Comparison of Totally Laparoscopic Total Gastrectomy and Open Total Gastrectomy for Gastric Cancer

Introduction

Gastric cancer is the second most common cause of cancer-related deaths in the world and the leading primary malignancy in Korea.^1,2 Early detection of gastric cancer has improved recently because of the popularity of cancer screening and diagnostic advances. Since laparoscopic gastrectomy was introduced by Kitano et al. ³ in 1994, it has been possible to treat both early and advanced gastric cancer successfully by surgical resection, which involves laparoscope-assisted gastrectomy with lymph node dissection.^4–7 Recently, several investigators have reported that totally laparoscopic gastrectomy using intracorporeal reconstruction is feasible and safe,^8–10 and totally laparoscopic distal gastrectomy is now being frequently performed for early gastric cancer because of the absence of wound retraction, the wide operating view, and tension-free anastomosis.^11–13 Most recently, it has been reported that laparoscopic total gastrectomy using intracorporeal reconstruction is feasible and safe.^14–18 Although there have been several reports of early surgical outcomes of various procedures for totally laparoscopic total gastrectomy (TLTG), an optimal method has not been established because of the technical difficulties involved and concern about postoperative complications, especially the reconstruction of esophagojejunostomies (EJs). Therefore, we have developed a secure and practicable technique for intracorporeal EJ and have performed TLTG successfully using endoscopic linear staplers. In this study we describe our experience with TLTG in the treatment of gastric cancer and compare the effectiveness of TLTG with conventional open total gastrectomy (OTG).

Patients and Methods

Surgical technique of TLTG

Posture and site of insertion of a trocar

The patient was placed in the reverse Trendelenburg position. A carbon dioxide pneumoperitoneum was formed from the umbilical port, with pressure maintained between 12 and 15 mm Hg. Total gastrectomy was performed using four trocar ports placed in a U-shaped pattern (Fig. 1A). To retract the liver, the attachment site of the lesser omentum to the right diaphragmatic cruse was intracorporeally sutured, and then a thread pulled by a suture-passer was tied onto the skin in the xiphoid process area (Fig. 1B). If the operating field was not adequate, an additional 5-mm trocar was inserted into the epigastric area to retract the liver.

OPEN IN VIEWER

FIG. 1.

(A) The first trocar is inserted at the umbilicus. The 5-mm and 12-mm trocars are then placed at the right and left upper quadrant sites. An endoscopic linear stapler for esophagojejunostomy is used through the trocar at the umbilicus. (B) Liver retraction during laparoscopic gastrectomy.

Dissection and removal of specimen

Dissection was begun by dividing the greater omentum from the midportion of the gastroepiploic arcade to the left gastroepiploic vessel. The lymph nodes around the left gastroepiploic and short gastric vessels were dissected, followed by the infrapyloric area. After dissection of the lymph nodes around the suprapyloric area, the duodenum was transected just below the duodenal bulb with an endoscopic linear stapler (Echelon™ Flex™ 60; Ethicon, Blue Ash, OH) with a closed stapler height of 1.5 mm. Subsequently, the lymph nodes around the common hepatic, proximal and distal splenic, celiac, and left gastric arteries were dissected (in that order), followed by dissection of the lymph nodes around the right paracardial and lesser curvature areas.

After all the lymph nodes were cleared, nearly two-thirds of the esophagus diameter was transected 2 cm above the gastroesophageal junction using the same type of endoscopic linear stapler (Fig. 2A), and the first intracorporeal suture was placed at the end of the stapled line to retract the esophageal stump (Fig. 2B). This suture was cut 15 cm from the esophageal stump, which was retracted by the first assistant during reconstruction of the EJ. The unstapled esophageal stump was then transected with laparoscopic scissors after the remnant stomach was grasped with a laparoscopic intestinal clamp to avoid cancer cell spillage (Fig. 2C). To make the lumen of the esophagus easier to detect, a second round of intracorporeal sutures was placed in the small esophagostomy of the esophageal stump, and it was pulled out of the abdomen through the right lower trocar; this prevented the esophageal mucosa and submucosa from slipping during reconstruction of the EJ (Fig. 2D). The specimen was subsequently removed through another 3–4-cm incision on the lower abdominal site, 2 cm from the suprapubic area (Fig. 3). After the specimen was removed, this incision site was closed by suturing the peritoneum to reinstate the pneumoperitoneum. The proximal resection margin of the specimen was examined pathologically.

OPEN IN VIEWER

FIG. 2.

Making an esophageal stump. (A) Nearly two-thirds of the esophagus diameter is transected 2 cm above the gastroesophageal junction using an endoscopic linear stapler. (B) The first intracorporeal suture is made at the end of the staple line of the esophageal stump. (C) The unstapled esophagus is transected with laparoscopic scissors after the remnant stomach is clipped with manual titanium clips to avoid spillage of cancer cells. (D) The second intracorporeal suture is made at the esophagostomy site of the esophageal stump.

OPEN IN VIEWER

FIG. 3.

(A and B) The specimen is removed through another 3–4-cm incision on the lower abdominal site, 2 cm from the suprapubic area.

The jejunum was then divided 20 cm below the ligament of Treitz using an endoscopic linear stapler (Echelon Flex 60), and an efferent loop was turned in a counterclockwise direction to reconstruct the EJ. An enterostomy of the jejunum was made on the antimesenteric side of the Roux-en-Y limb using laparoscopic scissors, and an endoscopic linear stapler (ETS Straight™ 45; Ethicon) with a closed staple height of 1.5 mm was inserted into the esophagostomy and enterostomy of the jejunum to form an EJ (Fig. 4B).

OPEN IN VIEWER

FIG. 4.

(A) The first assistant retracts the first thread in the operator's direction inside the abdominal cavity, and the second assistant retracts the second thread through the right lower trocar from outside the abdomen. (B) To make an esophagojejunostomy, an endoscopic linear stapler (ETS Straight 45) is inserted by the operator between the esophagostomy and enterostomy of the jejunum. (C) The remnant entry hole is closed with an endoscopic linear stapler (Echelon Flex 60) by the operator. (D) The esophagojejunal anastomosis is seen after completion of the reconstruction.

Reconstruction of anastomosis

To make an EJ, the two sutures were retracted during the reconstruction. The first assistant retracted the first suture in the direction of the operator inside the abdominal cavity (Fig. 4A), and the second assistant retracted the second suture out of the abdomen through the right lower trocar. This procedure is important for preventing the esophageal stump from slipping into the thoracic cavity; it also guides the introduction of the anvil of the endoscopic linear stapler into the small esophagostomy of the esophageal stump during reconstruction of the EJ. An endoscopic linear stapler (ETS Straight 45) was inserted by the operator between the esophagostomy and enterostomy of the jejunum to form an EJ (Fig. 4B). After that, the entry hole was closed with three sutures for tissue approximation, followed by stapling with a laparoscopic linear stapler (Echelon Flex 60) (Fig. 4C). In some cases, further suturing was necessary to complete the closure. Then, to construct the jejunojejunostomy, two small enterostomies of the jejunum were made: (1) on the antimesenteric side of the jejunum, 40 cm below the EJ, and (2) on the antimesenteric side of the afferent loop. The endoscopic linear stapler (Echelon Flex 60) was inserted into these enterostomies of the jejunum to construct a jejunojejunostomy, and the entry hole was closed with the endoscopic linear stapler (Echelon Flex 60). Finally, the mesenteric and the Peterson defects were sutured, and the operation was finished.

Results

The clinical backgrounds of the 346 patients are shown in Table 1. There was no significant difference in age, gender, body mass index, American Society of Anesthesiologists score, and previous abdominal operation history between the OTG and TLTG group.

All the TLTG operations were successfully performed using intracorporeal esophagojejunal anastomosis with an endoscopic linear stapler. None of the TLTG group required conversion to open surgery or other laparoscopic anastomosis techniques. The extent of lymph node dissection was similar in the two groups. All the operations were curative.

Table 2 summarizes the early surgical outcomes. We found no differences between the OTG and TLTG groups in operation time or postoperative change in hematocrit. However, times to first flatus and to commencement of a soft diet and the length of postoperative hospital stay were significantly shorter in the TLTG group; in addition, the number of analgesic doses administered was significantly lower, and pain scored on the VAS lasted less time. Table 3 gives details of the postoperative complications in the two groups. There were significant differences in terms of overall and mild postoperative complications. Although the severe postoperative complication rate was similar in the two groups, there was no instance of anastomosis leakage in the TLTG group (Table 3).

Postoperative complications developed in 21.7% (45 cases) of the patients who received OTG. The postoperative complication rate for the TLTG group was significantly lower at 10.0% (14 cases). The mild postoperative complications (7 cases, 5.0%) in the TLTG group were partial obstruction of the anastomosis site (n=2), pneumonia (n=2), wound infection (n=1), paralytic ileus (n=1), and ischemia of the left lateral segment of the liver (n=1). Severe postoperative complications (7 cases, 5.0%) in the TLTG group were extraluminal bleeding (n=2), intraluminal bleeding (n=1), duodenal stump leakage (n=1), mechanical ileus (n=2), and transmesenteric herniation (n=1).

One of the instances of extraluminal bleeding in the TLTG group was stopped by laparoscopic suturing around the suprapancreatic area; the other patient had a large hematoma and was managed by inserting a pig-tail catheter with antibiotics and transfusion. The patient with duodenal stump leakage was managed by a combination of laparoscopic tube duodenostomy and conservative therapy, whereas the two cases with mechanical ileus were managed by laparoscopic adhesiolysis. Finally, the patient with internal herniation was managed by laparoscopic hernia reduction. There were no operative deaths in either group during the first 30 postoperative days.

Table 4 presents the pathology findings for the two groups. The groups did not differ significantly in terms of distal resection margin (P=.577), distribution of tumor depths (P=.059), or presence of lymph node metastasis (P=.056). However, there were significant differences in tumor size (P=.011), number of retrieved lymph nodes (P=.039), and proximal resection margin (P=.005).

Fluorography revealed smooth passage of contrast medium through the anastomoses in the TLTG group without leakage or stenosis (Fig. 5). Postoperative endoscopy demonstrated larger anastomoses in the TLTG group than in the OTG group 6 months after operation (Fig. 6).

OPEN IN VIEWER

FIG. 5.

Fluorography reveals smooth passage of contrast medium through the anastomosis without leakage or stenosis.

OPEN IN VIEWER

FIG. 6.

Postoperative endoscopy. (A) The lumen of the anastomosis 6 months after open total gastrectomy using a circular stapler. (B) The large lumen of the anastomosis 6 months after totally laparoscopic total gastrectomy using an endoscopic linear stapler.

Discussion

Since the first report of laparoscopy-assisted distal gastrectomy in 1994, ³ the procedure has been widely adopted for gastric cancer.^4,21,22 Recently, many workers have described totally laparoscopic distal gastrectomy using intracorporeal reconstruction. More recently, it has been shown to have several advantages over laparoscopy-assisted distal gastrectomy, including smaller wounds, less invasiveness, and greater security.^9,10,23,24 The laparoscopic distal gastrectomy technique is fully established, and many patients have undergone laparoscopic distal gastrectomy for gastric cancer.^9–13,23 In contrast, the number of laparoscopic total gastrectomies has been limited by the small number of expert surgeons who perform the procedure because of its technical challenges and lack of knowledge of long-term outcomes.^14–18,25 Reconstruction of the EJ is technically one of the most difficulty aspects of the TLTG method. However, several investigators have reported on its safety and feasibility.^14–18 Previously, we performed extracorporeal EJ and reported that laparoscopy-assisted total gastrectomy was superior to conventional OTG as it was associated with earlier bowel movements, less pain, and shorter hospital stays. ²⁶ However, it is sometimes difficult to complete anastomoses by minilaparotomy in narrow and deep operating fields. In our experience, one difficulty is the process of clamping the distal esophagus through a small incision. Another is inserting a straight needle into the purse-string clamp and extracting it. A final difficulty is anvil insertion into the esophagus. This procedure tends to be affected by the size of the patient, and in obese patients we had to enlarge the incision. ²⁶ Therefore, beginning in October 2008 we started performing intracorporeal EJ after laparoscopic total gastrectomy using an endoscopic linear stapler to exploit the advantage of laparoscopy-assisted total gastrectomy. We believe that we had reached a plateau in the learning curve with our 71st patient by January 2011, in terms of early surgical outcomes such as operation time, bowel movement, and postoperative complications. We had performed over 1500 laparoscopic gastrectomies before this study commenced and believe we are justified in claiming that we were well up the learning curve for performing TLTG. The purpose of this study was to establish a TLTG procedure using an endoscopic linear stapler and to compare it with the conventional procedure in gastric cancer.

Reconstruction of the intracorporeal EJ after laparoscopic gastrectomy is technically the most difficult aspect of TLTG. In our intracorporeal EJ method, we create two intracorporeal sutures to prevent slipping of the esophageal stump during reconstruction of the EJ: one at the end of the staple line and one at the small esophagostomy of the esophageal stump. The first assistant pulls the first suture toward the operator in the abdominal cavity, while the second assistant pulls the second suture thread toward the operator's side from outside the abdominal cavity. This retraction prevents the anastomosis from falling into the thoracic cavity while simultaneously making it safe. Furthermore, the operator is able to insert an endoscopic linear stapler without great difficulty between the esophagostomy and enterostomy of the jejunum to make an EJ. As a result, we are able to minimize the size of the remnant entry hole of the EJ and successfully complete the EJ (Fig. 4D).

Our study showed that TLTG using an endoscopic linear stapler could be performed safely with the intended benefit of less invasive surgery and favorable early surgical outcomes as assessed by more rapid postoperative recovery. In our series, the time required for TLTG did not differ from that for conventional OTG (137 versus 144 minutes, P=.381). When performed by a skilled and experienced surgeon, laparoscopic gastrectomy does not require any more time than open gastrectomy. ²⁷

In our study, TLTG was indeed safe. The postoperative complication rate was lower than in the OTG group. Although there was no significant difference in severe postoperative complications between the two groups, there was no leakage of anastomosis in the TLTG group. However, in TLTG, one needs to create a wide operating view in order to construct the EJ safely. In our approach to TLTG, we are able to widen the operation field by retracting the liver. In addition, in our method we reconstruct the EJ with endoscopic staplers and three staggered rows of staples, whereas in OTG it is created by only two rows of staples; the three staggered rows of staples provide better staple line security than the two rows of staples. In addition, in terms of the lumen size for the esophagojejunal anastomosis, the lumen is larger in the TLTG (Fig. 6B) than with the double stapling method using a circular stapler (Fig. 6A). Among our patients there were 2 cases of mild anastomotic stricture of the EJ, and these cases were managed by conservative therapy. Clearly the TLTG technique provides acceptable surgical outcomes not only in terms of mild postoperative complications but also anastomosis-related complications.

At present, surgical resection is the treatment of choice for gastric cancer, but the extents of resection and lymphadenectomy during gastrectomy remain controversial. However, a recent randomized trial confirmed the survival benefit of D2 lymph node dissection while controlling for the morbidity of lymph node dissection.^28–30 Therefore, we intend to continue performing D2 lymphadenectomy for gastric cancer. In this study, although the number of lymph nodes harvested was significantly lower in the TLTG group, the average number of lymph nodes harvested was over 30 in both groups; hence there was no clinical difference in this respect between the OTG and TLTG groups.

There are very few definite guidelines that cite literature sources and make recommendations regarding the desirable distance between tumor and resection margin. In a search of Medline we found only a few articles that made such recommendations.^31–33 Bozzetti et al. ³² evaluated 343 patients with gastric cancer, including advanced cancer, with special reference to the outlines of the tumor, assessed grossly, together with other macroscopic or microscopic features of the tumor. They reported that neither proximal nor distal infiltration occurred at a distance >3 cm in lesions confined to the mucosa, submucosa, and muscularis. Jang et al. ³³ reported that middle-third advanced gastric cancers that were treated by curative resection were not affected by the length of the proximal resection margin and the extent of gastric resection. In our procedure, the gross margin was free, and frozen-sections of the proximal resection line were examined to ensure that R0 resections had been performed.

In the present study, early surgical outcomes after TLTG were good, with reduced pain, rapid recovery of bowel movement, shorter hospital stay, fewer mild postoperative complications, and fewer anastomosis-related complications. We acknowledge that selection bias affecting the study populations in this retrospective study may have influenced the comparison of early surgical outcomes of OTG and TLTG. However, in this study, although TLTG aimed to cure via more radical lymphadenectomy, the oncologic safety of TLTG has not yet been established. Therefore, a large, randomized, and prospective study is needed to evaluate the effectiveness of TLTG for gastric cancer.

In conclusion, our TLTG method is safe and reliable. We believe that it should be preferred to OTG as a surgical treatment for gastric cancer. However, long-term follow-up is needed to evaluate any oncologic effects.