Since we reported the details of SITG in a previous paper, the SITG with D1+ LND is described briefly here.10 The only point of difference is that in the present case series, we used a laparoscopic scope holder (Laparostat) instead of a scopist (i.e., solo surgery). The patient was placed in the lithotomy position with reverse Trendelenburg positioning. However, the hip and knee joints were straightened and not bent so as not to limit the instruments' movements. The surgeon sat between the patient's legs. The scope holder was place on the left side rail of the operating table and covered by a sterile plastic bag (Fig. 1). This scope holder has a low profile, providing more working space for the surgeon with no interference between surgeon and scopist and with minimal clashes between the instruments and scope near the single port, which is in contrast to the conditions associated with additional human scope manipulation. A commercial 4-hole single port (Gloveport; Nelis, Bucheon, Korea) was inserted into a longitudinal 2.5-cm long transumbilical incision. The abdominal cavity was insufflated with carbon dioxide at a pressure of 11 to 13 mmHg. No additional assistant port was used. No assistant or scopist was present in any of the surgeries. We used a 10-mm flexible high-definition laparoscopic scope (Endoeye flexible HD camera system; Olympus Medical Systems Corp., Tokyo, Japan) that was manipulated by the surgeon. The conventional laparoscopic grasper was used in nearly all procedures while the curved long grasper was used for single-port surgery (Olympus Medical System Corp.) when operating on the lesser curvature side, including suprapancreatic LND. We used a laparoscopic automatic linear stapler (I-drive 45 Purple and 60 Purple; Covidien, Minneapolis, MN, USA). We performed routine total gastrectomy with D1+ LND (1; 2; 3; 4sa, sb, d; 5; 6; 7; 8a; 9; 11p, 11d; and 12a), including partial omentectomy (Fig. 2). After complete exposure of the esophagus by division of the anterior and posterior vagus nerves, the esophagus was transected using a linear stapler (Fig. 3A). The specimen was delivered through the single umbilical incision without any extension.

After the specimen's proximal resection margin was examined histopathologically, we used semi-loop reconstruction to create an intracorporeal anastomosis after SITG. After a jejunal loop located approximately 20 cm distal to the Treitz ligament was transected by a linear stapler (I-drive 60 Tan; Covidien), a side-to-side jejunojejunostomy was created using the I-drive 45 Tan (Covidien) to construct a 40 to 45 cm Roux-en-Y limb. The common entry opening was then closed in double layers by barbed sutures (V-loc; Covidien). Next, the Roux-en-Y limb was brought up via the antecolic route to form an esophagojejunostomy. A stay suture made using 3-0 monofilament thread was passed from the anterior part of the mid-portion of the staple line and retracted through the single port. This traction suture is important for manipulating the esophagus (Fig. 3B). An esophageal opening was created at the mid-portion of the esophagus using the hook in cutting mode (Fig. 3C), which allowed for easy access to the esophageal lumen with nasogastric tube guidance (Fig. 3D). The opening of the jejunum was created in a similar manner using the hook in cutting mode.

First, the stapler side of a linear stapler was inserted into the jejunal opening with a motion similar to that of pulling up socks; the stapler side could then easily enter the jejunum without separate human assistance. The jaw of the linear stapler was partially closed to prevent the jejunum slipping from the stapler (Fig. 4A). The anvil side of the linear stapler was finally inserted into the opening in the esophagus while the traction suture was controlled with the grasper. By maintaining friction between the two serosal surfaces, it is possible to pull either the esophagus or jejunum using the grasper to adjust the length as needed. In general, a first stapling with a 3-cm length is adequate for preventing stenosis (Fig. 4B).

After firing the stapler, we checked for bleeding along the stapling line and lumen. The common entry opening was closed with a 23-cm barbed suture (3-0 V-Loc suture on a V-20 needle; Covidien). Before starting the V-Loc suture, one stay suture was placed on the corner of the near side of the staple site; this stay suture was used to visualize the blind corner while the common entry opening was sutured (Fig. 4C). Using 23-cm 3-0 V-Loc sutures for both layers, a full-thickness inner layer closure was made proceeding from the corner of the far side of the staple site to the near side using a continuous technique (Fig. 4D). Once the full-thickness layer was completed, the second seromuscular layer was closed by returning to the far side corner using the same barbed suture (Fig. 4E). After the last stitch was made, the suture was simply cut without any knots or additional sutures (Fig. 4F). At this point, a flexible scope with a laparoscopic scope holder provided a stable field of view for carrying out intracorporeal suturing. This is normally a difficult task because even a scopist'ws minute tremors are maximized due to the leverage effect along with trembling caused by the movements of the heart and diaphragm. In this case, the suturing of the common entry opening could be accomplished without much difficulty in the stable field of view provided by the laparoscopic scope holder. The intracorporeal esophagojejunostomy was then complete.