Abstract
Background: Although robotic surgery has several advantages over other minimally invasive surgery (MIS) techniques for rectal cancer surgery, the uptake in Canada has been limited owing to a perceived increase in cost and lack of training. The objective of this study was to determine the impact of access to robotic surgery in a Canadian setting.
Methods: We conducted a retrospective cohort study involving consecutive adults undergoing surgical resection for rectal cancer between 2017 and 2020. The primary exposure was access to robotic surgery. Outcomes included MIS utilization, short-term outcomes, total cost of care, and quality of surgical resection. We completed univariate and multivariate analyses.
Results: We included 171 individuals in this cohort study (85 in the prerobotic period and 86 in the robotic period). The 2 groups had similar baseline characteristics. A higher proportion of individuals underwent successful MIS in the robotic phase (86% v. 46%, p < 0.001). Other benefits included a shorter mean length of hospital stay (5.1 d v. 9.2 d, p < 0.001). The quality of surgical resection was similar between groups. The total cost of care was $16 746 in the robotic period and $18 808 in the prerobotic period (mean difference −$1262, 95% confidence interval −$4308 to $1783; p = 0.4).
Conclusion: Access to robotic rectal cancer surgery increased successful completion of MIS and shortened hospital stay, with a similar total cost of care. Robotic rectal cancer surgery can enhance patient outcomes in the Canadian setting.
Laparoscopic surgery for colorectal cancer has several advantages over open techniques, including decreased postoperative pain, shorter length of stay, decreased morbidity, and improved quality of life.1,2 Although there were initial concerns regarding the safety of laparoscopic surgery, similar long-term oncologic outcomes have been found when compared with open techniques. 2,3 Despite the well-described benefits, adoption of laparoscopic colorectal cancer surgery has been slower in Canada than in other settings,4 with only 52% of those with colon cancer and 34% of those with rectal cancer having undergone a laparoscopic resection between 2014 and 2015.5,6 Barriers to adopting laparoscopic surgery include a lack of formal training, inadequate operating room time, and, in the case of rectal cancer, technical and ergonomic challenges of operating in the pelvis.7–10
Robotic surgery is an alternative to laparoscopic surgery. Using articulating tools, 3D vision of the anatomy, and greater precision in dissection, the field of robotic surgery has addressed the many technical issues of laparoscopic rectal surgery. Barriers to implementation of robotic surgery within a Canadian setting primarily relate to a lack of training and perceived additional costs of robotic surgery.11 In a recent report, our group questioned whether robotics increased costs when used routinely for benign and malignant colorectal disease.12 A criticism of this study12 was the overly broad inclusion criteria, as sigmoid resections were included, as was benign pathology. Many have advocated that in resource-limited settings, robotics could be considered primarily for the surgical management of rectal cancer.13,14 Thus, the objective of this study was to assess the impact of robotic surgery for rectal cancer at a regional cancer centre in Canada. We aimed to assess changes in the utilization of minimally invasive surgery (MIS), short-term outcomes, total cost of care, and quality of dissection after robotic surgery was available to patients with rectal cancer.
Methods
We conducted a retrospective cohort study involving consecutive individuals undergoing rectal cancer surgery between 2017 and 2020 at Kingston Health Sciences Centre, in Kingston, Canada. Kingston Health Sciences Centre is a regional cancer centre with a catchment area of approximately 495 000. The 3 participating surgeons are fellowship-trained colorectal surgeons, whose elective practice is exclusively colorectal pathology. One surgeon (S.V.P.) trained at an accredited American fellowship with dedicated robotic training. The other 2 surgeons (P.H.M. and A.C.-M.) did not receive robotic training during fellowship. Proctoring was provided for the first 2 cases. Our team has a well-established protocol for enhanced recovery after surgery, which has been in place for about 10 years. In addition, consistent discharge criteria have been used as part of this protocol. We included consecutive individuals with rectal cancer undergoing either low anterior resection or abdominoperineal resection for rectal cancer. Those with recurrent rectal cancer, those requiring multivisceral resection, or those with a previous colon resection were excluded, as this group was ineligible for MIS at our centre.
This study complies with the Strengthening the Reporting of Observational Studies in Epidemiology requirements.
Study variables
The primary exposure for this study was access to robotic surgery, and categorized as prerobotic or robotic period. The robotic surgery platform was available for use beginning on Mar. 1, 2019. Before this date, individuals were offered laparoscopic surgery or open surgery. After this date, individuals were offered robotic, laparoscopic, or open surgery. Selection of surgical approach (i.e., robotic, laparoscopic, or open) was at the discretion of the operating surgeon and determined based on an individual’s previous abdominal surgeries, body habitus, and anatomic variation, as well as the location and extent of the rectal cancer. An a priori sensitivity analysis was planned to compare the surgical approaches directly (i.e., robotic, laparoscopic, or open surgery). For the sensitivity analysis assessing surgical approach, conversions were included in the group they were originally a part of (i.e., laparoscopic or robotic surgery) and not the open group.
The primary outcomes of this study included the successful completion of MIS, total length of hospital stay, operative time, short-term complications, total hospital costs, and quality of dissection. Success of MIS was defined as an attempt and completion of MIS, without conversion to open surgery. Minimally invasive surgery included robotic or laparoscopic surgery. Conversion was defined as an open incision used for a purpose other than specimen extraction. We considered patients with a preplanned open pelvic dissection to be in the open surgery category. Total length of hospital stay included the index in-patient hospital stay (in days) plus any subsequent hospital days from readmissions within 30 days. We assessed the following 30-day complications: readmission for any reason, emergency department visit, anastomotic leak, and unplanned reoperation. An emergency department visit was counted separately from a readmission. We defined an anastomotic leak as that requiring intervention (i.e., antibiotics, percutaneous drain insertion, or operative intervention).
Total hospital costs included the costs of the index hospital stay and any costs associated with an emergency department visit or readmission. Costs were obtained from the Ontario Case Costing submission data. Total hospital cost includes those incurred during an inpatient stay as well as a portion of indirect costs, attributable based on workload measures. Operating room costs included patient-specific supplies, direct nursing and other labour costs, and attributable indirect operating costs. Hospital stay costs included in-patient nursing and other labour costs (i.e., allied health), nutrition and food services, laboratory, medical imaging, pharmacy, intensive care unit costs (if applicable), and any other direct costs and attributable indirect costs. The calculated total hospital costs comply with Ontario Health’s requirements to determine funding rates for quality-based procedures. Costs are calculated at the end of the fiscal year. We did not include the capital cost of the robotic platform as this was funded entirely through philanthropic donation. Similarly, service contract costs for robotic or laparoscopic equipment, and the capital costs of laparoscopic towers or other equipment used for laparoscopic or open procedures were not included.
The quality of surgical excision was based on the quality of the total mesorectal excision and circumferential radial margin positivity. We defined circumferential radial margin positivity as cancer less than 1 mm from the border of the specimen.15
We identified patient demographic characteristics, including age, sex, body mass index, and American Society of Anesthesiologists classification. Pretreatment stage was identified from the synoptic magnetic resonance imaging report, which was completed in every included case. We collected type of resection (low anterior resection v. abdominoperineal resection), use of ostomy, and surgical approach (laparoscopic v. robotic v. open). Patients undergoing laparoscopic-assisted transanal total mesorectal excision were included in the laparoscopic surgery category.
Data sources and analysis
We collected most data retrospectively through the electronic medical record. The exception was costing data, which were provided by the finance department, based on required data submission to Cancer Care Ontario, Ontario Health. Surgical details were extracted from both the surgeon’s operative note and the corresponding perioperative data information report. We identified inpatient hospital details, including length of stay, readmission, and emergency department visits through the electronic medical record. Readmission or emergency department visits at another hospital within the province were included and identified through the provincial electronic medical record. We extracted diagnostic imaging and pathology data from synoptic reports created for the management and reporting of rectal cancer.
We completed all statistical analysis in Stata 16.1 (StataCorp LLC). Continuous variables were reported as means with standard deviations or medians with interquartile ranges. We determined difference in means and significance using either the Student t test (2 variables) or linear regression (≥ 3 variables). We calculated differences between proportions using the χ2 test. Linear regression was used for multivariate analysis.
Ethics approval
Research ethics approval was obtained from the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board.
Results
Between April 2017 and December 2020, 191 individuals underwent surgical resection for rectal cancer. Twenty individuals were excluded for recurrent rectal cancer, multivisceral resection, or previous colon cancer resection, leaving 171 individuals included in this study. A total of 85 individuals were included in the prerobotic period, which included 56 laparoscopic procedures (traditional n = 42; transanal total mesorectal excision n = 14) and 29 open procedures. In the robotic period, 86 individuals were included, which included 75 robotic procedures, 6 laparoscopic procedures (traditional n = 4; transanal total mesorectal excision n = 2), and 5 open procedures (Figure 1).
Baseline characteristics, including demographics, surgery type, and neoadjuvant therapy, were similar between groups (Table 1). With comparison of participants by surgical approach (robotic v. laparoscopic v. open), patient characteristics and disease characteristics were also similar (Table 2).
Minimally invasive surgery was attempted more frequently during the robotic period (prerobotic 66% v. robotic 93%, p < 0.001), with fewer conversions to open procedures (prerobotic 17/56 [30%] v. robotic 6/80 [8%], p < 0.001). Successful MIS occurred more frequently during the robotic period than during the prerobotic period (prerobotic 46% v. robotic 86%, p < 0.001) (Table 3). Comparing robotic with laparoscopic surgery, fewer conversions occurred in the robotic group (8% v. 27%, p = 0.03) (Table 4).
The mean total length of stay was reduced in the robotic period (9.2 v. 5.1 d, p < 0.001), and the mean operative time was similar (4.3 v. 4.3 h, p = 0.9). Comparing the techniques directly showed that the mean length of hospital stay was lowest in the robotic surgery group (5.1 d) followed by the laparoscopic surgery group (6.9 d) and the open surgery group (12.3 d, p < 0.001). Operative time was lowest in the open surgery group (3.1 h), followed by the robotic surgery group (4.4 h) and the laparoscopic surgery group (4.9 h, p < 0.001) (Table 4).
The risks of emergency department visit (38% v. 33%, p = 0.5), readmission to hospital (17% v. 22%, p = 0.4), unplanned reoperation (5% v. 7%, p = 0.4), and anastomotic leak (11% v. 5%, p = 0.3) were similar between the prerobotic and robotic period (Table 3). The risk of these complications was also similar between robotic, laparoscopic, and open approaches (Table 4).
We found that the quality of the total mesorectal dissection (p = 0.4) and margin positivity (1.3% v. 2.3%, p = 0.6) was similar between the prerobotic and robotic period (p = 0.4). We did not observe a difference in the completeness of the mesorectal dissection or margin positivity when the techniques were compared directly (Table 4).
The mean cost of care was $16 746 in the robotic period and $18 808 in the prerobotic period (mean difference −$1262, 95% confidence interval [CI] −$4308 to $1783; p = 0.4) (Table 3). Comparing total mean costs of the techniques directly did not demonstrate a statistically significant difference (robotic surgery $17 329 v. laparoscopic surgery $15 910 v. open surgery $20 048, p = 0.2) (Table 4). An adjusted analysis was completed, adjusting for age, BMI, receipt of neoadjuvant therapy, and sex (Table 5). This analysis demonstrated that robotics was not associated with a difference in total cost of care (−$497, 95% confidence interval −$3265 to $2270, p = 0.7). Sensitivity analysis comparing robotic with laparoscopic and open approaches was also completed (Table 6). No difference in cost was found on this analysis.
Discussion
This study reports the clinical and cost impact of adopting robotic surgery into routine use for rectal cancer at a Canadian cancer centre. We found that access to robotic surgery resulted in an increase in successful completion of MIS and a reduced length of stay, without an increase in the total cost of care.
The present study has similar findings to those of several published studies from other settings. A meta-analysis from 201816 included 5 randomized controlled trials published between 2008 and 2017. This meta-analysis included 671 patients (robotic n = 334, laparoscopic n = 337), with most (n = 471) being from the Robotic vs Laparoscopic Resection for Rectal Cancer trial.17 Pooled analysis found a lower conversion risk (pooled risk robotic surgery 7.3% v. laparoscopic surgery 12.9%) with similar perioperative complications, duration of hospital stay, and quality of excision. The recently published Robotic Versus Laparoscopic Surgery for Middle and Low Rectal Cancer trial (not included in the above meta-analysis)18 compared robotic (n = 620) and laparoscopic (n = 620) surgery for middle and low rectal cancer. This study found a lower risk of conversion to open surgery, and shorter length of hospital stay in the robotic group, with similar other short-term outcomes between groups.
An important limitation of the above studies, when considering real-world utilization and generalizability, were the inclusion criteria. To be included in any of these trials, an individual must have been eligible for either a robotic approach or a laparoscopic approach, which does not reflect real-world practice. A systematic review of 41 observational studies identified differences between patients offered robotic versus laparoscopic rectal cancer surgery.19 Those undergoing robotic surgery were older, more likely to be male, more likely to have undergone neoadjuvant therapy, and had more distal tumours than those undergoing laparoscopic surgery. Each of these factors has been identified as a risk factor for technical complexity and increased conversion to open surgery, and often preclude a laparoscopic approach.20–23 Thus, the benefits of robotic surgery are likely underestimated in the above-cited trials. Our study results confirm this hypothesis, as shown by a larger proportion of patients to be offered MIS (93% v. 66%) with fewer conversions to open surgery (8% v. 30%), and substantial reductions in length of stay.
The comparative cost between robotic rectal surgery and other techniques in a Canadian setting has previously been assessed by Ramji and colleagues.24 This study found an increased cost in the robotic surgery group ($18 273 v. laparoscopic $11 493 v. open $12 558), which differs from our results. The difference in findings is most likely explained by the robotic system used and the technical approach. Ramji and colleagues used an older robotic platform, which was limited in its ability to operate in more than 1 quadrant. For this reason, a hybrid approach was used to allow for both pelvic (completed robotically) and left upper quadrant (completed laparoscopically) dissection, which incurred the full costs of both laparoscopic and robotic surgery. We used a newer platform that allowed for multiquadrant robotic surgery and hence did not incur laparoscopic surgery expenditures.
The design of this study was its primary strength. By comparing the period before and after the establishment of a robotic program, we were able to evaluate the influence of access on real-world patient outcomes. This approach helps mitigate the selection bias and performance bias that can exist when comparing laparoscopic with robotic surgery directly.25 Individuals eligible for robotic surgery may not be eligible for laparoscopic surgery (and thus undergo an open approach), and those eligible for laparoscopic surgery are potentially more likely to have favourable anatomy and outcomes.
In this study, we comprehensively assessed the effect of robotic surgery on a rectal cancer program at a Canadian regional cancer centre. Unlike previous publications, the current study offers a real-world assessment of the impact of access to the robotic platform on clinical outcomes and costs. We have demonstrated that access to robotic surgery results in increased utilization of MIS and improved clinical outcomes without the increased costs identified in previous studies.
Limitations
This is an observational study at risk of selection bias and confounding. We attempted to mitigate the risk of selection bias by including consecutive and unselected patients undergoing robotic surgery before and after robotic surgery was available. To mitigate confounding, we completed an adjusted analysis of cost of care. Similar to the unadjusted analysis, no differences in total costs was found based on period (robotic v. prerobotic) or approach (robotic v. laparoscopic v. open).
Conclusion
We have shown that access to robotic surgery significantly increases the successful completion of MIS and reduces hospital length of stay, with no increase in the total cost of care in patients with rectal cancer. The results of this study can be applied to similar settings across Canada.
Acknowledgement
The authors acknowledge Kathleen Wattie Barnett for her support in providing costing data for this paper.
Footnotes
This work was presented at the Canadian Surgery Forum, September 2022, Toronto, Canada.
Contributors: Sunil Patel, Lisa Zhang, Shaila Merchant, Antonio Caycedo-Marulanda, and P. Hugh MacDonald conceived and designed the study. Vanessa Wiseman and Sunil Patel acquired the data. Sunil Patel analyzed the data. All authors interpreted the data. Sunil Patel drafted the article, which all authors revised. All authors gave final approval of the version to be published and agreed to be accountable for all aspects of the work.
Data sharing: Researchers may contact the corresponding author to request access to the data used in this work. Reasonable requests will be considered, and data will be made available in Excel format.
Competing interests: Sunil Patel reports that Intuitive Surgical and Johnson & Johnson have provided the Department of Surgery at Queen’s University funding for a colorectal surgery fellow. No other competing interests were declared.
- Accepted January 4, 2023.
This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY-NC-ND 4.0) licence, which permits use, distribution and reproduction in any medium, provided that the original publication is properly cited, the use is noncommercial (i.e., research or educational use), and no modifications or adaptations are made. See: https://creativecommons.org/licenses/by-nc-nd/4.0/