Surgical outcomes of 500 robot-assisted minimally invasive esophagectomies for esophageal carcinoma

Introduction

Radical esophagectomy with lymphadenectomy is the main therapeutic modality for localized esophageal cancer. However, open esophagectomy (OE) with lymphadenectomy has considerably high rates of morbidity (26–41%) and mortality (4–10%) (1).

Recently, robot-assisted minimally invasive esophagectomy (RAMIE) has increasingly been applied to surgery for esophageal cancer (2). RAMIE provides convenience to the operator by enabling precise and meticulous dissection with a 10-time magnified three-dimensional field view and an ambidextrous articulated instrument. It also enables dissection of the entire mediastinum through a superior surgical view at the hiatus and thoracic inlet level in esophageal cancer surgery, which is helpful especially for upper mediastinal lymph node (LN) dissection involving the recurrent laryngeal nerve (RLN) (3). Additionally, it enables robot-sewn anastomosis within the thoracic cavity. However, intrathoracic robot-sewn anastomosis is a highly challenging procedure, and a multicenter study demonstrated a disappointing anastomotic leakage (AL) rate of 33% (4). Oncologic outcomes of RAMIE are comparable to outcomes of open surgical methods (5,6), with lower postoperative morbidity, better short-term quality of life and functional recovery, and lesser postoperative pain than with OE (6).

However, RAMIE requires sufficient experience to gain technical proficiency. Studies on the learning curve of RAMIE have shown that at least 20 to 80 procedures are required to achieve proficiency (7-9).

In this study, we aimed to identify the experience needed to reduce complications and become technically proficient in RAMIE through the learning curves. In addition, we described the surgical and oncologic outcomes after RAMIE based on an experience of 500 cases. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-637/rc).

Methods

Patients

We retrospectively reviewed 503 consecutive patients with primary esophageal cancer who underwent the RAMIE procedure, performed by Kim YH for esophageal cancer, from December 2008 to February 2021. Patients with histology other than esophageal carcinoma (n=3) were excluded. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The Institutional Review Board of Asan Medical Center approved this study (IRB No. 2021-1812; approval date: December 24, 2021). The need for informed consent was waived owing to the retrospective nature of study and patient data anonymization.

The contraindications for RAMIE procedure were patients with expected severe lung adhesion, and with highly suspected direct adjacent structural invasion of tumors and metastatic LNs in preoperative imaging modalities. However, if re-evaluation of imaging tests performed 1 month after neoadjuvant chemoradiotherapy (nCRT) showed remarkable remission and clear boundaries between adjacent structures, we included them as candidates for RAMIE. Finally, we reviewed 500 patients who underwent RAMIE to identify the learning curves, the surgical and oncologic outcomes.

Surgical technique of RAMIE

RAMIE was initiated using the da Vinci S system in 2008. Since the introduction of the da Vinci Xi system, RAMIE has been mainly carried out using the da Vinci Xi surgical system (Intuitive Surgical, Inc., Sunnyvale, CA, USA).

The RAMIE procedure consists of thoracic esophagectomy by robotic system, abdominal, and with or without cervical procedure. During the abdominal procedure, conduit was formed using the stomach, jejunum, or colon. Gastric conduit formation using a minimally invasive technique was performed by an experienced gastric surgeon. However, during the learning phase, there were instances of conduit-related issues, leading to the temporary use of an open laparotomy approach by the thoracic surgeon for conduit formation. In the developing phase, a change in the conduit formation strategy was introduced. The thoracic surgeon provided instructions to the gastric surgeon to create a gastric conduit using the same strategy and perform laparoscopic or robotic surgery entirely starting from the 82nd case.

The sequential thoracic procedure was performed in the left semi-prone position. Three 8-mm ports were utilized for the robotic arms, while one 12-mm port was designated for the first assistant. Carbon dioxide (CO₂) gas insufflation into the thoracic cavity was carried out as a standard procedure, maintaining a CO₂ pressure of 6 mmHg. For intra-thoracic anastomosis, the camera port at the 5th intercostal space was extended by approximately 4 cm to allow for the insertion of a circular stapler into the thoracic cavity. The specimen was subsequently extracted through this port. Following partial transection of the esophagus at the highest level of the thoracic esophagus, a purse-string suture was performed using a robotic arm equipped with Maryland forceps and a needle holder. The anvil of the stapler was inserted into the proximal esophagus, and the purse-string suture was secured. The conduit in the abdomen was then pulled up into the thoracic cavity, and an esophago-gastric anastomosis was performed. Routinely, anastomosis of the conduit and proximal thoracic esophagus was performed at the level of the brachiocephalic artery. For anastomosis, the 25 mm sized circular stapler was routinely considered first, while manual suture anastomosis was considered for relatively short conduits. There is no specific definition for “relatively short” in terms of a precise length. When a thoracic surgeon pulls up the conduit to the anastomosis site and there is a lack of length and a tight sensation, it is described as “relatively short”. When the surgeon perceived the conduit to be “relatively short”, we documented this in the surgical records. Fuchs et al. also showed that intra-thoracic esophagogastrostomy is possible by extending the incision of the previously used port without rib cutting (10). Intra-thoracic anastomosis was performed if the tumor was located below the carina. Otherwise, the anastomosis was performed at the cervical esophagus or above. The description of the anastomosis site can be found in Table S1. Bilateral cervical node dissection (CND) was performed by a head and neck surgeon. The indications of CND are as follows; advanced T-stage upper thoracic esophageal cancer, cervical esophageal cancer, or a suspected/confirmed cervical/highest mediastinal LN metastases. The detailed technique of RAMIE in our center has been previously reported (3).

Postoperative management

Sips of water were commenced from postoperative day 3, and if there were no significant events during the surgery, the resection margin of the anastomosis exhibited a complete donut shape, there are no signs of aspiration or leakage, and there is no sudden high fever, then the diet was advanced on the next day. Esophagography is a crucial examination for detecting postoperative anastomotic leakage. However, there have been instances where esophagography could not be conducted on time due to scheduling constraints, leading to delayed initiation of oral intake or potential complications such as aspiration pneumonia caused by the contrast medium. Esophagography to evaluate anastomosis site leakage and conduit was only performed on patients with suspected leakage because in patients with no intraoperative event, the sensitivity to detect AL may be low (5).

If symptoms of hoarseness were observed in the patient and vocal cord palsy (VCP) was suspected, a vocal cord examination was performed by head and neck specialist. In the case of unilateral VCP, hyaluronic acid (HA) injection was recommended regardless of whether the patients’ aspirated or not, to reduce respiratory complications following surgery and to reduce the anxiety of hoarseness. In the case of bilateral VCP, if the airway was well preserved, the patient was carefully observed, and the diet was proceeded with caution; otherwise, tracheostomy was performed.

Postoperative complications and outcome assessment

Most complications were evaluated based on the joint definitions of the Society of Thoracic Surgeons and the European Society of Thoracic Surgeons (11). The complications of AL were evaluated according to the Esophageal Complication Consensus Group guidelines (ECCG) (12). Minor leak of anastomotic leakage that does not require surgical therapy corresponds to anastomotic leak type I and type II of ECCG, and major leak that requires surgical therapy corresponds to type III of ECCG (12).

Statistical analysis

Statistical analysis was performed using SPSS, version 18.0 for Windows (IBM Corp, Armonk, NY, USA). All P values <0.05 were considered statistically significant. The χ² test or Fisher’s exact test were used to compare categorical variables. The Student’s t-test was used for comparisons of continuous variables between two groups. To compare continuous variables among groups, the one-way analysis of variance was used.

To evaluate oncologic outcome, overall survival and disease-free survival of RAMIE were obtained using Kaplan-Meier analysis. The definition of disease-free survival is the length of time a patient survives without signs or symptoms of cancer after operation. A cumulative sum (CUSUM) curve was created to calculate the learning curve (13). The significant change point on this curve was the highest or lowest peak. A positive slope indicates that the value corresponding to this point is greater than the average, and a negative slope corresponds to a value lower than the average.

Results

Patient characteristics

Between December 2008 and February 2021, 500 consecutive patients who underwent RAMIE were evaluated. The baseline patient characteristics are presented in Table 1. The mean age was 63.3±8.2 years. Most of the patients were male (n=447, 89.4%). Most histology types were squamous cell carcinoma (n=486, 97.2%). There were 17 (3.4%), 65 (13.0%), 169 (33.8%), and 249 (49.8%) patients with cervical, upper thoracic, mid thoracic, and lower thoracic esophageal tumors, respectively. The nCRT was administered to 113 patients (22.6%). In the clinical stage, cT1 was the most common with 335 cases (67.0%), and clinical N positive was 125 cases (25.0%).

Operative and pathologic outcomes

Operative and pathologic outcomes are summarized in Table 2. The mean total operation time and thoracic phase time were 391±81 and 146±43 minutes, respectively. The sites of anastomosis varied the thoracic esophagus, and the cervical esophagus, accounting for 282 cases (56.4%), and 202 cases (41.4%), respectively. Regarding conduits, the stomach accounted for the most with 479 cases (95.8%). Regarding the anastomosis method, circular end-to-end anastomosis (EEA) was used in 467 cases (93.4%). Three-field LN dissection was done in 89 cases (17.8%). The mean number of harvested LNs was 34.4±13.6.

Postoperative complications

The postoperative morbidity and mortality are presented in Table 3. The median length of hospital stay was 14 [12–18] days. VCP occurred in 148 (29.6%) cases. Most of the VCPs were transient (89.1%). The rates of AL and anastomotic stricture (AS) were 6.6% and 17.6%, respectively.

Variables related to AL and AS are shown in Tables S2,S3, respectively. The rate of AL was lower in the cervical esophagus (4.0%), than in the thoracic esophagus (7.8%). Regarding the conduit, the stomach showed the lowest leakage rate (5.6%), and nCRT increased the leakage rate (12.4%). When intra-thoracic anastomosis was performed, the incidence rate was the lowest (9.9%), and the short conduit length increased the incidence of stricture (34.8%). The rate of stricture was low in patients receiving nCRT (7.1%). VCP was associated with anastomosis site, and CND (Table S4). The incidence of VCP was lower in intra-thoracic anastomosis (23.4%) than in other sites. CND increased the incidence of VCP (26.4%). Re-operation within 90 days was performed in 38 (7.6%) patients. The most common re-operation was thoracic duct ligation (n=10) (Table S5).

Assessment of learning periods

The CUSUM curve for variables showed each significant change point, which is shown in Table 4 and Figures 1,2. The change points for the number of harvested LNs, and the rate of VCP and AL were about 50^th case. After 150 cases, significant improvements in thoracic and total procedure time, reoperation within 90 days, length of stay, and AS were observed. Through these change points, the learning period could be classified into the learning, the developing, and the stable phase.

Figure 1 Cumulative sum analysis curves for RAMIE. (A) Harvested lymph nodes, (B) vocal cord palsy, (C) anastomotic leakage, and (D) thoracic procedure time. Approximately 50 procedures are required to achieve early proficiency. CUSUM, cumulative sum analysis; RAMIE, robot-assisted minimally invasive esophagectomy.

Figure 2 Cumulative sum analysis curves for RAMIE. (A) Reoperation within 90 days, (B) length of stay, (C) anastomotic stricture, and (D) total operation time. Extensive experience of more than 150 procedures is needed for quality stabilization. CUSUM, cumulative sum analysis; RAMIE, robot-assisted minimally invasive esophagectomy.

Surgical outcomes according to learning periods

With increasing proficiency, the total operation time, thoracic procedure time, and length of stay were significantly reduced, and the incidence of intraoperative transfusion, re-operation within 90 days, and AS decreased. As experience accumulated, the number of patients receiving nCRT and CND increased gradually. As the phase changed, the main operation type changed from the McKeown operation to the Ivor Lewis operation (Table 5).

Oncologic outcomes

Median follow-up period was 35 months (range, 17–57 months). The 5-year overall survival rate was 70.6%, and the 5-year disease-free survival rate was 70.6% (Figure S1).

Discussion

In the current study, we described the surgical and oncologic outcomes of 500 cases of RAMIE. The learning period for RAMIE was divided into 3 phases according to CUSUM analysis. As the most important factors for beginners, improvement of VCP, AL and an increase in the number of harvested LNs were evident at the beginning of 50 cases. After experience was sufficiently accumulated, change points of operation time, reoperation within 90 days, length of stay, and AS appeared after 150 cases.

Previously, there have been excellent studies on the learning curve and surgical outcomes of RAMIE. Sarkaria et al. demonstrated tips that were useful during surgery and management of critical complications from 100 cases of totally minimally invasive RAMIE (14). Han et al. analyzed the learning curve in 124 cases of Ivor Lewis RAMIE and it demonstrated that 51 cases were required to gain technical proficiency (15). Yang et al. analyzed the experience of 400 McKeown RAMIE and demonstrated that the 40 cases comprised the learning phase and the quality outcomes improved after 80 cases (9). To our knowledge, our 500 cases of RAMIE performed by a single surgeon constitute one of the largest experiences. Our study showed changes in the type of operation performed by a surgeon as he became proficient in RAMIE and the flow of changes in surgical outcomes.

There were three notable changes in the stable phase. First, the increase in the Ivor Lewis operation in the stable phase indicates that it was possible to apply an appropriate surgical method according to the location of esophageal cancer. It means that unnecessary incisions and damage to the patient can be reduced at the cervical level, thereby reducing complications (16,17). The VCP rate was 23.4% for intra-thoracic anastomosis and 37.8% for anastomosis at cervical level or above, with the intra-thoracic anastomosis rate being significantly lower (P=0.001). Compared to previous studies, the rate of vocal cord palsy in this study is higher at 29.6%. This elevated rate can be attributed to the extensive en bloc lymph node dissection in the superior mediastinum. The improved accessibility afforded by RAMIE enables surgeons to reach deeper regions of the thoracic cavity, which would not have been possible with conventional surgery. As a result, we performed nearly 35 lymph node harvesting procedures during RAMIE, a significantly higher number compared to other studies where the average ranged from 18 to 30 lymph nodes (18,19). Similar findings have been reported in other studies, where vocal cord palsy occurred in 26% and 29% of cases after 44 and 38 lymph node harvesting procedures, respectively (20,21). Compared to anastomosis at cervical or above level group (20.4±20.0 d), the intra-thoracic anastomosis group (16.9±15.5 d) had a significantly shorter mean length of hospital stay (P=0.029). In this study, as the rate of the Ivor Lewis operation increased, conduit related complications decreased. Intra-thoracic anastomosis has a benefit in terms of the length of the conduit, which can reduce the incidence of stricture. In contrast to general expectations, our study revealed a higher rate of intrathoracic AL compared to cervical anastomosis (7.8% vs. 4.0%). Performing intrathoracic anastomosis in the context of RAMIE can be a challenging procedure. During the learning phase, the implementation rate of RAMIE Ivor-Lewis was low due to a lack of proficiency, resulting in a higher incidence of AL. However, the cervical procedure of RAMIE McKeown, which was similar to open McKeown, was performed with proficiency, leading to a lower incidence of AL. As the proficiency in RAMIE increased, we observed a gradual increase in the number of intrathoracic anastomoses and a decrease in the incidence of intrathoracic AL. This finding was evident in the stable phase, where despite a rise in the proportion of Ivor-Lewis RAMIE (67%), there was a decrease in the overall occurrence of AL. Nevertheless, we consider the 7.8% rate of intrathoracic AL to be an acceptable outcome (Table S6).

Second, the number of patients who received nCRT for locally advanced esophageal cancer increased in the stable phase. Due to the dense adhesion between the esophagus and surrounding tissues induced by nCRT (22), it might be difficult to find the surgical plane, especially between the trachea, left main bronchus and esophagus. As the surgeon’s experience grew, the number of challenging cases with RAMIE also increased. In patients with suspected T4 invasion, if there was a therapeutic response after nCRT and the tumor was deemed resectable, OE was initially recommended to address the injury to the surrounding structures during dissection when the experience with RAMIE was insufficient. As experience with RAMIE increased, the ability to perform meticulous and precise dissection improved, thanks to the magnified surgical view and articulated surgical arm, which minimized damage to surrounding structures. This can be attributed to the ease of accessing and the enhanced accuracy in performing mediastinal lymph node harvest during RAMIE. Weindelmayer et al. demonstrated that RAMIE allows for a more accurate lymphadenectomy, resulting in a higher number of thoracic lymph nodes compared to OE, without an increase in other complications (23). Therefore, based on these experiences, it can be concluded that RAMIE is a feasible approach for locally advanced esophageal cancer.

Third, the number of patients receiving CND in the stable phase increased, which was associated with the increased number of patients with locally advanced esophageal cancer described above. This is the reason why the rate of VCP increased in the stable phase. However, when it comes to long-term outcome, the permanent VCP rate decreased in the stable phase, suggesting that most nerve injuries were transient palsy due to traction or thermal injury (Table 5). However, in terms of long-term outcomes, the rate of permanent VCP decreased during the stable phase, suggesting that the majority of nerve injuries were transient and likely caused by traction or thermal injury (Table 5). The relatively high rate of vocal cord palsy in our technique highlights an area that requires improvement, and we are currently making efforts to perform a more meticulous recurrent laryngeal nerve lymph node dissection.

The rate of AL and the details of conduit formation should be taken note of. The rate of AL decreased from 10% in the learning phase to 2% in the developing phase, which might be due to several procedures applied during conduit formation. Gentle manipulation was performed to preserve the serosa of the gastric conduit, and at least 2 cm of omental fat tissue was left in the great curvature (up to the fundus level) to preserve vascular flow. However, the AL rate increased to 7.4% again in the stable phase. We carried out additional analysis to identify the risk factors for AL (see Table S2). As a result, the nCRT was significantly associated with an increase in the incidence rate of AL, which is also reported in several studies (24,25), The increased rate of AL in the stable phase might be attributed to the increase in patients with nCRT. Unlike previous studies (25,26), the rate of AL was higher with intrathoracic anastomosis than with cervical anastomosis. We believe this is related to the high proportion of intrathoracic anastomosis among patients who received nCRT (64.6%, 73/113).

In our study, the rate of AS gradually decreased (Table 5). The incidence of AS was significantly lower when the site of anastomosis was the intra-thoracic esophagus, when the conduit length was not short, and when nCRT was performed (Table S3). Ischemia is an important risk factor for AS (27). If the conduit is short, tension occurs in the conduit at the anastomosis site, which leads to a decrease in the vascular flow. Based on these findings, the decrease in stricture occurrence in the stable phase seems to be influenced not only by the improvement in surgical technique but also by the proportion of the Ivor Lewis operation increases. Most patients who received nCRT underwent intrathoracic anastomosis (64.6%, 73/113), which is thought to be the reason for association of nCRT with lower incidence of AS.

This study had notable limitations. First, selection bias is inherent in a retrospective study conducted at a single institution. Second, RAMIE was performed by a single thoracic surgeon. Therefore, it may be challenging to generalize the learning curve and surgical outcomes of this study to other settings. Third, the clinical outcomes in this study were obtained through a well-established collaborative system involving highly experienced surgeons in their respective fields. Therefore, we believe that the impact of collaboration on the learning curve in our study is likely to be minimal. However, it may be challenging to generalize our learning curves to surgeons who do not have access to a similar collaborative system. Fourth, the patient population in our study has different characteristics compared to esophageal cancer patients in the Western world. Therefore, it may be challenging to generalize our research findings. Fifth, we do not have data on the learning curve or surgical outcomes specifically related to robotic-sewn anastomosis. However, a recent study by Huscher et al. demonstrated excellent results with an AL rate of 10% in 40 patients who underwent intrathoracic robotic-sewn anastomosis (28). This indicates that as proficiency with RAMIE increases, complete robotic Ivor Lewis esophagectomy becomes feasible. Last, it should be noted that the patient population in our study consisted predominantly of early-stage patients who did not receive nCRT. It is important to recognize that as the proportion of patients receiving nCRT increases, the learning curve may be longer, and the surgical outcomes may not be as favorable.

Conclusions

Approximately 50 procedures are required to achieve early proficiency, and extensive experience of more than 150 procedures is needed for quality stabilization. Appropriate anastomosis according to the cancer location can reduce patient morbidity and mortality. Specifically for the Ivor-Lewis operation, the damage caused by additional cervical incision was reduced, and the required conduit length was reduced, resulting in a decrease in tight conduit, which was thought to be related to better outcomes. For surgeons beyond the learning curve, RAMIE could be a safe and feasible option for patients with locally advanced-stage esophageal cancer.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-637/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-637/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-637/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-637/coif). The 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 Institutional Review Board of Asan Medical Center approved this study (IRB No. 2021-1812; approval date: December 24, 2021). The need for informed consent was waived owing to the retrospective nature of study and patient data anonymization.

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. Sabra MJ, Smotherman C, Kraemer DF, et al. The effects of neoadjuvant therapy on morbidity and mortality of esophagectomy for esophageal cancer: American college of surgeons national surgical quality improvement program (ACS-NSQIP) 2005-2012. J Surg Oncol 2017;115:296-300. [Crossref] [PubMed]
2. Kingma BF, de Maat MFG, van der Horst S, et al. Robot-assisted minimally invasive esophagectomy (RAMIE) improves perioperative outcomes: a review. J Thorac Dis 2019;11:S735-42. [Crossref] [PubMed]
3. Yun JK, Chong BK, Kim HJ, et al. Comparative outcomes of robot-assisted minimally invasive versus open esophagectomy in patients with esophageal squamous cell carcinoma: a propensity score-weighted analysis. Dis Esophagus 2020;33:doz071. [Crossref] [PubMed]
4. Kingma BF, Grimminger PP, van der Sluis PC, et al. Worldwide Techniques and Outcomes in Robot-assisted Minimally Invasive Esophagectomy (RAMIE): Results From the Multicenter International Registry. Ann Surg 2022;276:e386-92. [Crossref] [PubMed]
5. van der Sluis PC, Ruurda JP, Verhage RJ, et al. Oncologic Long-Term Results of Robot-Assisted Minimally Invasive Thoraco-Laparoscopic Esophagectomy with Two-Field Lymphadenectomy for Esophageal Cancer. Ann Surg Oncol 2015;22:S1350-6. [Crossref] [PubMed]
6. van der Sluis PC, van der Horst S, May AM, et al. Robot-assisted Minimally Invasive Thoracolaparoscopic Esophagectomy Versus Open Transthoracic Esophagectomy for Resectable Esophageal Cancer: A Randomized Controlled Trial. Ann Surg 2019;269:621-30. [Crossref] [PubMed]
7. Zhang H, Chen L, Wang Z, et al. The Learning Curve for Robotic McKeown Esophagectomy in Patients With Esophageal Cancer. Ann Thorac Surg 2018;105:1024-30. [Crossref] [PubMed]
8. van der Sluis PC, Ruurda JP, van der Horst S, et al. Learning Curve for Robot-Assisted Minimally Invasive Thoracoscopic Esophagectomy: Results From 312 Cases. Ann Thorac Surg 2018;106:264-71. [Crossref] [PubMed]
9. Yang Y, Li B, Hua R, et al. Assessment of Quality Outcomes and Learning Curve for Robot-Assisted Minimally Invasive McKeown Esophagectomy. Ann Surg Oncol 2021;28:676-84. [Crossref] [PubMed]
10. Fuchs HF, Müller DT, Leers JM, et al. Modular step-up approach to robot-assisted transthoracic esophagectomy-experience of a German high volume center. Transl Gastroenterol Hepatol 2019;4:62. [Crossref] [PubMed]
11. Fernandez FG, Falcoz PE, Kozower BD, et al. The Society of Thoracic Surgeons and the European Society of Thoracic Surgeons general thoracic surgery databases: joint standardization of variable definitions and terminology. Ann Thorac Surg 2015;99:368-76. [Crossref] [PubMed]
12. Low DE, Alderson D, Cecconello I, et al. International Consensus on Standardization of Data Collection for Complications Associated With Esophagectomy: Esophagectomy Complications Consensus Group (ECCG). Ann Surg 2015;262:286-94. [Crossref] [PubMed]
13. Yap CH, Colson ME, Watters DA. Cumulative sum techniques for surgeons: a brief review. ANZ J Surg 2007;77:583-6. [Crossref] [PubMed]
14. Sarkaria IS, Rizk NP, Grosser R, et al. Attaining Proficiency in Robotic-Assisted Minimally Invasive Esophagectomy While Maximizing Safety During Procedure Development. Innovations (Phila) 2016;11:268-73. [Crossref] [PubMed]
15. Han Y, Zhang Y, Zhang W, et al. Learning curve for robot-assisted Ivor Lewis esophagectomy. Dis Esophagus 2022;35:doab026. [Crossref] [PubMed]
16. Sabra MJ, Alwatari YA, Wolfe LG, et al. Ivor Lewis vs Mckeown esophagectomy: analysis of operative outcomes from the ACS NSQIP database. Gen Thorac Cardiovasc Surg 2020;68:370-9. [Crossref] [PubMed]
17. Ozawa S, Koyanagi K, Ninomiya Y, et al. Postoperative complications of minimally invasive esophagectomy for esophageal cancer. Ann Gastroenterol Surg 2020;4:126-34. [Crossref] [PubMed]
18. Sarkaria IS, Rizk NP, Finley DJ, et al. Combined thoracoscopic and laparoscopic robotic-assisted minimally invasive esophagectomy using a four-arm platform: experience, technique and cautions during early procedure development. Eur J Cardiothorac Surg 2013;43:e107-15. [Crossref] [PubMed]
19. Cerfolio RJ, Wei B, Hawn MT, et al. Robotic Esophagectomy for Cancer: Early Results and Lessons Learned. Semin Thorac Cardiovasc Surg 2016;28:160-9. [Crossref] [PubMed]
20. Kim DJ, Hyung WJ, Lee CY, et al. Thoracoscopic esophagectomy for esophageal cancer: feasibility and safety of robotic assistance in the prone position. J Thorac Cardiovasc Surg 2010;139:53-9.e1. [Crossref] [PubMed]
21. Park SY, Kim DJ, Yu WS, et al. Robot-assisted thoracoscopic esophagectomy with extensive mediastinal lymphadenectomy: experience with 114 consecutive patients with intrathoracic esophageal cancer. Dis Esophagus 2016;29:326-32. [Crossref] [PubMed]
22. Qin Q, Xu H, Liu J, et al. Does timing of esophagectomy following neoadjuvant chemoradiation affect outcomes? A meta-analysis. Int J Surg 2018;59:11-8. [Crossref] [PubMed]
23. Weindelmayer J, De Pasqual CA, Turolo C, et al. Robotic versus open Ivor-Lewis esophagectomy: A more accurate lymph node dissection without burdening the leak rate. J Surg Oncol 2023;127:1109-15. [Crossref] [PubMed]
24. Klevebro F, Elliott JA, Slaman A, et al. Cardiorespiratory Comorbidity and Postoperative Complications following Esophagectomy: a European Multicenter Cohort Study. Ann Surg Oncol 2019;26:2864-73. [Crossref] [PubMed]
25. Fabbi M, Hagens ERC, van Berge Henegouwen MI, et al. Anastomotic leakage after esophagectomy for esophageal cancer: definitions, diagnostics, and treatment. Dis Esophagus 2021;34:doaa039. [PubMed]
26. Kassis ES, Kosinski AS, Ross P Jr, et al. Predictors of anastomotic leak after esophagectomy: an analysis of the society of thoracic surgeons general thoracic database. Ann Thorac Surg 2013;96:1919-26. [Crossref] [PubMed]
27. Rice TW. Anastomotic stricture complicating esophagectomy. Thorac Surg Clin 2006;16:63-73. [Crossref] [PubMed]
28. Huscher CGS, Cobellis F, Lazzarin G. Intrathoracic Robotic-Sewn Anastomosis During Ivor Lewis Esophagectomy for Cancer: Back to Basics? J Gastrointest Surg 2023;27:1034-41. [Crossref] [PubMed]