’s atmosphere, increasing the amount of carcinogenic UV light reaching the earth’s surface. Expressed in kg CFC-11 equivalents (Acero, 2015).
• Waste: Total amount of waste expressed in kilograms, such as for example plastics or (disposable) instruments.

Search and select (Methods)

The databases Pubmed (via NCBI), Embase (via OVID), Web of Science (via Webofscience), Cochrane (via Cochrane library) and Emcare (via OVID) were searched with relevant search terms from 2000 until 7 December 2021. The detailed search strategy is depicted under the tab Methods. The systematic literature search resulted in 202 hits. Studies for this module were selected based on the following criteria:

• Systematic reviews in which searches were performed in at least two databases, with a detailed search strategy, risk of bias assessment and results of individual studies available, randomized controlled trials, (observational) comparative studies, Life Cycle Assessments;
• Full-text English or Dutch language publication; and
• Studies according to the PICO. This included studies that compared robot-assisted surgery with conventional laparoscopic or open surgery and included at least one of the outcomes conform the PICO.

Three studies were included for full text analysis. After reading the full text, all three studies were included in the literature summary of this module.

Results

Three studies were included in the analysis of the literature, which were all Life Cycle Assessments (LCAs). Important study characteristics and results are summarized in Appendix 1 ‘Evidence table of LCAs’. The quality assessment of the studies is summarized in Appendix 2 ‘Critical appraisal of LCAs’.