Ivor‐Lewis oesophagectomy: A standardized operative technique in 11 steps

Standardization of robotic oesophagectomy can benefit both patients and surgeons by decreasing complications, shortening the learning curve and improving surgical training.


| BACKGROUND
Oesophageal resection with en-bloc lymphadenectomy is the cornerstone of multimodal therapy of locally advanced oesophageal carcinoma. 1 The use of minimally invasive techniques (MIE) has become increasingly important because it has been shown that MIE reduces surgical trauma and postoperative morbidity with the same surgical radicality. [2][3][4] However, conventional thoraco-laparoscopic procedures are limited by two-dimensional vision and limited freedom of movement. These technical limitations can impair the feasibility of MIE and its further establishment, 5,6 and lead to a long learning curve and greater risks for the patient. 7 In recent years, robot-assisted technology has been increasingly used in oesophageal surgery. Limitations of conventional MIE techniques can be overcome by the greater mobility of the instruments, the guided, high-resolution camera and the surgeon's ability to work with a total of four arms. 8 Numerous reports have been published on the technique for robot-assisted MIE (RAMIE) and early postoperative outcomes. 5,9,10 However, there is inconsistent use of the term 'robotic-assisted technique', and some centres actually report a hybrid technique involving conventional laparoscopic or thoracoscopic techniques. 11,12 Furthermore, reported techniques for the anastomosis differ greatly, and there are many different options regarding the positioning of the trocars and the incision for specimen removal. Therefore, the procedure is anything but standardized. [13][14][15] Our centre was the first German centre to perform a completely robot-assisted Ivor-Lewis oesophagectomy. 8,16 Since then, we have gradually adapted and standardized our technique. We also played a leading role in a joint project with seven German robotic centres, where we drafted a proposal for a standard version of the operation. 17 One way to adjust the learning curve step by step, and to ensure the safety of the patients, is modular division of the operative procedure into individual surgical steps. This has been successfully demonstrated in colorectal surgery 18 and pancreatic surgery. [19][20][21] Based on our experience with over 250 RAMIE procedures and more than 2000 robotic procedures overall, we have divided this operation into 11 steps (see Table 1). The surgeons train using these steps in modules, until the entire robotic procedure can be performed. The aim of our study is to present these steps individually, and to demonstrate the helpful tips and tricks that in our experience lead to a rapid increase in the learning curve.

| Patient selection
At the beginning of a robotic program, attention should be paid to patient selection. We recommend selecting patients with few comorbidities, a BMI <30, and small tumours without infiltration of other structures. As approximately 80% of patients with oesophageal cancer are treated with neoadjuvant chemotherapy or combined radio-chemotherapy, the majority of patients in our cohort has received neoadjuvant treatment. When implementing this technique, we would advise selecting patients that have not received any neoadjuvant treatment as this can increase technical difficulty. With increasing surgical experience, restrictions can be lifted.

| Operating room configuration
We use the DaVinci Xi System™ (Intuitive) for the Ivor-Lewis procedures. Besides standardization of the procedure itself, it is essential to standardize the operating room setup. The recommended configuration of the operating room is as shown by Egberts et al. 17

| Patient positioning
The patient is placed in a supine position and intubated with a double-lumen endotracheal tube. After completion of the abdominal

| Thoracic phase
The ports for the thoracic phase are placed as follows: R4 (da Vinci   Table 2).
For clear visualization of the operative field, it is essential to retract the liver. We use a method for static retraction that allows all da Vinci instrument arms to be free for dynamic use and localized static retraction where necessary. A liver retractor (e.g., Reveel Endoscopic Retractor; Retraction Limited) or laparoscopic grasper is placed through a 5-mm port in the right flank (LR). The liver is usually sufficiently elevated when this hook is fixed to the lateral drape with a clamp.

| Step 3: Gastric and distal oesophagus mobilization
For gastric mobilization, we follow a medial to lateral approach

| Step 4: Gastric tube construction
The  After preparation of the gastric tube, the end of the first stapling line is tagged with a Vicryl suture as a marker. This marker helps to estimate the stomach tube length after the gastric pull up.

| Step 5: Feeding jejunostomy
We regularly perform a percutaneous jejunostomy at the end of the

| Step 6: Trocar placement, robot docking, and transection of the azygos vein and thoracic duct
Single-lung ventilation is introduced, and the port placement is executed as shown in Figure 4. The da Vinci Xi ® patient cart is also docked from the right side of the patient. Once docked, targeting is performed using the azygos vein as the target. The endoscope is inserted via Arm 3 and will later be switched to Arm 2 for anastomosis.

| Step 7: En-bloc resection of the oesophagus and lymphadenectomy
Depending

| Step 8: Insertion of the stapler anvil and purse-string suture
Different options for intra-thoracic oesophagogastrostomy have been described in literature. We will describe the circular stapled anastomosis, as we find this to be the best option. We recommend using a minimum diameter of 28 mm, so that enough tissue can be grabbed to avoid the 'doughnut' reaching into the staple line. During this part of the procedure, it is helpful to optimize the view by switching the endoscope to R1 or R2.  Figure 8A). The anvil is secured with a purse-string suture using a StratafixTM Spiral ® (Ethicon). This makes it possible to control and precisely dose the tension after each stitch. The stapler anvil is then secured with a second suture using a Roeder-Loop (Ethicon) ( Figure 8B).

| Step 9: Gastric tube pull-up
The gastric tube is pulled up into the thoracic cavity through the hiatus. This manoeuvre should be done extremely carefully because of the risk of damaging the fragile gastric tube. We recommend performing the pull-up with laparoscopic graspers via the assistant port because of the lack of haptic feedback from the robot. Special attention should be given not to twist the gastric tube, by keeping the lesser curvature toward the lateral chest wall. The conduit should be carefully grasped without damaging the right gastroepiploic arcade. Sometimes, it is necessary to pull up the omentum majus step-by-step in obese patients because it can be blocked at the hiatus. The stomach is pulled into the thoracic cavity through the extended hiatus up to the Vicryl marker stitch, which ensures that the maximum available length of the stomach tube is utilized.

| Step 10: Circular stapled intrathoracic anastomosis
Now the intact minor gastric curvature of the transected specimen is brought out of the chest, through the previously created mini thoracotomy. The perfusion of the gastric conduit and oesophageal  Figure 9B). Ensure that a tension-free, non-twisted gastric tube with adequate perfusion is created. The gastric conduit is then straightened by the assistant, the stapler anvil is carefully grabbed with the large needle driver via Arm 4, and the anvil shaft slid over the trocar to perform the anastomosis ( Figure 9C). ( Figure 9D). We think this is mainly due to the inconsistent use of the term 'robotic-assisted, minimally invasive oesophagectomy' in the literature and the lack of a standardized procedure. 8 The advantages of Use of a standardized technique with high-quality standards for robotic oesophagectomy is necessary to decrease the chances of related complications and to benefit from the advantages of minimally invasive surgery. A recent study has shown that the learning curve of RAMIE can be decreased from 70 to 24 cases when following a structured training program. 15 Thus, our standardized approach could lead to a shorter learning curve, improved surgical training, and shorter operative times-benefitting patients and surgeons and easing implementation into other centres. Standardization could also result in a progressive reduction in morbidity, with possibly better oncological outcomes. 9 As more centres use a standardized technique, results will become homogenous and reproducible, and will enable the much-needed multi-centre, randomized controlled trials that are necessary before this technique can become the gold standard for minimally invasive surgery in oesophageal cancer. Honest and intensive exchange with comparison of the results of specialized centres is essential. With standardization and use of the dual console DaVinci Xi, it is easily possible to use a stepwise approach to introduce this technique to other surgeons.

| Procedure completion
Nevertheless, it is important that the entire surgical team is trained and has sufficient experience in robotic surgery. 24,25