Robotic-assisted minimally invasive Ivor Lewis esophagectomy within the prospective multicenter German da Vinci Xi registry trial

The study was designed as a multicenter prospective registry investigator-initiated trial. Five German university centers at Berlin, Dresden, Hamburg, Heidelberg, and Kiel participated in the prospective German da Vinci Xi Registry trial. The trial was in compliance with the ethical principles of Helsinki, and the protocol was approved by the responsible independent ethics committees of all participating centers, i.e., the local ethics committee at TU Dresden (EK296072017), Christian Albrechts University in Kiel (AZD421/13, D451/19), Heidelberg University Faculty of Medicine (S-341/2017), Charité University Hospital in Berlin (EA4/084/17) and University Medical Center Hamburg-Eppendorf (PV5591).

The study was supported with a research grant by Intuitive (Sunnyvale, CA, US).

All consecutive patients who underwent elective RAMIE with an intrathoracic circular stapled end-to-side esophagogastrostomy at each study site between Oct 15, 2017, and Jun 5, 2020, were considered for inclusion if they met the following criteria: age ≥18 years and written informed consent. All patients with esophageal cancer or an indication for esophagectomy who were suitable for a minimally invasive approach using gastric tube reconstruction were considered for the primary robotic-assisted minimally invasive approach. There was no standard selection criteria for the avoidance of the robotic approach. The da Vinci Xi robotic surgical systems were available for all scheduled patients without limitations. One center was delayed in starting RAMIE because of a concurring randomized trial of open versus total minimally invasive laparothoracoscopic esophagectomy. Surgeons’ and patients’ preferences as well as availability of the da Vinci robotic system were taken into account regarding the choice of the procedure in the latter center. The exclusion criteria were emergency operations, patients with a survival prognosis of less than 1 month, operations for which the da Vinci Xi system was not approved (according to the manufacturer), and pregnancy. The study was managed and monitored by the local study centers at the participating sites. Intraoperative documentation was performed by authorized senior surgeons.

The basic steps of the surgical technique used for all operations were published recently [7, 9]. After such consensus of the basic surgical steps was obtained, a proctoring during initial operations was conducted by the most experienced surgeons [14]. All operations were performed on a da Vinci Xi surgical system (Intuitive, Sunnyvale, CA, US). Briefly, the patients were placed in a supine and 15°–20° reverse Trendelenburg position for the abdominal part. The four da Vinci Xi trocars were inserted on a horizontal line usually above the umbilicus supplemented with one additional assistant trocar. Lymphadenectomy included lymph node stations along the hepatic and splenic arteries centrally toward the left gastric artery and the celiac trunk. The left gastric vessels were divided using clips, and the lymph node package ventral to the aorta at the hiatus was resected en bloc with the esophagogastric junction. The gastric tube was trimmed to a semicircumference of 45 mm using linear stapling devices 45 or 60 mm in length beginning at the angular notch. The semicircumference of 45 mm was calculated to allow a circular anastomosis with a 28-–29-mm diameter. Indocyanine green (ICG) fluorescence analysis was routinely performed to define the optimal perfusion margin of the gastric tube. For the thoracic part, the patients were turned into a left lateral to semiprone position, and the right lung was not ventilated. The four da Vinci Xi trocars were placed in a banana-shaped fashion between the fourth and tenth intercostal spaces of the right hemithorax. One additional assistant trocar was used. After docking, the azygos vein was divided, and the esophagus was dissected, including the adjacent paraaortal and mediastinal lymph node stations. The thoracic duct was identified and clipped. If preoperative imaging analysis indicated lymph node metastases along the recurrent nerves, an extended paratracheal lymphadenectomy was performed, e.g., lymph node level two. After complete resection, the esophagus was divided and closed with a purse-string suture. The proximal resection margin of the esophagus was assessed by intraoperative frozen section analysis. At this point, the esophageal resection specimen was extracted using a minithoracotomy at a trocar site, and the stapler anvil was inserted into the oral esophageal stump. Consecutively, the end-to-side esophagogastrostomy was stapled through the minithoracotomy, and the proximal part of the stomach was stapled 2 cm from the esophagogastrostomy using a linear stapling device. The esophagogastrostomy anastomosis was reinforced using either a robotic-assisted running suture or an omental wrap or both. A nasogastric tube was intraoperatively placed distally to the esophagogastrostomy, and one to two chest tubes were placed according to the SOP at each center at the time of operation.

Postoperative morbidity was assessed using the classification of postoperative complications according to Clavien-Dindo [4, 6] with substantiation by the “Japan Clinical Oncology Group” [12]. The following complications were considered: anastomotic leak (AL), pneumonia, respiratory insufficiency, readmission to the intensive care unit (ICU), disorder of gastrointestinal passage, recurrent laryngeal nerve palsy, cardiac arrhythmia, chylothorax, postoperative hemorrhage, wound dehiscence, surgical site infection, intrathoracic fluid collections, mediastinitis, empyema/pyothorax, thromboembolic events, acute renal failure, cardiac decompensation, bacteremia/sepsis, and multiple organ failure.

Based on the recommendations of the Esophagectomy Complications Consensus Group (ECCG), AL was defined as a full-thickness gastrointestinal defect involving the esophagus, anastomosis, stapler line, or conduit irrespective of presentation or method of identification [17]. Diagnosis of AL was made on the basis of contrast leakage on CT, endoscopically or both. For the definition of pneumonia, at least one major and two minor criteria had to be fulfilled (major: new infiltrate in chest imaging; minor: fever >38.5°C or hypothermia <36.5°C, new elevation or permanent high infection markers (leucocytes/C-reactive protein/procalcitonin), productive cough with sputum, pathogen detection) [11].

Textbook outcome was defined as R0 resection with no conversion, a lymph node yield ≥15, no complications of Clavien-Dindo ≥3a, no reinterventions or reoperations, no readmission to the ICU, length of hospital stay ≤21 days, no hospital readmission, and no mortality within 90 days postoperatively [14].

The SPSS (version 27.0.0.0) software package was used for statistical calculation and data plots. The significance level for all calculations was set at p=0.05. The operative time was defined from the start of the operative procedure with a skin incision at the abdomen until skin suturing of the thoracic part, including docking and repositioning times. For learning curve analysis, a subgrouping of 15 consecutive patients was performed for each center. Moreover, we accomplished a case grouping (n=15) for each surgeon with more than 30 cases and studied the median reduction in the operative time of subsequent cases compared with the initial 15 cases. For further investigation of the learning curve, a cumulative sum (CUSUM) analysis of the total operative time was performed. This technique is a graphical method to transform raw data into a running total of differences from the group average. Therefore, a chronological arrangement of all cases from the first to the last by the center (or by the leading surgeon, respectively) was performed. Then, CUSUM values were calculated according to the following formula: CUSUM = Σ (x _i−μ), where x _i is the total operative time of the individual case and μ is the mean operative time of the corresponding center or leading surgeon [30, 36]. Finally, the CUSUM values were plotted on the vertical axis according to their case number on the horizontal axis. Learning curves could be determined by visual interpretation of the chart. The end of the learning curve was predefined as inflection of the curve to a plateau or decrease.

In total, 220 patients were included in the analysis (center 1: 72; center 2: 41; center 3: 83; center 4: 10; center 5: 14). The median age of the patients was 64 years (IQR 58–72), and 85.5% (n=188) were male. Two-thirds of the patients had significant comorbidities (ASA ≥III), and the median BMI was 26.2 kg/m² (IQR 23.6–29.4). No further information of race or ethnicity of the patients was collected routinely in the Xi trial. The indications for esophagectomy were adenocarcinoma in 81.4%, squamous cell carcinoma in 15.5% and other diseases (malignant and nonmalignant) in 3.2% of the cases. Most of the patients had neoadjuvant treatment, including chemoradiation (32.7%) or chemotherapy (47.7%) (Table 1). The distribution of pTNM stage and UICC stage is shown in Table 1.

A totally robotic-assisted operation of both the abdominal and thoracic part (RAMIE) was accomplished in 189 cases (85.9%), whereas a hybrid minimally invasive approach (hRAMIE) with a robotic thoracic part and open abdominal part was performed in 25 cases (11.4%). A hybrid robotic abdominal operation with open thoracotomy was performed in 6 cases (2.7%). Laparoscopy or thoracoscopy was not used alternatively for hybrid approaches. The main reasons for a hybrid approach were a learning phase strategy (n=14), extended dissection for lymph node metastasis (n=4), and adhesions/prior surgery (n=4). Overall conversion to an open procedure was necessary in 16 cases (7.3%). The most frequent reasons for conversion were adhesions (n=5) and intraoperative bleeding (n=4). Extended thoracic resection because of lung infiltration was necessary in 7 cases. One center routinely placed jejunal feeding tubes during the abdominal part. A circular stapler size of ≥28 mm was used in most cases (84%). The median blood loss was 200 ml (IQR 80–400), and the median operative time (OT) was 425 min (IQR 335–527) (Table 2).

Oncological resection with microscopically tumor-free margins (R0) was achieved in 92.9% of patients. Reresection because of a tumor-infiltrated resection margin reported by the intraoperative frozen section examination (which was available for all cases) was required in only two cases. The median number of resected lymph nodes was 25 (IQR 19–30) (Table 2).

Twenty-one percent (n=46 patients) of the total cohort developed major postoperative complications (CDC grade ≥3b) (Table 3). The rate of postoperative anastomotic leakage was 13.2% (n=29). Most patients (82.8%; n=24) with AL were successfully treated using endoluminal approaches (predominantly endoluminal vacuum therapy), whereas reoperation was indicated in 5 cases. Thereby in most cases, an esophageal diversion as well as an insertion of a jejunal feeding tube was performed. Furthermore, 27 patients underwent postoperative endoscopic interventions for indications other than AL. The rate of postoperative AL differed between the centers, and the two centers with >70 cases had leak rates of 7.0% and 12.0% (p=0.213), respectively (Fig. 1).

Postoperative pneumonia was diagnosed in 19.5% of cases. A reoperation, either minimally invasive (n=9) or open (n=15), was indicated in 5 cases because of AL (including esophagobronchial fistula), chylothorax (n=3), postoperative ileus (n=3), pulmonary air leak (n=2), or ischemia of the gastric tube (n=2) (Table 3).

The median postoperative intensive care unit stay was 2 days (IQR 1–4.7), and the median hospital stay was 15 days (IQR 12–24). The 90-day readmission and mortality rates were 8.2% and 3.6%, respectively. The causes of death were multiorgan failure (n=3), aspiration pneumonia (n=1), cardiac tamponade (n=1), hemorrhagic shock (n=1), myocardial infarction (n=1), and tumor progression (n=1).

Overall, 67 patients (31.6%) had a defined textbook outcome with an optimal intra- and postoperative course.

The median total operative time within the first 15 cases of each center (n=69) was 488 min, which was significantly longer than that in the consecutive case groupings (p=0.024). The median operative time subsequently dropped from 439 min (cases 16–30; n=45) to 402 min (cases 31–45; n=41) (p=0.126). Although the median operative time was further reduced to 349 min after >60 cases (n=35), the difference was not statistically significant when compared with cases 46–60 (p=0.089; n=30) (Fig. 2a). Similar results were observed if the median operative time of the thoracic part only was analyzed: the median operative time of cases 1–15 was longer than that of the consecutive case groupings (p=0.023). However, a significant reduction in the operative time for the abdominal part was not seen until a caseload >60 (p≤0.001) (Suppl. Fig. 1). The three most experienced surgeons of the participating centers could significantly improve their individual median operative time by approximately −4.8% after cases 16–30 (n=45) and approximately −11.6% after >30 cases (n=50) (p≤0.021). Likewise, there was a trend toward a further operative time reduction from cases 16–30 to cases >30, but without statistical significance (p=0.057) (Fig. 2b).

The pooled CUSUM graph for all centers showed a peak (inflection point) with a slow decrease after 22 cases, indicating the end of the learning curve for the total operative time (Fig. 3a). The CUSUM graphs for the three centers with more than 22 RAMIE procedures revealed different end points of the learning curve: the inflection point in center 1 was at 22 cases, center 2 reached a plateau after 13 cases, and center 3 reached a plateau after just 10 cases (Fig. 3b). The CUSUM analysis for the leading surgeons of the three most experienced centers showed the same end points of the learning curve of surgeons B and C as for their related centers 2 and 3 (Fig. 3c). This finding is not surprising because the leading surgeon in these two centers performed (nearly) all procedures (78.3% and 100%, respectively). In center 1, three surgeons routinely performed RAMIE, which explains the longer learning curve for this center. The point of inflection for leading surgeon A of center 1 was at the 9th case (Fig. 3c).

Based on the CUSUM analysis, perioperative outcome parameters were compared in relation to the pre- and post-learning curve cohort (≤22 and >22 cases). The latter cohort was operated on with less blood loss (p<0.001), a shorter operative time (p<0.001), and a lower rate of postoperative pneumonia (p=0.046). Additionally, there was a trend toward a lower conversion (11.1 to 4.6%; p=0.061) and 90-day readmission rate (12.2to 5.4%, p=0.059) after >22 cases. Other outcome parameters, including major complications CDC ≥3b, AL rate, textbook outcome, and intensive care parameters, were not significantly different between the two groups (Table 4).

All operated cases (n=220) were included in a univariate analysis to identify predictive factors for the development of AL (Suppl. Tab. S1). However, none of the tested variables significantly correlated with the occurrence of AL.