Comparative Assessment of Laparoscopic and Endoscopic Cooperative Surgery With Laparoscopic Wedge Resection for the Treatment of Gastric Submucosal Tumors

We retrospectively assessed 137 patients who underwent curative surgical resection for gSMTs at Tokyo Medical University Hospital, Japan, between January 2004 and January 2023. Of these 137 patients, patients who underwent the following surgical methods were excluded: open wedge resection (n = 53), hand-assisted laparoscopic wedge resection (n = 1), conversion laparoscopy to open wedge resection (n = 1), proximal gastrectomy (n = 1), simultaneous colorectal resection (n = 1), and inverted LECS¹⁹  (n = 1).

Figure 1 shows the flow diagram of the 79 consecutive patients registered in this study. Patients who underwent curative gSMT resection by LWR or LECS were compared. The areas of comparison were patient characteristics, surgical outcomes, postoperative courses, histopathological examination results, surgical margin, and resected specimen area. The areas were obtained from medical chart reviews, histological slides, and pathological reports. Herein, a 2:1 propensity score matching (PSM) was performed to reduce potential selection bias resulting from the number difference of each group, and to increase the observational evidence level.

Figure 2 shows the surgical algorithm for gSMTs in our hospital. LWR was not performed for tumors located on the lesser curvature, near the EGJ, or close to the pylorus to prevent postoperative stricture resulting from the relatively wide extent of resection. The indications for LWR were tumors showing an extraluminal growth pattern, whereas the indications for LECS were not determined by tumor growth pattern. Tumors located on the lesser curvature were treated by LECS to prevent excessive gastric deformation.

As shown in Fig. 3, the tumor and resected specimen areas were determined by approximating an ellipse using the major and minor axes of the tumor and specimen. The tumor and resected specimen areas were calculated using the formula abπ, where “a” indicates the major axis length and “b” represents the minor axis length of the resected specimen.

Written informed consent for the treatments was provided by all patients. The research protocol was approved by the Ethics Committee for Clinical Research of Tokyo Medical University, Japan (Approval No.: T2022–0122).

During the LWR, the patient was under general anesthesia and in a supine position. A split leg position was used to allow the scope assistant to stand between the patient’s legs. A camera port was inserted into the umbilical lesion using an open method. This port was used for inflation with a pneumoperitoneum pressure of 10 mmHg using carbon dioxide. Three or 4 additional ports were added in the upper abdomen according to tumor location. An extra port was added to lift the left liver lobe using a liver retractor if needed. After confirming the tumor location, the gastric marginal vessels around the tumor were dissected using an ultrasonic dissection device to control bleeding and for better tissue mobility. Local resection was performed to extract the tumor with a safe surgical margin using a laparoscopic linear stapling device inserted through a 12-mm port (Fig. 4a). A video of the LWR surgical technique is shown in Supplemental Digital Content 1a.

The surgical setting and anesthesia for LECS were similar to those for LWR. After vessel preparation of the excision areas, tumor peripheries were marked endoscopically. Specimens were removed by endoscopic submucosal incision and subsequent laparoscopic seromuscular dissection around the tumor periphery markings. After the complete tumor resection, the gastric wall defect was closed by laparoscopic hand suturing in all cases, with the minimum gastric transformation. Finally, LECS was completed after confirming the absence of bleeding and leakage (Fig. 4b). A video of the LECS surgical technique is shown in Supplemental Digital Content 1b.

Comparisons of the baseline characteristics and surgery-associated outcomes of the patients who underwent LWR or LECS were made by Fisher’s exact test or Pearson’s chi-square test for categorical variables, and by Mann-Whitney U test for continuous variables. PSM was applied to minimize the effects of confounding factors. In the logistic regression model, the covariates included were age, sex, body mass index (BMI), American Society of Anesthesiologists physical status (ASA-PS), tumor size, and tumor location. The surgical outcomes were compared using a 2:1 nearest neighbor PSM algorithm without replacement in a 2 (LWR) to 1 (LECS) ratio with a caliper value of 0.2 of the standard deviation. P values of < 0.05 were considered to indicate a statistically significant difference. Statistical analysis was conducted using R software for Windows (version 3.3.3; http://www.r-project.org/).

A comparative summary of the background and clinicopathological features of the LWR and LECS patients is shown in Table 1. Among the 79 patients, 61 underwent LWR and 18 underwent LECS. There were no significant differences in gender, age, BMI, ASA-PS, tumor location, and diagnosis between the LWR and LECS groups. GIST was the most frequent pathological diagnosis in both groups. Because of the difference in the number of patients between the 2 groups, a 2:1 PSM was conducted to reduce the influence of patient selection bias. Each group was well matched in terms of gender, age at surgery, BMI, ASA-PS, and tumor location.

Table 2 shows the clinicopathological characteristics of the patients after PSM. After PSM, 24 LWR patients and 12 LECS patients were matched. There were no differences in gender, age, BMI, ASA-PS, tumor location, and diagnosis between the LWR and LECS groups.

As shown in Table 3, surgical outcomes were also stratified to the 2 groups. There was no significant difference in the intraoperative blood loss, postoperative hospital stay, or resected tumor size. The mean intraoperative duration was significantly shorter in the LWR group than in the LECS group (LWR, 131.1 minutes; LECS, 195.2 minutes; P < 0.05). There were no Clavien-Dindo grade II or above complications in the 2 groups. There was no record of mortality during hospitalization, no patient reoperation within 30 days, and no rehospitalization in either group. Histopathological examinations revealed that every patient underwent curative resection with negative margins, irrespective of the surgical procedure.

In Fig. 5a, quantitative analysis of the resected specimen showed that the mean ratio of tumor diameter to resected specimen diameter (tumor diameter:specimen diameter) was significantly higher in the LECS group than in the LWR group (LWR, 60.3% ± 20.9; LECS, 78.9% ± 15.7; P < 0.05).

Similar to the tumor diameter:specimen diameter ratio, Fig. 5b shows that the mean ratio of tumor area to resected specimen area (tumor area:specimen area) was significantly higher in the LECS group than in the LWR group (LWR, 44.7% ± 24.8; LECS, 66.3% ± 22.9; P < 0.05).

Most gSMTs, including GISTs, show expansive growth in the gastric wall. They rarely metastasize to regional lymph nodes or other organs. Thus, treatment of gSMTs requires neither lymphadenectomy nor wide surgical resection margins.^3–5,20,21  The basic treatment choice for gSMTs (including nonmetastatic GISTs) is complete tumor excision²² ; however, there is still no consensus regarding the resection methods.

For tumors <5 cm in diameter, LWR has been performed as one of the most appropriate laparoscopic surgical techniques.³  Other than LWR, several retrospective studies have reported the feasibility and safety of laparoscopic resection for gSMTs.^13–15  Among these surgical methods, since the first report of Hiki et al in 2008,¹¹  LECS as a treatment for gSMTs has gradually increased in usage over the years. LECS has been developed to minimize the invasiveness of surgical interventions for gSMTs. Compared with LWR, LECS makes it possible to resect tumors located in the EGJ, lesser curvature, and even near the pylorus, locations that are difficult to be treated by LWR because of the wide extent of resection by a straight linear stapling device, potentially leading to postoperative gastric deformation or structuring. In the present study, after LECS introduction for gSMT treatment in 2018, gastrectomy was required in only 1 case, which was a circumferential SMT of the EGJ.

Consistent with previous reports,^13,23,24  we found that the operative duration was significantly shorter in the LWR group than in the LECS group. This was because of the simplicity of the surgical procedures in the LWR group. LWR only requires the use of a laparoscopic linear stapling device without circumferential incisions under laparoscopy or endoscopy. Although the use of linear stapling devices shortens the operative duration, their application has restrictions in specific tumor locations (i.e., lesser curvature, EGJ, or pylorus). This may lead to postoperative deformation or structuring and has disadvantages in creating an optimal minimum required surgical margin.

Interestingly, our results showed that the LECS group had a predominantly larger proportion of tumor specimens, and that we were able to show quantitative differences in the surgical margins between the LECS and LWR groups. Even though every patient underwent curative resection with negative margins, the mean ratio of tumor diameter to resected specimen diameter (i.e., tumor diameter:specimen diameter) and the mean ratio of tumor area to resected specimen area (i.e., tumor area:specimen area) were both significantly higher in the LECS group than in the LWR group. These findings suggest the high efficiency of LECS.

Several reasons may underlie these results. First, as LWR requires the use of a straight linear stapling device (multiple in cases), expansion of the normal stomach resection area around the tumor is more often experienced. Excessive gastric resection will lead to gastric deformation causing postoperative complications in the worst case. Second, specimens were extirpated after endoscopic peripheral dissection around the tumor, which avoided the unnecessary resection of the normal stomach wall in the LECS group. Closure of the gastric wall defect was performed using the laparoscopic hand-suturing technique in LECS. Hand-suturing closure and minimum incision around the tumor theoretically show a synergetic effect in preventing deformation of the remnant stomach.

Moreover, Shoji et al²  reported the economic benefits of the use of the laparoscopic hand-suturing technique over the use of a linear stapling device. Regarding surgical indications, our surgical algorithm for gSMTs was found to be suitable, showing no complications in any case. As the algorithm shows, LWR is a reasonable and time-saving surgical method for cases that are unlikely to show postoperative complications, severe deformation of the remnant stomach, or stricture.

This study has several limitations. First, this was a retrospective single-center analysis. Further accumulation of clinical cases and prospective randomized trials are needed to fully establish the most appropriate treatment options for gSMTs. Second, other treatment options for gSMTs such as CLEAN-NET²⁵  and NEWS²⁶  were not included in this study as there was no case requiring such treatment options, and this may have led to bias in the results. Third, postoperative upper gastrointestinal radiographic examinations using contrast media to show objective deformation of the remnant stomach were not performed in any case.

Operative procedures for gSMTs must be carefully selected according to the tumor location and characteristics. LECS is a less invasive surgical technique that avoids postsurgical complications. Moreover, LECS is potentially beneficial in minimizing ontologically acceptable surgical margins, which account for the functional and anatomic preservation of gSMTs.