Abstract
Background
Minimally invasive thoracic surgery, including robot-assisted thoracoscopic surgery (RATS), has improved outcomes by reducing postoperative pain. However, intercostal nerve trauma from transthoracic approaches remains a concern. The “outside the cage” (OTC) subcostal robotic approach may reduce pain while maintaining safety. This study aimed to evaluate the learning curve of this technique in major lung and anterior mediastinal resections.
Methods
We conducted a prospective single-center study from January 2024 to April 2025. All procedures were performed by a single experienced robotic surgeon using the Da Vinci Xi system. Cumulative sum (CUSUM) analysis evaluated learning curves for docking and console times in two groups: major lung resections and thymectomy/anterior mediastinal mass resections.
Results
Of 131 patients, 107 were operated on by one surgeon: 67 underwent major lung resection and 40 thymectomy or anterior mediastinal mass resection. CUSUM analysis of docking time identified three phases—training, improvement (cases 17–33 for thymectomy/anterior mediastinal mass resection, cases 31–38 for major lung resection), and mastery (case 34 onwards for thymectomy/anterior mediastinal mass resection, case 39 onwards for major lung resection)—in both groups. In contrast, no distinct learning curve was observed for console time, suggesting proficiency from the outset. No conversions occurred except one due to vascular invasion. At 30 days, no patients reported neuropathic pain or needed analgesics.
Conclusions
The OTC subcostal approach appears safe and rapidly adoptable for experienced robotic surgeons, with a brief learning curve limited to docking maneuvers. These findings support its reproducibility and potential wider adoption of this pain-reducing technique.
Keywords: Robot-assisted thoracoscopic surgery (RATS), outside the cage (OTC), subcostal approach, subxiphoid approach
Highlight box.
Key findings
• This study aims to describe the learning curve of the subcostal robotic approach.
What is known and what is new?
• Subcostal approach in thoracic surgery allows reduction of post-operative pain and painkiller consumption as well as a reduction in chronic neuropathic pain. However, the learning process of this technique has not been described yet and concerns can be raised regarding its adoption by the thoracic surgery community.
• This study founds a short learning process for the docking time and no learning process for the console time.
What is the implication, and what should change now?
• For an experienced thoracic surgeon, the subcostal technique appears to be safe, doable and can be learned quickly, suggesting its wider adoption among the thoracic surgery community.
Introduction
Over the past few decades, thoracic surgical techniques have undergone a major transformation, shifting from traditional open thoracotomies to minimally invasive procedures such as video-assisted thoracoscopic surgery (VATS) and robot-assisted thoracoscopic surgery (RATS) with the primary objective of reducing postoperative pain while maintaining good surgical outcomes (1,2).
While pain management has improved, conventional minimally invasive approaches, either VATS or RATS, can still be associated with postoperative pain, particularly neuropathic pain caused by intercostal nerve injury from trocar insertion. This can lead to increased opioid consumption, prolonged hospital stays, and delayed recovery (3). To minimize intercostal nerve trauma, several surgical teams have developed a subcostal approach. This was first described in 1999 by Kido et al. with a subxiphoid video assisted approach for resection of benign anterior mediastinal masses (4). Later on, a combined subxiphoid-subcostal VATS approach was described in 2022 by Rao et al., allowing resection of malignant anterior mediastinal masses with no differences in operative time or major complications comparing with a conventional transthoracic thoracoscopic approach (5). Regarding major lung resection with an extra thoracic video assisted approach, it was firstly described in 2014 by Liu et al. (6) who performed a sub-xiphoid single-incision left upper lobectomy.
The sub costal approach in the field of robotic surgery was firstly described in 2022 by Korean teams: Park et al. (7) described a uniportal technique using the DaVinci SP robot (Intuitive SurgicalÓ, Sunnyvale, CA, USA), while Vishnu et al. described a multiportal technique utilizing the DaVinci Xi robot (Intuitive SurgicalÓ) (8). Interest in the subcostal approach has continued to grow over the last few years, with teams sharing their experiences, whether with the SP (9) or XI system (10).
Since the DaVinci SP robot is not yet available for general thoracic surgery in France, our team adopted the “outside the cage” (OTC) subcostal approach, as described by Bulgarelli Maqueda et al. (10) for major lung resections and transposed the video-assisted approach described by Rao et al. (5) into a robotic-assisted approach. We published our initial findings in 2024 (11), demonstrating a reduction in postoperative pain and opioid consumption compared with the conventional transthoracic RATS approach. However, beyond these pain-related outcomes, comparative data on other peri- and postoperative endpoints such as complications, conversion, readmission or length of stay remain scarce, and the OTC approach has not yet been shown to be superior to the standard intercostal transthoracic RATS technique for these outcomes.
Because concerns have been raised regarding the learning curve of this technique, particularly the reproducibility of the subcostal docking, we sought to analyze the learning curve of this novel subcostal thoracic surgery technique after completing more than 100 procedures using the CUSUM method. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2062/rc).
Methods
Materials
We conducted a single center prospective study at the University Regional Hospital of Nancy, France. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was reviewed and approved by the institutional review board of University Regional Hospital of Nancy (IRB No. 2024PI016-489). Informed consent was taken from all the patients.
In January 2024, Department of Thoracic Surgery in the University Regional Hospital of Nancy switched from a transthoracic to an OTC robotic approach for all cases (excluding diaphragmatic surgery). Accordingly, all patients who underwent OTC RATS surgeries between January 2024 and April 2025 were included. All surgeries were performed by a single senior consultant who had already performed more than 400 conventional transthoracic robotic cases.
The Xi robotic system (Intuitive SurgicalÓ) was always used. The surgical approaches have been described previously (11). In summary:
❖ In case of major pulmonary resection: patients were placed in lateral decubitus. The operating table was flexed by about 20° at hip level. An anterior subcostal utility incision (3–4 cm) was made approximately 2 cm lateral to the xiphoid process, following the costal margin. The abdominal wall was opened by partially dividing the muscles with electrocautery down to the transversus abdominis fascia at the costal edge. A blunt finger dissection was then carried out to develop the access tract for the camera. A wound retractor with a GelPortTM system (Applied Medical, Amersfoort, The Netherlands) was inserted, enabling CO2 insufflation and placement of the first trocar. Under thoracoscopic visualization, a second 8-mm port was introduced 10 mm lateral to the first, beneath the costal margin and through diaphragmatic fibers. A third port was placed in a similar fashion, 10 mm lateral to the second, also transdiaphragmatic. The fourth port was positioned at the level of the 12th rib head through the diaphragm. For right-sided procedures, the surgeon’s right-hand instrument was assigned to the first port (12 mm); for left-sided procedures, it was assigned to the third port (12 mm). All remaining robotic ports were 8 mm. The camera was consistently placed through the second port. The right robotic arm was primarily used for stapling and dissection with a Maryland bipolar forceps, while the other arms carried ProGrasp forceps for retraction and lung handling. In addition, a 15-mm assistant port was created within the utility incision via the GelPortTM, allowing suction and lymph node extraction. The OTC subcostal docking setup for major lung resections, including patient positioning, subcostal utility incision, port placement and cart positioning, is illustrated in Video S1.
❖ For anterior mediastinum approach (used for anterior mediastinal mass resection or thymectomy), patients were positioned in a dorsal decubitus position. The table was bent by approximately 20° at the level of the costal margin. A 4 cm incision was made just below the xiphoid process and the abdominal muscles were partially divided with cautery. Finger-blunt dissection was then used to create a space beneath and to open the pleura bilaterally. A wound protector with a GelPortÓ system (Applied Medical) system was placed, allowing CO2 insufflation. A first 8 mm port was inserted through the GelPortÓ system and two additional right and left 8 mm ports were inserted under direct vision just above the diaphragm and under the costal margins. The corresponding docking configuration for thymectomy and anterior mediastinal mass resections is illustrated in Video S2.
Video S1.
OTC subcostal docking setup for major lung resections, showing patient positioning, subcostal utility incision, port placement and positioning of the robotic cart and arms. OTC, outside the cage.
Video S2.
OTC subcostal docking setup for thymectomy and anterior mediastinal mass resections, showing patient positioning, subcostal utility incision, port placement and positioning of the robotic cart and arms. OTC, outside the cage.
Age, sex, body mass index (BMI), American Society of Anesthesiologist (ASA) score, smoking history, comorbidities were collected for each patient. Intra operative variables included operative time (defined as console time), docking time, blood loss and conversion rates. Postoperative outcomes were prospectively recorded and included pain scores and analgesic consumption, duration of chest tube drainage, length of stay, postoperative complications and 30-day readmission. In the present manuscript, the learning-curve analysis was deliberately focused on the technical metrics (docking and console times), whereas the detailed analysis of pain outcomes has been presented in a separate matched-cohort study (11). Patients were assessed at the postoperative clinic around 30 days after surgery with a clinical examination and a chest X-ray (with a thorough examination of the diaphragm).
Statistical analysis
In the specific context of this study, learning was defined a priori as the process of technical familiarization and optimization of efficiency when adopting a new subcostal OTC docking configuration by a surgeon who had already completed the learning curve for conventional intercostal RATS. Accordingly, we focused on time-based technical metrics (docking and console times), rather than on a global institutional learning curve that would also integrate perioperative outcomes such as complications, conversion rates or length of stay. To evaluate this technical learning process, we used a cumulative sum (CUSUM) analysis of docking and console times as described below.
We performed an analysis of the learning curves for both operating time and docking time across the two surgical techniques. Learning curve analysis was based on the CUSUM method which is recognized as the reference method for studying learning curves (12).
The primary variable used for the learning curve analysis was operating time, measured in minutes. Cases were chronologically ordered and the CUSUM scores were plotted to identify trend changes corresponding to the phases of the learning curve. Distinct phases—training, improvement and mastery—were defined by changes in the slope of the CUSUM plot.
The CUSUM analysis was conducted using the R software environment (version 3.6.3; R Core Team, 2023), which provides robust statistical tools for such analyses. A modified CUSUM curve, incorporating a penalty for any deterioration in performance, was also calculated to better capture phases of regression during the learning process.
Results
From January 2024 to April 2025, 131 patients had a robotic subcostal surgery in Department of Thoracic Surgery in the University Regional Hospital of Nancy. Twenty-two surgeries were performed by surgeons who performed less than 30 robotic surgery and were excluded from the analysis. Two patients had a middle mediastinum mass resection and were excluded from the CUSUM analysis as the operating time was not comparable to the other procedures. Overall, 107 patients were included: 67 had a major lung resection and 40 had a thymectomy/anterior mediastinum resection (Figure 1). The characteristics of the population are described in Table 1 and the per operative data are described in Table 2. There was one conversion to sternotomy in the thymectomy/anterior mediastinal mass resection group due to the extension of the tumor requiring a resection of the left brachiocephalic vein.
Figure 1.
Flow chart of the study. OTC, outside the cage; RATS, robot-assisted thoracoscopic surgery.
Table 1. Characteristics of the population.
| Characteristics | Major lung resection (n=67) | Thymectomy/anterior mediastinal mass resection (n=40) |
|---|---|---|
| Age (years) | 59.6±16.8 | 58.4±17.3 |
| Sex | ||
| Male | 38 (56.7) | 14 (35.0) |
| Female | 29 (43.3) | 26 (65.0) |
| BMI (kg/m2) | 26.9±5.5 | 27.4±5.6 |
| Weight (kg) | 77.1±16.3 | 78.9±16 |
| ASA score | ||
| 1 | 1 (1.5) | 2 (5.0) |
| 2 | 44 (65.7) | 25 (62.5) |
| 3 | 22 (32.8) | 13 (32.5) |
| Smoking status | ||
| Non smoker | 18 (26.9) | 27 (67.5) |
| Former smoker | 26 (38.8) | 4 (10.0) |
| Active smoker | 22 (32.8) | 9 (22.5) |
| Tobacco consumption (pack years) | 19.7±22.6 | 17.6±22.4 |
| Past medical history | ||
| COPD | 19 (28.3) | 0 (0.0) |
| Hypertension | 28 (41.8) | 5 (12.5) |
| Myasthenia | 0 (0.0) | 12 (30.0) |
| Coronaropathy | 9 (13.4) | 2 (5.0) |
| Peripheral vascular disease | 13 (19.4) | 0 (0.0) |
| Diabetes | 5 (7.5) | 3 (7.5) |
| Chronic kidney disease | 5 (7.5) | 1 (2.5) |
| Cardiac arrhythmia | 5 (7.5) | 2 (5.0) |
| Heart failure | 4 (6.0) | 0 (0.0) |
| Lung cancer | 6 (9.0) | 0 (0.0) |
| Previous ipsilateral thoracic surgery | 2 (3.0) | 0 (0.0) |
| Previous contralateral thoracic surgery | 9 (13.4) | 0 (0.0) |
| Neoadjuvant treatment | 3 (4.5) | 2 (5.0) |
Data are presented as mean ± standard deviation or n (%). ASA, American Society of Anesthesiologist; BMI, body mass index; COPD, chronic obstructive pulmonary disease.
Table 2. Per operative data.
| Data | Major lung resection (n=67) | Thymectomy/anterior mediastinal mass resection (n=40) |
|---|---|---|
| Console time (min) | 101.94±42.12 | 99±41.02 |
| Docking time (min) | 17.39±6.58 | 17.06±6.64 |
| Blood loss (mL) | 83.5±100 | 43.75±52.22 |
| Conversion | 0 (0.0) | 1 (2.5) |
Data are presented as mean ± standard deviation or n (%).
Docking time
The mean docking time was 17.06 [standard deviation (SD) 6.64] min for thymectomies/anterior mediastinal mass resection and 17.39 (SD 6.58) min for major lung resections.
Figures 2,3 represent the CUSUM score with penalty of the learning curve of the docking time for major lung resection and thymectomy/anterior mediastinal mass resection respectively.
Figure 2.
CUSUM score representing the learning curve of the docking time for major lung resections. The dotted red line corresponds to the CUSUM decision limit. CUSUM, cumulative sum; LC, learning curve.
Figure 3.
CUSUM score representing the learning curve of the docking time for thymectomy and anterior mediastinal mass resection. The dotted red line corresponds to the CUSUM decision limit. CUSUM, cumulative sum; LC, learning curve.
For both docking techniques, the CUSUM analysis found the usual tree phases of the learning process: the initial training phase (cases 1–16 for thymectomy and anterior mediastinal mass resection, cases 1–31 for major lung resection), the improvement phase (cases 17–33 for thymectomy and anterior mediastinal mass resection, cases 31–38 for major lung resection), and the mastery phase (case 34 onwards for thymectomy and anterior mediastinal mass resection, case 39 onwards for major lung resection).
Console time
The mean console time was 99 (SD 41.02) min for thymectomies/anterior mediastinal mass resection and 101.94 (SD 42.12) min for major lung resections.
Figures 4,5 represent the CUSUM score of the learning curve of the console time for major lung resection and thymectomy/anterior mediastinal mass resection respectively.
Figure 4.
CUSUM score representing the learning curve of the console time for major lung resection. The dotted red line corresponds to the CUSUM decision limit. CUSUM, cumulative sum; LC, learning curve.
Figure 5.
CUSUM score representing the learning curve of the console time for thymectomy and anterior mediastinal mass resection. The dotted red line corresponds to the CUSUM decision limit. CUSUM, cumulative sum; LC, learning curve.
The CUSUM analysis did not find any change in the operative time for both techniques, suggesting that there was no learning process for an experienced robotic surgeon.
Postoperative outcomes
Postoperative outcomes are presented in Table 3. Two patients died with 30 days post-operatively, one from a pneumonia and one from acute decompensation of chronic liver disease.
Table 3. Postoperative outcomes.
| Outcomes | Major lung resection (n=67) | Thymectomy/anterior mediastinal mass resection (n=40) |
|---|---|---|
| Length of hospital stay (POD) | 5.08±5.13 | 2.77±2.48 |
| Duration of chest tube (POD) | 2.80±2.58 | 1.35±1.64 |
| Postoperative complication | ||
| Bowel obstruction | 3 (4.5) | 0 (0.0) |
| Gastrointestinal bleeding | 1 (1.5) | 0 (0.0) |
| Acute liver failure | 1 (1.5) | 0 (0.0) |
| Pneumonia | 15 (22.4) | 1 (2.5) |
| Prolonged air leak | 11 (16.4) | 0 (0.0) |
| Pneumothorax | 6 (9.0) | 0 (0.0) |
| Pleural effusion | 3 (4.5) | 0 (0.0) |
| Atrial fibrillation | 2 (3.0) | 2 (5.0) |
| Acute kidney injury | 1 (1.5) | 0 (0.0) |
| Stroke | 1 (1.5) | 0 (0.0) |
| Myasthenia crisis | 0 (0.0) | 1 (2.5) |
| Pulmonary embolism | 0 (0.0) | 0 (0.0) |
| Postoperative pneumoperitoneum | 12 (17.9) | 4 (10.0) |
| 30-day mortality | 3 (4.5) | 0 (0.0) |
| 30-day readmission | 5 (7.5) | 3 (4.5) |
| Reoperation within 30 days | 1 (1.5) | 0 (0.0) |
Data are presented as mean ± standard deviation or n (%). POD, postoperative days.
The readmission rate (within 30 days post-operatively) was 8.4% and only one patient had a reoperation with 30 days post-operatively due to prolonged air leak; 17.9% of the patients had a pneumoperitoneum described on the post-operative chest X-ray in the major lung resection group and 10% in the thymectomy/anterior mediastinal mass resection group. It was always treated conservatively.
Regarding the specific digestive complications, a post-operative ileus requiring a nasogastric tube insertion was seen for three patients in the major lung resection group and for zero patients in the thymectomy/anterior mediastinal mass resection group. One patient suffered from gastrointestinal hemorrhage due to bleeding from a stomach ulcer. It was treated endoscopically with clip application.
Postoperative control in clinic
At the 30-day post-operative consultation, no patients were taking painkillers and no patients recorded neuropathic pain. No abdominal pain or digestive complaints were reported, and no bowel obstruction or pneumoperitoneum was observed.
Discussion
Despite accumulating evidence that the subcostal OTC approach can reduce postoperative pain and opioid consumption compared with conventional intercostal RATS, it has not yet become widely accepted in the thoracic surgery community, partly because of concerns regarding technical complexity and reproducibility. In this context, the primary aim of the present prospective study was not to re-evaluate clinical outcomes, but to examine how an experienced robotic thoracic surgeon adapts to this new access route by analyzing the learning curves for both docking and console times.
The CUSUM analysis showed a typical three-phase learning process for docking time, with an initial training phase, an improvement phase and a mastery phase for both major lung resections and thymectomy/anterior mediastinal mass resections. In contrast, no distinct learning curve was observed for console time: the CUSUM curves remained essentially linear, suggesting stable operative performance from the outset. Taken together, these findings indicate that the main adaptation required when adopting the OTC approach concerns the docking phase, whereas once the robot is docked the surgeon operates in a configuration that is very close to a standard transthoracic RATS setup. Inside the chest, anatomical landmarks and surgical steps are identical to those of the conventional technique, with the main difference being a slightly more tangential view of the thoracic cavity. This likely explains why console skills acquired with the standard intercostal approach appear fully transferable to the subcostal configuration, and why no additional console learning curve was detected in our experienced surgeon.
Our operative and docking times also support the idea that the OTC configuration can be integrated into an established robotic program without major loss of efficiency. For major lung resections, the mean console time in our series (around 102 minutes) was within the range reported for conventional transthoracic RATS (13,14) and shorter than that reported in the first OTC series using a similar technical approach (10). For thymectomy and anterior mediastinal mass resections, the mean console time (around 99 minutes) was comparable to or shorter than that reported for subcostal VATS or multi-institutional robotic thymectomy series (5,15). Docking times around 17 minutes in both groups, with mastery after approximately 30–40 cases, are very similar to those described in transcostal RATS learning-curve series (16,17), suggesting that adoption of the subcostal configuration does not impose an additional docking burden compared with a standard transthoracic robotic program. As in transcostal RATS, the initial learning period mainly reflects the time required to standardize the subcostal incision, passage of the trocars through the abdominal wall and diaphragm, and adjustment of the boom position. Once these elements were standardized, docking became highly reproducible for the whole team.
Regarding safety, conversion to open surgery was rare and related to tumor extension rather than to the access route, and perioperative outcomes are detailed in Tables 2,3. Pneumoperitoneum and bowel obstruction in the immediate postoperative period are reported in Table 3, and no abdominal pain, bowel obstruction or pneumoperitoneum was observed at the 30-day postoperative visit. These findings are consistent with our previously published matched-cohort study, which showed improved pain outcomes without an excess of complications when using the OTC approach, and suggest that the subcostal configuration can be implemented safely when introduced by an already experienced robotic surgeon. In addition, although they were not included in the present analysis, 22 patients in our series were operated on entirely by a fellow using this approach without major complications, supporting its reproducibility and potential for teaching.
This study has several limitations. It is a single-center experience and the learning-curve analysis is restricted to one surgeon who was already highly experienced in transthoracic RATS; the results may therefore not be generalizable to surgeons at the beginning of their robotic experience. Furthermore, we focused deliberately on technical learning, assessed by docking and console times, and did not perform a formal comparative analysis of peri- and postoperative outcomes versus the standard intercostal approach, as these outcomes are currently being evaluated in a separate prospective study. Detailed pain data have already been reported in a previous matched-cohort study (11), and long-term results, including chronic pain and oncological outcomes, were beyond the scope of this work.
Despite these limitations, the strengths of our study include its prospective design, the homogeneous use of a single robotic platform and subcostal configuration, and the absence of inter-operator variability. For an experienced robotic thoracic surgeon, our findings suggest that the OTC subcostal approach is feasible and rapidly adoptable, with a short learning curve confined to docking maneuvers and no additional console learning curve. This information may help reassure other teams that, apart from mastering the specific subcostal docking, they will find a familiar operative environment at the console, which should facilitate wider evaluation and adoption of this pain-reducing technique in future comparative studies.
Conclusions
This study shows that for an experienced senior consultant in robotic thoracic surgery, the OTC technique appears to be doable, safe and quickly mastered. However, more data are necessary to confirm these preliminary results. Future comparative studies are warranted to directly compare the OTC subcostal approach with the standard intercostal transthoracic robotic technique in terms of peri- and postoperative outcomes.
Supplementary
The article’s supplementary files as
Acknowledgments
None.
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 and its subsequent amendments. This study was reviewed and approved by institutional review board of University Regional Hospital of Nancy (IRB No. 2024PI016-489). Informed consent was taken from all the patients.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2062/rc
Funding: None.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2062/coif). The authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2062/dss
References
- 1.Lim E, Batchelor T, Shackcloth M, et al. Study protocol for VIdeo assisted thoracoscopic lobectomy versus conventional Open LobEcTomy for lung cancer, a UK multicentre randomised controlled trial with an internal pilot (the VIOLET study). BMJ Open 2019;9:e029507. 10.1136/bmjopen-2019-029507 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.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. 10.1016/j.athoracsur.2013.07.117 [DOI] [PubMed] [Google Scholar]
- 3.Marshall K, McLaughlin K. Pain Management in Thoracic Surgery. Thorac Surg Clin 2020;30:339-46. 10.1016/j.thorsurg.2020.03.001 [DOI] [PubMed] [Google Scholar]
- 4.Kido T, Hazama K, Inoue Y, et al. Resection of anterior mediastinal masses through an infrasternal approach. Ann Thorac Surg 1999;67:263-5. 10.1016/s0003-4975(98)01210-7 [DOI] [PubMed] [Google Scholar]
- 5.Rao M, Salami A, Robbins A, et al. Subxiphoid-subcostal versus transthoracic thoracoscopic thymectomy: A safe and feasible approach. JTCVS Tech 2022;16:172-81. 10.1016/j.xjtc.2022.08.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Liu CC, Wang BY, Shih CS, et al. Subxiphoid single-incision thoracoscopic left upper lobectomy. J Thorac Cardiovasc Surg 2014;148:3250-1. 10.1016/j.jtcvs.2014.08.033 [DOI] [PubMed] [Google Scholar]
- 7.Park SY, Lee JH, Stein H, et al. Initial experience with and surgical outcomes of da Vinci single-port system in general thoracic surgery. J Thorac Dis 2022;14:1933-40. 10.21037/jtd-21-1739 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Vishnu R, Jung W, Chia-Hui Shih B, et al. Subcostal robotic-assisted pulmonary resection: Left lingular segmentectomy. JTCVS Tech 2022;16:149-52. 10.1016/j.xjtc.2022.08.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lee JH, Park TH, Kim HK. Robotic thoracic surgery using the single-port robotic system: Initial experience with more than 100 cases. J Thorac Cardiovasc Surg 2024;168:1513-1522.e2. 10.1016/j.jtcvs.2024.03.005 [DOI] [PubMed] [Google Scholar]
- 10.Bulgarelli Maqueda L, Têtu M, Guimarães Rocha Lima P, et al. Outside the Cage Subcostal RATS Lobectomy: Technical Aspects and Results of the First Series of a Novel Approach to Pulmonary Lobectomy. Innovations (Phila) 2023;18:519-24. 10.1177/15569845231217257 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Streit A, Walsh LC, Stasiak F, et al. Pain outcomes of outside-the-cage robotic thoracic surgery: a prospective matched-cohort study. J Robot Surg 2024;18:351. 10.1007/s11701-024-02108-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wohl H. The cusum plot: its utility in the analysis of clinical data. N Engl J Med 1977;296:1044-5. 10.1056/NEJM197705052961806 [DOI] [PubMed] [Google Scholar]
- 13.Yang F, Chen L, Wang H, et al. Optimizing surgical precision: a comparative study of three-port vs. four-port robotic-assisted lobectomy for NSCLC. J Cardiothorac Surg 2025;20:184. 10.1186/s13019-025-03436-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhang J, Wang Z, Wang Y, et al. Comparison of Long-Term Survival Between Robotic and Video-Assisted Lobectomy for Stage I NSCLC With Radiologic Solid Tumors: A Propensity Score Matching Study. Clin Lung Cancer 2025;26:e63-e72.e2. [DOI] [PubMed]
- 15.Marulli G, Maessen J, Melfi F, et al. Multi-institutional European experience of robotic thymectomy for thymoma. Ann Cardiothorac Surg 2016;5:18-25. 10.3978/j.issn.2225-319X.2015.08.13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Song G, Sun X, Miao S, et al. Learning curve for robot-assisted lobectomy of lung cancer. J Thorac Dis 2019;11:2431-7. 10.21037/jtd.2019.05.71 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Meacci E, Nachira D, Congedo MT, et al. Learning Curve of Robot-Assisted Thymectomy: Single Surgeon's 7-Year Experience. Front Surg 2022;9:860899. 10.3389/fsurg.2022.860899 [DOI] [PMC free article] [PubMed] [Google Scholar]