Short-term outcomes of laparoscopic local resection for gastric submucosal tumors: a single-center experience of 266 patients

Gastric SMTs are potentially life-threatening tumors. Although the majority of SMTs are benign, some have the potential to become malignant. GISTs make up a large portion of SMTs. It has been reported that the pathogenesis of GIST is closely related to mutations at c-Kit proto-oncogene, and GISTs that do not have c-Kit mutations might be correlated with gain-of-function mutations of platelet-derived growth factor-a (PDGFRa) [12]. Secondary to these findings, imatinib mesylate (Gleevec®) therapy, which can inhibit the intracellular kinase activities of CD117 and PDGFRa, has been manifested to increase the overall tumor control rate of GIST by 85%. It has been report that Gleevec® is the standard treatment for unresectable or metastatic GIST in many countries [13]. However, resection remains the first option for primary GIST. On the one hand, it can establish the diagnosis; on the other hand, it may be curative. SMTs that are not obviously benign should be excised as assumed GISTs. Patients with a diagnosis for GISTs are mainly treated by laparoscopic local resection [14].

Lukaszczyk and colleagues reported in 1992 the use of laparoscopy for gastric GIST resection in a patient [15]. Since then, there have been multiple small series using the laparoscopic approach for these kinds of tumors. In this study, LLR was successfully performed within an acceptable operation time. It is lowest of the average blood loss and morbidity rate, and mortality hardly happened. All patients started oral feeding earlier, the average oral intake day is 3.2 days (range: 2 to 6 days) after the operation. Hospital stays were also short and acceptable. Pathologic examination of the surgical specimens showed that all surgical margins were microscopically tumor free (R0 resection). Although there are no randomized studies comparing laparoscopic versus open approach in the management of gastric SMTs, several retrospective studies have shown the advantages of the laparoscopic approach in treating gastric SMTs, with similar disease free survival, mortality and oncologic outcomes comparable with the open approach [16‐24]. It is not only the technical potential of laparoscopic excision, but also its effectiveness facilitated by the data.

LLR can be distributed into ELWR or LTR according to the development mode, tumor measurement, and site. ELWR is the most ordinary method and the first thought for gastric SMTs. We can use the laparoscopic vision or the haptic retroaction of the laparoscopic tools to locate lesions straightly for the tumors which were located in the anterior wall along the greater curvature of the stomach. Considering the remarkable superfluity and mobility of the stomach in these sites, a laparoscopically stapled gastric wedge excision often is always greatly potential. With a project for a stapled wedge excision, it is easy to come near laparoscopically for tumors in this site. In order to avoid cardiac stricture for tumors located in the fundus, the left gastroepiploic artery should be transected. The gastrophrenic ligament and gastrosplenic ligament should be dissected for mobility of the stomach and then making sure there is enough distance between the cutting line and the left side of the cardia when applying the second staple. Otherwise it would cause stricture due to the Endo-GIA position being too close to the cardia. Because the stomach where tumors along the lesser curvature are short of redundancy, and the lesser curvature is restricted in length, so it is tough to treat laparoscopically. To make the posterior wall of the stomach clear, we are supposed to do dissection to part the stomach from the greater omentum. Then, the lesser curvature was everted and the tumor was removed. The vagus nerve branch (Latarjet nerves) and blood vessels in the lesser omentum should be protected.

For the reason that ELWR has the chance of stenosis or deformity conducing to excessive excision of the normal gastric wall, it is difficult to apply to tumors which are located close to the gastric inlet or outlet [5]. In our series, there seemed to be 65 cases with tumors which were located close to the esophagogastric junction or the pylorus which were thought improper to undergo ELWR by laparoscopic stapling. We had better utilize ultrasonic coagulating shears with the excision margin paralleling the round edge of the tumor to reduce the fine tissue loss and to hold back luminal narrowing to operate manual excision, on condition that the tumor has exogastric development and is located on the anterior wall. Then, the laparoscopic intracorporeal hand-sewn method was used to close the incision. It is advisable to use the laparoscopic transgastric method for the excision of an intraluminal tumor which is located at the posterior wall of the stomach, which offered straight observation of the lesion and inner stomach, and brings greater command of the surgical margin. [25, 26]. The outcomes of our retrospective research proved that the process was secure and effective. Two cases had postoperative gastric intraluminal bleeding, but this was cured with a conservation treatment. In our experience, the exact site of tumors is crux of this method, which is using endoscopic before the operation, marking with titanium clips, intact redeployment of the EGJ or pylorus before excision, and step cutting transversely along the foundation of the tumor. However, restrictions related to attached bleeding venture and intraperitioneal contanmination by seeping gatric juices are common.[26, 27]. The endoscopic linear staple resection line of the stomach wall is weak and leakage and bleeding occur easily. Therefore, intracorporeal hand-sewn was used in our center to reinforce this area. Intraperitioneal contamination with gastric juice seemed to be a dominating problem of this technique, as the gastric cavity requires opening temporarily. Hence, it is necessary to operate abundant decompression of the stomach before gastrotomy and comprehensive irrigation of the operating area after closing the gastrotomy for preventing abdominal or wound infection.

There is another technique designed to resect tumors located near the EGJ or pylorus is laparoscopic intragastric resection [26]. This technique involves a difficult procedure to set up the view before resection. By blowing up a balloon sticked on the trocars, the gastric wall is supposed to be attached to the abdominal wall after inserting several trocars into the gastric lumen through the gastric wall. This method not only offers an abundant operative area, but also generates less deformity of the EGJ compared with an extragastric approach. However, the need of specific ballon-type ports and the trouble originated from inserting the ports into stomach restricted its feasibility. In addition, if a tumor is larger than 4 cm, the intragastric resection is inapposite, because it is difficult to retract the large specimen orally. We did not use this method in our center.

If a relatively large tumor is located at EGJ and antrum, the surgeons must consider the problems incurred by LLR that include the possibility of stenosis and deforming the gastric inlet or outlet. For tumors located at antrum or large tumors in the lower stomach close to antrum, we recommended laparoscopic distal gastrectomy (LDG). For tumors at EGJ or large tumors in the middle or upper body we recommend laparoscopic total gastrectomy (LTG) instead of laparoscopy proximal gastrectomy (LPG) due to the relatively lower rate of reflux esophagitis. Another tip, observation by flexible endoscope during the resection of the GIST is recommended to avoid gastric inlet or outlet narrowing. During endoscope examination, we looped the pylorus with a silk band to stop gas from entering the intestine, which would interfere with vision and subsequent manipulation.

Several limitations of this study warrant mention and require special attentions in the interpretation. First, because it was a retrospective study performed at a single institution, case selection was inevitably affected by bias. Second, the uneven surgical skills of the different surgeons might result in flaws of the study. Third, it’s a one-arm study and long-term outcomes were not evaluated because of the short observation period. Therefore, randomized controlled trials or prospective comparative studies with long-term follow-up are necessary to adequately evaluate the status of LLR for gastric SMTs.