Initial outcomes of a robotic thoracic fellowship programme—a feasible and safe way to train the next generation

Introduction

Thoracoscopic surgery has made significant progress since it was first introduced in the early 1990s and now facilitates a significant number of thoracic surgical procedures (1-3). Video-assisted thoracoscopic surgery (VATS) lobectomy for lung cancer has been associated with less pain, fewer in-hospital complications and shorter hospital stay, without compromising early oncologic outcomes. Patients who undergo VATS procedures for lung resection experience superior post-operative functional recovery, lower re-admission rates and no difference in disease-free and overall survival (4). Despite constant advancements, both in technology and in training, surgeons called to perform VATS procedures may still be challenged by technical limitations, such as impaired vision and lack of instrument manoeuvrability, which can make complex resections harder to perform (5,6).

The introduction of robotic-assisted thoracoscopic surgery (RATS) has improved intra-operative vision, increased precision of movement, and provided CO₂ insufflation technology and novel stapling instrumentation to thoracic surgeons (7-9). RATS lobectomy has been demonstrated to carry lower conversion rates when compared to VATS lobectomy. It has also been associated by some authors with a shorter inpatient length of stay, although it must be acknowledged that there are several potential confounders when assessing the length of stay associated with a novel technique (10-12). At present, the DaVinci platform (Intuitive Surgical, Sunnyvale, CA, USA) is the most utilised robotic platform within thoracic surgery (13). The range of DaVinci models deliver physiological hand tremor filtration, powerful image magnification, seven degrees of instrument freedom and tri-dimensional high-definition vision (8,14). The DaVinci Xi platform, when compared to the previous model, DaVinci Si, offers faster docking, better robotic arm positioning and the ability to ‘port hop’ with the camera, and is already yielding positive results within other surgical specialities (15,16).

The accelerated adoption of new technology and novel surgical techniques has resulted in an increased demand for robotic training programmes within thoracic surgery (17). However, initial training with RATS has been mainly focused on already established surgeons, creating a lag in trainee experience with the robot (18). Currently, there is no consensus with regards to establishing a structured robotic training program, and multiple centres have been autonomously working to develop a systematic approach of teaching robotic surgery with purpose of increasing trainee autonomy whilst still maintaining high standards of patient care (19,20).

The aim of our study was to analyse patient outcomes from the inception period of a dedicated robotic thoracic surgery fellowship programme; comparing outcomes between trainer and trainee to confirm that the programme was feasible and safe. We also aimed to describe the way in which our programme is organised and conducted—so that it may be reproduceable by other institutions establishing their own robotic training programmes going forward. We present this article in accordance with the STROBE reporting checklist (available at https://ccts.amegroups.com/article/view/10.21037/ccts-24-16/rc).

Methods

We conducted a retrospective review of a prospectively populated database in a single thoracic surgery centre in the United Kingdom (UK). Patients’ clinical data were collected from surgical databases, operative logbooks, and electronic patient records. Individual patients’ consent was waived for this study. All operations were performed by, or under the supervision of, a single expert thoracic surgeon trainer. The trainer was a surgeon with 5 years’ robotic thoracic surgery experience and 11 years’ consultant practice. He had extensive training experience, holding a position as Intuitive European proctor and a previous position as previous role as regional training committee chair. We included all lung cancer patients undergoing anatomic lung resections, both lobar and sub-lobar (complex and simple segmentectomy), with the DaVinci Xi surgical system between January 2018 to October 2021 in the analysis. Simple segmentectomy was defined as a resection that creates one, linear intersegmental plane, whereas complex segmentectomy was defined as a resection that creates several intersegmental planes.

We collected demographic and clinical data including age, sex, body mass index (BMI), modified medical research council (mMRC) dyspnoea score, forced expiratory volume in 1 second (FEV₁), and diffusion capacity of the lung for carbon monoxide (DLCO). Perioperative outcomes measured included operation time, fissure difficulty score, blood loss, number of lymph nodes sampled, stapler firings and overall complexity score.

Fissure difficulty score was based on the integrity of the interlobar fissure, and was devised as follows: complete with entirely separate lobes (1 point), complete visceral cleft but parenchymal fusion at the base of the fissure (2 points), visceral cleft evident for part of the fissure (3 points), complete fusion of the lobe with no evident fissure line (4 points) (21). Overall complexity was calculated considering the type of anatomic resection, presence of extensive adhesions, redo or completion procedures, and requirement for extensive or complex resections, including bi-lobectomy, bronchoplasty or bronchial sleeve resection (Table S1).

The fissure difficulty score was agreed on by both trainer and trainees on an individual case by case basis, creating a shared consensus around operation difficulty and therefore allowing for direct comparison of results between trainers and trainees. The score was devised solely for this purpose—it was not our intention to create a validated or verified score which would be reproduceable in other settings—development of such a score is outside of the scope of this study.

Post-operative data collected included number of complications, length of stay and pain severity. We recorded conversion to thoracotomy when incision extended beyond 4 cm or when a rib spreader was utilised. Prolonged air leak was defined as lasting more than 5 days post-operatively.

Operations were conducted under general anaesthesia with a double lumen endotracheal tube. The standard set up involved positioning of the patient cart at the back of the patient. As per the trainer’s routine practice, patients were placed in lateral decubitus and steep reverse Trendelenburg position, to allow wider range of vertical motion for the robotic arms by avoiding conflicts with the pelvis (22). The camera port is first introduced at the level of the 8^th intercostal space in the posterior axillary line; three further ports are then inserted under thoracoscopic vision, distanced equally following a horizontal line (roughly 3 finger’s breadth apart), to facilitate maximum ergonomic movement. We do not routinely use a utility port, a choice which we feel maximises the control that the console surgeon has over the operative field. This also goes toward maintaining our totally endoscopic approach, and helps to preserve our capnothorax, routinely established at 6 mmHg. The development of a new generation of surgeon controlled endowristed robotic staplers with the DaVinci Xi platform has limited the use of non-robotic staplers to only rare instances in which tissue density prevented the robotic staplers from firing automatically or the robotic staplers had reached their 12-fire limit (in which case a VATS stapler was used for reasons of economy).

We routinely performed regional anaesthesia to aid in postoperative pain control, with the use of 40 mL levobupivacaine 0.25% administered in divided doses as thoracoscopic intercostal blocks superior and inferior to the level of the ports at the beginning of the procedure, and through a paravertebral catheter inserted at the end of the procedure. Routine post-operative pain control involved a paravertebral infusion of levobupivacaine, and a multi-modal analgesia protocol including paracetamol, ibuprofen if not contraindicated, regular dihydrocodeine and breakthrough morphine or oxycodone. Our multi-modal approach to postoperative analgesia is employed with the goal of limiting opiate use and its consequent complications in so far as possible, in keeping with enhanced recovery after thoracic surgery (ERAS) guidelines for thoracic surgery (23). Routine use of patient-controlled analgesia (PCA) pumps was avoided due to the risk of opiate toxicity in our relatively frail, elderly patient cohort. High risk patients were routinely recovered in a high dependency unit for the first postoperative night. All other patients were cared for initially in a post-anaesthetic recovery room and subsequently at ward level, using a standardised ERAS protocol for thoracic surgery (23).

Robotic thoracic fellowship programme

A robotic thoracic fellowship programme was commenced in January 2018. The period covered by this study is the inception period for the programme, running until October 2021 and coinciding with the coronavirus disease 2019 (COVID-19) pandemic. The challenges posed by COVID-19 operative training for surgical trainees have been well documented elsewhere (24). Fellowship training was carried out on the DaVinci Xi platform. At the commencement of the programme all fellows were thoracic surgeons who were nearing the end of their training—approaching or having just passed the point of specialist certification. All had performed at least 50 VATS lung resections prior to the commencement of the fellowship programme as first operator.

Training initially commenced with a series of online modules via Intuitive’s ‘DaVinci Learning’ platform. Once these modules were passed the fellow then transitioned to console-based simulation training using the DaVinci simulation platform for the surgeon console. Once basic knowledge of and familiarity with the system has been obtained by the fellow and assessed by the trainer, the fellow transitioned to performing simple RATS cases, such as wedge resections and excision of uncomplicated paravertebral or pleural based lesions, under the direct supervision of a consultant trainer using the dual console setup. Progression was facilitated by pre-operative case briefings (utilising videos of previous similar operations performed by the consultant trainer), peri-operative one-to-one tutoring, post-operative video analysis and gradually increasing case complexity. In this way fellows gradually progressed to completing more complex resections with increasing independence. The dual-console setup with the DaVinci Xi platform greatly enhanced the efficiency and safety of training—allowing the trainer to take and give control of the operation at various points as needed, and monitor the fellow’s progress with the enhanced three-dimensional (3D) vision that the platform provides. There were no specific criteria for case selection for fellow led operating; moreover, all cases were deemed suitable for fellow led operating at least in part. In more complex cases control would be handed over to the consultant trainer as needed during the case depending on the fellow’s level of proficiency and experience.

Statistical analysis

Computations were performed using Microsoft Excel (Microsoft, WA, USA) and statistical analysis was performed using the 26th version of SPSS Statistics (IBM, NY, USA). Measures were expressed as means and standard deviations (SD). Levene’s test was performed for equality of variances and two-tailed t-test was used to analyse all outcomes, with the exception of length of stay, which was analysed with a Mann-Whitney test as it was not normally distributed in our cohort. Rejection of the null hypothesis at the level of 5% was adopted.

Ethical consideration

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study did not require local ethics board approval, as permission is explicitly sought for data collection on normal patient care as part of the normal consent process for operation (these data are used to contribute to a national database to monitor outcomes). Patients were advised that trainees may be involved in performing procedures as is standard practice at a teaching hospital. There were no changes to what was considered normal care/normal practice, and patients were not randomised. As such individual consent for this retrospective analysis was also waived.

Results

A dedicated thoracic robotic fellowship program was implemented in our unit, during a challenging period overlapping the COVID-19 pandemic. A total of 187 robot-assisted anatomic lung resections were performed during the inception period of the program (72 males, 115 females, mean age 69 years); 123 lobectomies and 66 segmentectomies. Forty-five patients had an MRC dyspnoea score of 1 (24%), 116 had a score of 2 (62%), 25 had a score of 3 (13%) and one had a score of 4 (0.5%).

One hundred and fourteen operations were performed (60.96%) by the trainer and 73 (39.04%) by fellows as primary operators. Patients operated on by fellows were slightly older (71.47 vs. 67.77 years, P=0.06); the reasons for this are unclear—although statistically significant it is unlikely to be clinically significant. There was no statistically significant difference in any of the other preoperative patient factors measured (BMI, FEV₁, DLCO) (Table 1).

As would be expected, the operation time was longer for the fellow-operated cohort than for the trainer-operated cohort, although the difference failed to reach statistical significance (218 vs. 202 minutes, P=0.16). Again, as expected the fissure difficulty score was slightly lower in the fellow-operated cohort (2.10 vs. 2.35, P=0.07), as was overall complexity score (1.47 vs. 1.54, P=0.39), however neither of these differences succeeded in reaching statistical significance. Operations performed by fellows had significantly less recorded blood loss (87.14 vs. 165.66 mL, P=0.008). Importantly the number of lymph nodes resected (3.82 vs. 3.71, P=0.55) was similar in both groups. In the overall cohort, 87% of patients had 2 or more mediastinal lymph nodes sampled, and station 7 was sampled in 96% of patients—the mediastinal station which is perhaps most difficult to access via a VATS approach to the posterior hilum.

There was no statistically significant difference observed between patients operated on by fellows and trainers with regard to any of the post-operative variables (complications score 1.36 vs. 1.20, P=0.38, length of stay 7.12 vs. 6.12, P=0.48, pain severity 0.62 vs. 0.63, P=0.91), although it is worth noting that length of stay was longer for patients operated on by fellows by 1 day, even though this failed to reach statistical significance.

The overall complication rate for our cohort was 36% (68/187). Of these 40% were grade 1, 37% were grade 2, 13% were grade 3 and 10% were grade 4, according to the Clavien-Dindo classification (25); 2.7% (5/187) of cases underwent controlled conversion to an open procedure, all of these were cases operated on by the trainer. Three of these conversions were due to bleeding, one due to the inability to achieve single lung isolation and one due to adhesions. No fellow-led cases underwent conversion to open. The 30-day mortality was 1.6% (3/187) and the 90-day mortality was 2.1% (4/187). Of the four patients who died within 90 days of surgery causes of death included pericardial tamponade secondary to effusion of unknown origin on postoperative day 1, respiratory failure secondary to pneumonia (2 patients), and combined respiratory failure secondary to pneumonia and left ventricular failure. In terms of completeness of resection the R0 resection rate was 96.5% for the trainer and 94.5% for the fellows (n=4 in each group).

Discussion

As operative techniques continue to evolve, technology has assumed a more prominent role in the future of surgery; we are now called to embrace an environment of quick progression, driven by rapid technological advancements. The introduction of computer aided robot assisted surgery in the new millennium has overcome technical challenges associated with VATS, providing new tools to facilitate minimally invasive surgery, and has been widely adopted by thoracic surgeons (26,27).

The construction of a robotic theatre in our department to accommodate the introduction of the new DaVinci Xi robotic platform (Intuitive Surgical), dedicated to thoracic surgical procedures, resulted in a significant shift of routine practice from video-assisted lobectomies to robotic-assisted sub-lobar resections. This shift in our department occurred as the result of a London based low dose computed tomography (CT) lung cancer screening study (the SUMMIT trial—ClinicalTrials.gov ID NCT03934866), in which we were a participating unit. As such there was an increase in the burden of small nodules coming forward for resection in patients with compromised lung function. The overall resection rate at our unit, which ranged from 31% to 43% over the trial period is consistently amongst the highest in the UK. Our approach has since been supported by the release of the JCOG 0802 and CALGB 14503 randomised controlled trials—both of which support sub-lobar resection as non-inferior to lobectomy with respect to 5-year disease free survival (28,29). We believe that the robotic platform, with its improved ergonomics, instrument dexterity and 3D vision as compared to VATS has broadened surgeons’ ability to perform more complex anatomical resections, and as such early on we recognised the need to establish a fellowship program aimed at senior trainees wishing to develop sub-specialty interest and skills in the evolving field of robotic sub-lobar resection.

Results from implementation of a robotic fellowship training programme

Our study demonstrates that the training of surgical fellows in RATS with an appropriately formulated teaching programme is safe, feasible and has no negative impact on patient outcomes. Our results yielded no significant difference in overall outcomes between trainer and fellow. We did demonstrate that operations performed by trainees had significantly less recorded blood loss, while also having a slightly (non-significantly) lower resection and overall complexity score. Although statistically significant it is unlikely that the recorded difference in blood loss (79 mL) is clinically significant. It is worth noting however that reduced blood loss results in a cleaner operating field—something which is needed for training. Previous studies have demonstrated that operative time, estimated blood loss, and length of stay were similar regardless of level of fellow participation within operations (18,20). Cerfolio et al. previously reported their teaching centre outcomes, demonstrating that by splitting robotically assisted lobectomy into 19 discrete steps, good patient outcomes can be preserved (20). Our study’s positive patient outcomes with a differing approach suggest there is a multitude of teaching modalities within robotic surgery that we must continue to explore and rationalise. We would like to highlight the importance of video recording of the procedures with preliminary analysis performed early after the end of the operation. There is a wealth of knowledge to be gained by examining key steps of the operations, as well as more technically challenging aspects, so that improvements can be considered and adopted as early as in the same operative day. We also favour return to the console at any time to continue developing skills of fine movements and tissue handling which may not be common in routine practice, to prepare for more technically challenging operations and unplanned incidents. We have demonstrated similar levels of R0 resection between fellows and trainers (94.5% and 96.5% respectively), suggesting that fellow-led operating remains oncologically safe.

We acknowledge that length of stay is dependent on multiple factors outside the surgeon’s control—for example resident led ward rounds, patients’ social circumstances, how far away patients live from the hospital where they receive their operations (prohibiting discharge with drains in situ), other medical co-morbidities etc. Length of stay is therefore an unreliable metric of surgical and more specifically RATS performance. However, we included it because existing evidence relating to reduced length of stay for RATS vs. VATS/open surgical cases has been used in business cases for the development of RATS lung resection programmes all over the world. Surgical departments must remain cost-conscious, especially when developing fellowship programmes. And so, the absolute length of stay is not our primary concern—moreover we wish to communicate the fact that fellowship programmes like ours do not result in significantly increased patient-related costs, and therefore cost concerns should not be a reason for more centres not to adopt these programmes.

Specific outcomes from the DaVinci Xi surgical system

Our study showed that 2.7% (5/187) of cases overall underwent controlled conversion to an open procedure. Previous literature examining robotic pulmonary resections, has demonstrated conversion rates to thoracotomy as high as 19% (30). To manage operative complications, we favour direct conversion to a thoracotomy incision instead of transitioning from a RATS to VATS approach; with our accumulated knowledge and experience over the years, we feel that the improved dexterity offered by the robotic instruments, for surgeons who have completed their learning curve, can only be surpassed by hand movements alone. Given that conversion is a surgical emergency requiring extensive co-ordination of the surgical, scrub, anaesthesia and circulating teams we practice this regularly as a team in a simulated setting at our unit. The fact that all of the cases which converted to open were cases operated on by the trainer perhaps reflects the more complex case-mix that is reserved for the more experienced operator (as reflected in the slightly higher resection difficulty and overall complexity scores); nonetheless it is encouraging that none of the fellow-led cases required conversion to open.

Our overall complication rate of 36% (68/187) on the DaVinci Xi console was 5% less than a previous study conducted by Kent et al. (31). Our results yielded a 30- and 90-day operative mortality of 1.6% and 2.2% respectively. Nasir et al. published outcomes post lobectomy and segmentectomy demonstrating 30- and 90-day operative mortality of 0.25% and 0.5% (32). Overall, our outcomes with the DaVinci Xi system are consistent with previously published literature, however variation in results reinforces the fact that that the surgical system alone does not determine patient outcomes. Factors that confound and influence patient outcomes include but are not limited to patient co-morbidities and selection for surgery, institutional peri-operative practice, and surgeon experience. There clearly remains variability in published outcomes associated with RATS, however our results contribute to the growing body of literature that confirms the feasibility of robotic pulmonary resections.

Study limitations

Our study is limited by selection bias inherent to retrospective studies. The study’s design could not control for the differing levels of experience amongst fellows, the presumed increase in experience of our single consultant trainer over time, and the natural differences in institutional practices. The scoring systems we have used in this study are not validated and our cohort is relatively small. Analysis of overall complication rates was performed, but more detailed analysis of rates for specific complications peri- and post-operatively was beyond the remit of this study. Operations were performed in a large tertiary centre under the care of a single expert surgeon, hence results may not be generalisable to other institutions.

Conclusions

Implementation of a dedicated robotic fellowship programme was feasible and safe, and resulted in no significant differences in patient outcomes between patients operated on by trainers and patients operated on by fellows.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://ccts.amegroups.com/article/view/10.21037/ccts-24-16/rc

Data Sharing Statement: Available at https://ccts.amegroups.com/article/view/10.21037/ccts-24-16/dss

Peer Review File: Available at https://ccts.amegroups.com/article/view/10.21037/ccts-24-16/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (https://ccts.amegroups.com/article/view/10.21037/ccts-24-16/coif). S.S. is a European Proctor for Intuitive Surgical. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study did not require local ethics board approval, as permission is explicitly sought for data collection on normal patient care as part of the normal consent process for operation (these data are used to contribute to a national database to monitor outcomes). Patients were advised that trainees may be involved in performing procedures as is standard practice at a teaching hospital. There were no changes to what was considered normal care/normal practice, and patients were not randomised. As such individual consent for this retrospective analysis was also waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Kumar A, Asaf BB. Robotic thoracic surgery: The state of the art. J Minim Access Surg 2015;11:60-7. [PubMed]
2. Landreneau RJ, Mack MJ, Hazelrigg SR, et al. Video-assisted thoracic surgery: basic technical concepts and intercostal approach strategies. Ann Thorac Surg 1992;54:800-7. [PubMed]
3. Roviaro G, Rebuffat C, Varoli F, et al. Videoendoscopic pulmonary lobectomy for cancer. Surg Laparosc Endosc 1992;2:244-7. [PubMed]
4. Lim E, Harris RA, McKeon HE, et al. Impact of video-assisted thoracoscopic lobectomy versus open lobectomy for lung cancer on recovery assessed using self-reported physical function: VIOLET RCT. Health Technol Assess 2022;26:1-162. [PubMed]
5. Gonzalez-Rivas D. VATS lobectomy: surgical evolution from conventional VATS to uniportal approach. ScientificWorldJournal 2012;2012:780842. [PubMed]
6. Dieter RA Jr, Kuzyçz GB. Complications and contraindications of thoracoscopy. Int Surg 1997;82:232-9. [PubMed]
7. Veronesi G. Robotic thoracic surgery: technical considerations and learning curve for pulmonary resection. Thorac Surg Clin 2014;24:135-41. v. [PubMed]
8. Zirafa CC, Romano G, Key TH, et al. The evolution of robotic thoracic surgery. Ann Cardiothorac Surg 2019;8:210-7. [PubMed]
9. Bodner J, Wykypiel H, Wetscher G, et al. First experiences with the da Vinci operating robot in thoracic surgery. Eur J Cardiothorac Surg 2004;25:844-51. [PubMed]
10. Baig MZ, Razi SS, Agyabeng-Dadzie K, et al. Robotic-assisted thoracoscopic surgery demonstrates a lower rate of conversion to thoracotomy than video-assisted thoracoscopic surgery for complex lobectomies. Eur J Cardiothorac Surg 2022;62:ezac281. [PubMed]
11. Li C, Hu Y, Huang J, et al. Comparison of robotic-assisted lobectomy with video-assisted thoracic surgery for stage IIB-IIIA non-small cell lung cancer. Transl Lung Cancer Res 2019;8:820-8. [PubMed]
12. Ma J, Li X, Zhao S, et al. Robot-assisted thoracic surgery versus video-assisted thoracic surgery for lung lobectomy or segmentectomy in patients with non-small cell lung cancer: a meta-analysis. BMC Cancer 2021;21:498. [Crossref] [PubMed]
13. Terra RM, Leite PHC, Dela Vega AJM. Global status of the robotic thoracic surgery. J Thorac Dis 2021;13:6123-8. [Crossref] [PubMed]
14. Ashrafian H, Clancy O, Grover V, et al. The evolution of robotic surgery: surgical and anaesthetic aspects. Br J Anaesth 2017;119:i72-84. [Crossref] [PubMed]
15. Protyniak B, Jorden J, Farmer R. Multiquadrant robotic colorectal surgery: the da Vinci Xi vs Si comparison. J Robot Surg 2018;12:67-74. [Crossref] [PubMed]
16. Ozben V, Cengiz TB, Atasoy D, et al. Is da Vinci Xi Better than da Vinci Si in Robotic Rectal Cancer Surgery? Comparison of the 2 Generations of da Vinci Systems. Surg Laparosc Endosc Percutan Tech 2016;26:417-23. [Crossref] [PubMed]
17. Shahin GMM, Brandon Bravo Bruinsma GJ, Stamenkovic S, et al. Training in robotic thoracic surgery-the European way. Ann Cardiothorac Surg 2019;8:202-9. [Crossref] [PubMed]
18. White YN, Dedhia P, Bergeron EJ, et al. Resident training in a new robotic thoracic surgery program. J Surg Res 2016;201:219-25. [Crossref] [PubMed]
19. Linsky PL, Wei B. Training in robotic thoracic surgery. J Vis Surg 2018;4:1. [Crossref] [PubMed]
20. Cerfolio RJ, Cichos KH, Wei B, et al. Robotic lobectomy can be taught while maintaining quality patient outcomes. J Thorac Cardiovasc Surg 2016;152:991-7. [PubMed]
21. Craig SR, Walker WS. A proposed anatomical classification of the pulmonary fissures. J R Coll Surg Edinb 1997;42:233-4. [PubMed]
22. Stamenkovic S, Melfi F. Right robotic-assisted thoracoscopic lower lobectomy. Ann Cardiothorac Surg 2019;8:279-85. [Crossref] [PubMed]
23. Batchelor TJP, Rasburn NJ, Abdelnour-Berchtold E, et al. Guidelines for enhanced recovery after lung surgery: recommendations of the Enhanced Recovery After Surgery (ERAS®) Society and the European Society of Thoracic Surgeons (ESTS). Eur J Cardiothorac Surg 2019;55:91-115. [Crossref] [PubMed]
24. Lund J, Sadler P, McLarty E. The effect of COVID-19 on surgical training. Surgery (Oxf) 2021;39:829-33. [Crossref] [PubMed]
25. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009;250:187-96. [Crossref] [PubMed]
26. Melfi FM, Menconi GF, Mariani AM, et al. Early experience with robotic technology for thoracoscopic surgery. Eur J Cardiothorac Surg 2002;21:864-8. [Crossref] [PubMed]
27. Soliman BG, Nguyen DT, Chan EY, et al. Impact of da Vinci Xi robot in pulmonary resection. J Thorac Dis 2020;12:3561-72. [Crossref] [PubMed]
28. Saji H, Okada M, Tsuboi M, et al. Segmentectomy versus lobectomy in small-sized peripheral non-small-cell lung cancer (JCOG0802/WJOG4607L): a multicentre, open-label, phase 3, randomised, controlled, non-inferiority trial. Lancet 2022;399:1607-17. [Crossref] [PubMed]
29. Altorki N, Wang X, Kozono D, et al. Lobar or Sublobar Resection for Peripheral Stage IA Non-Small-Cell Lung Cancer. N Engl J Med 2023;388:489-98. [Crossref] [PubMed]
30. Augustin F, Bodner J, Wykypiel H, et al. Initial experience with robotic lung lobectomy: report of two different approaches. Surg Endosc 2011;25:108-13. [Crossref] [PubMed]
31. Kent M, Wang T, Whyte R, et al. Open, video-assisted thoracic surgery, and robotic lobectomy: review of a national database. Ann Thorac Surg 2014;97:236-42; discussion 242-4. [PubMed]
32. Nasir BS, Bryant AS, Minnich DJ, et al. Performing robotic lobectomy and segmentectomy: cost, profitability, and outcomes. Ann Thorac Surg 2014;98:203-9. [PubMed]