Can better surgical outcomes be obtained in the learning process of robotic rectal cancer surgery? A propensity score-matched comparison between learning phases

Evaluation of an established colorectal robotic programme at an NHS district general hospital: audit of outcomes and systematic review of published data

INTRODUCTION

Robotic-assisted surgery (RAS) offers potential advantages over traditional surgical approaches. This study aimed to assess outcomes from a district general hospital (DGH) robotic colorectal programme against published data.

MATERIALS AND METHODS

The robotic programme was established following simulator, dry/wet lab training, and proctoring. We performed a case series analysing technical, patient, and oncological outcomes extracted from a prospective database of colorectal RAS cases (2015-2022). A registered systematic review (PROSPERO CRD42022300773; PubMed, Web of Science, EMBASE) of single-centre colorectal series from established robotic centres (n>200 cases) was completed and compared to local data using descriptive summary statistics. Risk of bias assessment was performed using an adapted version of the Cochrane ROBINS-I tool.

RESULTS

Two hundred thirty-two RAS cases were performed including 122 anterior resections, 56 APERs, 19 rectopexies, and 15 Hartmann's procedures. The median duration was 325 (IQR 265-400) min. Blood loss was < 100 ml in 97% of cases with 2 (0.9%) cases converted to open. Complications (Clavien-Dindo 3-5) occurred in 19 (8%) patients, with 3 (1.3%) deaths in < 30 days. Length of stay was 7 (IQR 5-11) days. In 169 rectal cancer cases, there were 9 (5.3%) cases with a positive circumferential or distal margin and lymph node yield of 17 (IQR 13-24). A systematic review of 1648 abstracts identified 13 studies from established robotic centres, totaling 4930 cases, with technical, patient, and oncological outcomes comparable to our own case series.

CONCLUSIONS

Outcomes from our robotic colorectal programme at a UK DGH are comparable with the largest published case series from world-renowned centres. Training and proctoring together with rolling audit must accompany the expansion of robotic surgery to safeguard outcomes.

Learning curve of robotic rectal surgery using risk-adjusted cumulative summation: a 5-year institutional experience

PURPOSE

Outline learning phases of robot-assisted laparoscopic surgery for rectal cancer and compare surgical and clinical outcomes between each phase of robot-assisted laparoscopic surgery and the mastery phase of conventional laparoscopic surgery.

METHODS

From 2015 to 2020, 210 patients underwent rectal cancer surgery at Sendai Medical Center. We performed conventional laparoscopic surgery in 110 patients and, laparoscopic surgery in 100 patients. The learning curve was evaluated using the cumulative summation method, risk-adjusted cumulative summation method, and logistic regression analysis.

RESULTS

The risk-adjusted cumulative summation learning curve was divided into three phases: phase 1 (cases 1-48), phase 2 (cases 49-80), and phase 3 (cases 81-100). Duration of hospital stay (13.1 days vs. 18.0 days, respectively; p = 0.016) and surgery (209.1 min vs. 249.5 min, respectively; p = 0.045) were significantly shorter in phase 3 of the robot-assisted laparoscopic surgery group than in the conventional laparoscopic surgery group. Blood loss volume was significantly lower in phase 1 of the robot-assisted laparoscopic surgery group than in the conventional laparoscopic surgery group (17.7 ml vs. 79.7 ml, respectively; p = 0.036). The International Prostate Symptom Score was significantly lower in the robot-assisted laparoscopic surgery group (p = 0.0131).

CONCLUSIONS

Robot-assisted laparoscopic surgery for rectal cancer was safe and demonstrated better surgical and clinical outcomes, including a shorter hospital stay, less blood loss, and a shorter surgical duration, than conventional laparoscopic surgery. After experience with at least 80 cases, tactile familiarity can be acquired from visual information only (visual haptic feedback).

CLINICAL TRIAL REGISTRATION

UMIN reference no. UMIN000019857.

Assessing the learning curve of robot-assisted total mesorectal excision: a multicenter study considering procedural safety, pathological safety, and efficiency

PURPOSE

Evidence regarding the learning curve of robot-assisted total mesorectal excision is scarce and of low quality. Case-mix is mostly not taken into account, and learning curves are based on operative time, while preferably clinical outcomes and literature-based limits should be used. Therefore, this study aims to assess the learning curve of robot-assisted total mesorectal excision.

METHODS

A retrospective study was performed in four Dutch centers. The primary aim was to assess the safety of the individual and institutional learning curves using a RA-CUSUM analysis based on intraoperative complications, major postoperative complications, and compound pathological outcome (positive circumferential margin or incomplete TME specimen). The learning curve for efficiency was assessed using a LC-CUSUM analysis for operative time. Outcomes of patients before and after the learning curve were compared.

RESULTS

In this study, seven participating surgeons performed robot-assisted total mesorectal excisions in 531 patients. Learning curves for intraoperative complications, postoperative complications, and compound pathological outcome did not exceed predefined literature-based limits. The LC-CUSUM for operative time showed lengths of the learning curve ranging from 12 to 35 cases. Intraoperative, postoperative, and pathological outcomes did not differ between patients operated during and after the learning curve.

CONCLUSION

The learning curve of robot-assisted total mesorectal excision based on intraoperative complications, postoperative complications, and compound pathological outcome did not exceed predefined limits and is therefore suggested to be safe. Using operative time as a surrogate for efficiency, the learning curve is estimated to be between 12 and 35 procedures.

Parallel, component training in robotic total mesorectal excision

There has been widespread adoption of robotic total mesorectal excision (TME) for rectal cancer in recent years. There is now increasing interest in training robotic novice surgeons in robotic TME surgery using the principles of component-based learning. The aims of our study were to assess the feasibility of delivering a structured, parallel, component-based, training curriculum to surgical trainees and fellows. A prospective pilot study was undertaken between January 2021 and May 2021. A dedicated robotic training pathway was designed with two trainees trained in parallel per each robotic case based on prior experience, training grade and skill set. Component parts of each operation were allocated by the robotic trainer prior to the start of each case. Robotic proficiency was assessed using the Global Evaluative Assessment of Robotic Skills (GEARS) and the EARCS Global Assessment Score (GAS). Three trainees participated in this pilot study, performing a combined number of 52 TME resections. Key components of all 52 TME operations were performed by the trainees. GEARS scores improved throughout the study, with a mean overall baseline score of 17.3 (95% CI 15.1-1.4) compared to an overall final assessment mean score of 23.8 (95% CI 21.6-25.9), p = 0.003. The GAS component improved incrementally for all trainees at each candidate assessment (p < 0.001). Employing a parallel, component-based approach to training in robotic TME surgery is safe and feasible and can be used to train multiple trainees of differing grades simultaneously, whilst maintaining high-quality clinical outcomes.

Learning curves for minimally invasive total mesorectal excision beyond the competency phase - a risk-adjusted cumulative sum analysis of 1000 rectal resections

AIM

The learning curve of total mesorectal excision (TME) by minimally invasive surgery (MIS) beyond the competency phase has not been adequately reported with large numbers or using a statistical control limit. The aim of this work was to study the learning curve of MIS TME in the proficiency phase.

METHOD

Risk-adjusted (RA) cumulative sum (CUSUM) and RA Bernoulli CUSUM charts were plotted for sequential MIS TME performed by a surgical team over 1000 cases. Surgical failure, a composite endpoint of conversions, complications of grade IIIA or above, R1 resections and inadequate nodal yield were used to monitor the performance.

RESULTS

The RA CUSUM detected an inflection point around the 600th operation. Two peaks were identified that could be traced back to probable causes of surgical failure. Similar inflection points were detected at the 450th case for laparoscopic TME and the 367th case for sphincter preservation. No single definite threshold point was noticed for robotic or abdominoperineal operations. At no point did the curves cross the safety threshold. The probability of surgical failure reduced with increasing experience in the multivariate regression (OR 0.899, p = 0.000). This association persisted irrespective of the surgical approach (laparoscopic versus robotic) or the type of operation (sphincter preservation versus abdominoperineal resection).

CONCLUSION

The learning curves for MIS TME did not cross the safety threshold beyond the competency phase. However, a 10% reduction of relative risk in surgical failure was observed for every 100 cases operated on.

The learning curve of laparoscopic, robot-assisted and transanal total mesorectal excisions: a systematic review

BACKGROUND

The standard treatment of rectal carcinoma is surgical resection according to the total mesorectal excision principle, either by open, laparoscopic, robot-assisted or transanal technique. No clear consensus exists regarding the length of the learning curve for the minimal invasive techniques. This systematic review aims to provide an overview of the current literature regarding the learning curve of minimal invasive TME.

METHODS

A systematic literature search was performed. PubMed, Embase and Cochrane Library were searched for studies with the primary or secondary aim to assess the learning curve of either laparoscopic, robot-assisted or transanal TME for rectal cancer. The primary outcome was length of the learning curve per minimal invasive technique. Descriptive statistics were used to present results and the MINORS tool was used to assess risk of bias.

RESULTS

45 studies, with 7562 patients, were included in this systematic review. Length of the learning curve based on intraoperative complications, postoperative complications, pathological outcomes, or a composite endpoint using a risk-adjusted CUSUM analysis was 50 procedures for the laparoscopic technique, 32-75 procedures for the robot-assisted technique and 36-54 procedures for the transanal technique. Due to the low quality of studies and a high level of heterogeneity a meta-analysis could not be performed. Heterogeneity was caused by patient-related factors, surgeon-related factors and differences in statistical methods.

CONCLUSION

Current high-quality literature regarding length of the learning curve of minimal invasive TME techniques is scarce. Available literature suggests equal lengths of the learning curves of laparoscopic, robot-assisted and transanal TME. Well-designed studies, using adequate statistical methods are required to properly assess the learning curve, while taking into account patient-related and surgeon-related factors.

Learning Curve of Robotic-Assisted Total Mesorectal Excision for Rectal Cancer

Background

Although some studies have assessed the learning curve of robotic-assisted total mesorectal excision for rectal cancer, most studies included limited sample sizes, no study used postoperative complications as an independent variable to analyze the learning curve of robotic rectal surgery, and no study evaluated the influence of the learning curve on long-term oncologic outcomes.

Methods

Clinical data on consecutive patients who underwent robotic-assisted total mesorectal excision for rectal cancer by a single surgeon between January 2015 and December 2018 at the First Affiliated Hospital of Nanchang University were retrospectively collected. The cumulative sum (CUSUM) and risk-adjusted cumulative sum (RA-CUSUM) were used to visualize the learning curve of operation time and postoperative complications (CD ≥ grade II). Comparisons of clinical outcomes at different learning phases analyzed by RA-CUSUM were performed after propensity score matching.

Results

A total of 389 consecutive patients were included in the analysis. The numbers of patients needed to overcome the learning curves of operation time and postoperative complications of robotic-assisted laparoscopic surgery for rectal cancer were 34 and 36, respectively. The learning process was divided into two phases based on RA-CUSUM: the learning phase (1st-36th cases) and the mastery phase (37th-389th cases). Before matching, the mastery phase had more patients with older age, lower tumor location, and neoadjuvant therapy. After matching, the two phases exhibited similar characteristics. The operation time, intraoperative blood loss, postoperative hospital stay, and postoperative complications in the mastery phase were reduced compared with the learning phase, with a median follow-up of 35 months, and the long-term oncologic outcomes were not significantly different between the two phases.

Conclusions

An experienced laparoscopic surgeon initially implements robotic-assisted total mesorectal excision for rectal cancer, surgical outcomes improved after 36 cases, and the learning curve seemingly did not have an obvious impact on long-term oncologic outcomes.

Malfunction Events in the US FDA MAUDE Database: How Does Robotic Gynecologic Surgery Compare with Other Specialties?

STUDY OBJECTIVE

To review malfunction events (MEs) related to the use of the da Vinci robot reported to the United States Food and Drug Administration Manufacturer and User Facility Device Experience in the last 10 years and compare gynecologic surgery with other surgical specialties.

DESIGN

A retrospective review.

SETTING

Manufacturer and User Facility Device Experience database.

PATIENTS

Reports from 2010 to 2020 with keywords "Davinci" and "Intuitive".

INTERVENTIONS

Report review.

MEASUREMENTS AND MAIN RESULTS

There were 679 reports included in the final analysis. Most MEs occurred intraoperatively (81.7%) and were related to robotic instrument malfunction (84.5%), and 30% required an instrument switch to complete the procedure. Conversion to open and laparoscopic surgery was required in 3.1% and 1.3% of MEs, respectively. Injury to the patient occurred in 15.6% of MEs. Of the reported injuries to patients, 6.6% were related to robotic malfunction, 49.2% to instrument malfunction, and 18% to surgeon or staff misuse of the robotic system, and 15.6% were complications inherent to the procedure, not related to the robotic system. Of all the reported MEs, 4.4% were related to robot malfunction, 1.5% to console malfunction, 73.3% to Intuitive accessory malfunction, 11.2% to other accessory malfunction, 4% to surgeon or staff misuse of robotic system, and 3% to complications inherent to the procedure. Comparison between gynecologic surgery and other surgical specialties showed that 14.4% of issues were solved intraoperatively in gynecologic surgery vs 13.7% in other specialties (p = .185). The procedure was completed robotically in 85.2% in gynecologic surgery vs 84% in other specialties, laparoscopically 4.6% vs 3.7%, and open in 10.2% vs 12.4%, respectively (p = .883). In gynecologic surgery, reported MEs were made by patients in 14.8% vs 4.8% in other specialties, manufacturer in 78.4% vs 74.2%, and operating room staff in 2.3% vs 16.1%, respectively (p = .007). Injury to patient was similar in gynecologic surgery compared with other specialties (35.1% vs 23.4%, p = .122). Gynecologic and other specialty MEs did not state the need for procedure rescheduling (0% vs 0%).

CONCLUSION

Most reported robotic MEs occurred intraoperatively, were related to robotic instrument malfunctions, and required an instrument switch. Most surgeries are completed robotically, but conversion to either an open or laparoscopic approach was reported in 4.4%. Of the 114 reported injuries, 47.4% were Clavien-Dindo grade III+. There were no differences noted in patient injury between gynecologic surgery and other specialties.

Teaching and learning robotic surgery at the dual console: a video-based qualitative analysis

Robotic-assisted surgery (RAS) involves training processes and challenges that differ from open or laparoscopic surgery, particularly regarding the possibilities of observation and embodied guidance. The video recording and the dual-console system creates a potential opportunity for participation. Our research, conducted within the department of visceral surgery of a big Swiss, public, academic hospital, uses a methodology based on the co-analysis of video recordings with surgeons in self-confrontation interviews, to investigate the teaching activity of the lead surgeon supervising a surgeon in training at the dual console. Three short sequences have been selected for the paper. Our analysis highlights the skills-in-construction of the surgeon in training regarding communication with the operating team, fluency of working with three hands, and awareness of the whole operating site. It also shows the divergent necessities of enabling verbalization for professional training, while ensuring a quiet and efficient environment for medical performance. To balance these requirements, we argue that dedicated briefing and debriefing sessions may be particularly effective; we also suggest that the self-confrontation video technique may be valuable to support the verbalization on both the mentor’s and the trainee’s side during such debriefing, and to enhance the mentor’s reflexivity regarding didactic choices.

Robotic sphincter saving rectal cancer surgery: A learning curve analysis

BACKGROUND

Longer operation time is one of the major obstacles in front of the proposed benefits of robotic rectal surgery. We intended to evaluate the learning process for robotic surgery in sphincter saving rectal cancer surgery.

METHODS

The learning curve was evaluated using the cumulative sum (CUSUM) method. The variable evaluated for learning curve calculation was the operative time.

RESULTS

The learning curve was divided into two phases: initial 52 operations comprised phase 1 and the following 44 operations represented phase 2. Interphase comparisons showed that phase 2 patients had shorter operation times (323.3 ± 102.8 vs. 379.9 ± 108.7 min, p = 0.011), less blood loss (37.2 ± 51.0 vs. 87.7 ± 124.8 mL, p = 0.009), longer distal resection margins (4.5 ± 4.3 vs. 2.5 ± 1.7 cm, p = 0.008), and higher rates of grade 3 mesorectal completeness (p = 0.001).

CONCLUSION

In this study, we saw that the cut-off level in the learning curve of a laparoscopically experienced surgeon could be beyond the numbers reported in the literature.