Abstract
Objective
Although endoscopic resection (ER) is an established technique for gastric subepithelial tumours (SETs), comprehensive data on its efficacy and safety remain limited. This study aimed to evaluate the efficacy and safety of ER in patients with gastric SETs and to identify risk factors associated with procedure-related complications.
Materials and methods
This retrospective study included 483 patients who underwent ER for gastric SETs between February 2012 and May 2023. Patient demographics, tumour characteristics, and clinical outcomes were evaluated. Multivariate analysis was performed to identify risk factors for complications.
Results
The median tumour size was 1.2 cm (range: 0.5–6.0 cm). The complete resection rate was 93.8%. Complications occurred in 6.8% of cases, including perioperative bleeding (2.1%), perforation (1.7%), and both (0.4%), with 0.8% of cases requiring conversion to surgery. Larger tumours notably increased the risks of incomplete resection (odds ratio [OR] = 1.605, 95% confidence interval [CI]: 1.066–2.416, p = 0.023), perioperative bleeding (OR = 2.004, 95% CI: 1.099–3.653, p = 0.023), and perforation (OR = 3.476, 95% CI: 1.830–6.602, p < 0.001). Additionally, an irregular tumour shape significantly elevated the risk of incomplete resection (OR = 10.771, 95% CI: 3.452–33.605, p < 0.001). Conversely, tumours located in the middle third (OR = 0.068, 95% CI: 0.014–0.341, p = 0.001) and the upper third (OR = 0.211, 95% CI: 0.051–0.884, p = 0.033) of the stomach substantially reduced the risk of incomplete resection.
Conclusions
ER is an effective and generally safe treatment modality for managing gastric SETs. However, larger tumour size and irregular shape are significant risk factors for adverse outcomes.
Keywords: Endoscopic resection, subepithelial tumours, efficacy, safety, adverse outcomes
KEY MESSAGES
This study involving 483 patients with gastric SETs demonstrated a high complete endoscopic resection rate of 93.8%.
The overall safety profile of ER for gastric SETs is favourable.
Larger tumour size and irregular shape are major predictors for adverse outcomes during ER procedures.
Introduction
Gastric subepithelial tumours (SETs) represent a diverse spectrum of neoplasms that originate from various layers of the gastric wall, spanning from the deep mucosa to the serosa, and each characterized by its unique cellular lineage. The histological subtypes of these tumours include gastrointestinal stromal tumours (GISTs), leiomyomas, and several less common entities such as schwannomas, neuroendocrine tumours, and lipomas. Gastric SETs have witnessed a significant increase in prevalence, largely due to the widespread utilization of esophagogastroduodenoscopy and endoscopic ultrasonography (EUS). The detection rates during routine endoscopic examinations range from 0.3% to 0.76% [1,2].
Current guidelines recommend endoscopic surveillance for asymptomatic gastric SETs and GISTs under 2 cm in size that exhibit no high-risk characteristics on EUS [3–5]. However, repeated surveillance may impose a psychological burden on patients and could potentially lead to delayed diagnosis and treatment of malignant lesions. Consequently, there is growing interest in a proactive approach to managing gastric SETs, advocating for their immediate resection upon detection to facilitate definitive histological diagnosis and mitigate the risk of malignant progression [6,7].
Traditionally, surgical resection has been the standard approach for removing gastric SETs. However, the invasiveness of surgical interventions, along with associated morbidity and mortality, has prompted the exploration of less invasive options. Endoscopic resection (ER) has emerged as a promising alternative, enabled by advancements in endoscopic techniques and equipment [1,8,9]. ER presents several advantages over surgery, including its minimally invasive nature, preservation of gastric anatomy and function, shorter hospital stays, and reduced healthcare costs [10–12]. Despite these benefits, there is still limited literature on the real-world efficacy and safety profile of ER for gastric SETs, and previous studies have insufficiently evaluated the risks of incomplete resection and procedure-related complications [2,13,14]. This study aims to address this deficiency by evaluating the efficacy and safety of ER for gastric SETs while identifying risk factors associated with incomplete resection, perioperative bleeding, and perforation.
Methods
Patients
We screened the patients who underwent ER of gastric SETs at the First Affiliated Hospital of Xiamen University between February 2012 and May 2023 for potential inclusion in this study. The inclusion criteria were: (1) absence of potential malignant features such as indistinct margins with signs of infiltration into the adjacent tissues, or heterogeneous echogenicity during endoscopic or EUS examination (UM-2R, 12 MHz; UM-3R, 20 MHz, Olympus Optical, Tokyo, Japan), (2) no signs of lymph node involvement or distant metastasis outside the gastrointestinal tract as determined by computed tomography (CT). Exclusion criteria included: (1) patients with multiple gastric SETs or those with concurrent oesophageal SETs, and (2) individuals with coagulation disorders. This retrospective study was conducted in accordance with the ethical standards of the Declaration of Helsinki and was granted an exemption from ethical review by the ethics committee of the First Affiliated Hospital of Xiamen University. The Institutional Review Boards of the First Affiliated Hospital of Xiamen University allowed to waive the informed consent due to the retrospective nature of this study.
Endoscopic procedures
All endoscopic procedures were performed by experienced endoscopists who had previously conducted over 2500 cases of endoscopic mucosal resection and more than 150 cases of endoscopic submucosal dissection (ESD). The ER techniques included ESD, endoscopic submucosal excavation (ESE), endoscopic full-thickness resection (EFTR), and submucosal tunnelling endoscopic resection (STER). The choice of these treatment modalities depended on factors such as tumour size, growth characteristics, and the depth of invasion. ESD was employed for lesions restricted to the submucosal layer, whereas ESE was typically chosen for tumours arising from the muscularis propria (MP) that exhibited intraluminal growth. EFTR was utilized for more deeply seated MP tumours with outward growth or those attached to the serosa. Meanwhile, STER was used for SETs located near regions such as the cardia, fundus, or greater curvature, where submucosal tunnelling could be effectively performed. The procedural steps for these ER techniques have been extensively described in previous studies [15,16]. In summary, the ESD procedure consisted of: (1) marking the lesion circumferentially with either needle knife or argon plasma coagulation (APC); (2) injecting a saline solution combined with indigo carmine under the submucosa to create a lifting cushion; (3) performing a circumferential incision, followed by submucosal dissection, and (4) addressing any visible small blood vessels on the wound surface using haemostatic forceps or APC, with defect closure using endoclips if required. ESE and EFTR are adaptations of the ESD technique. In ESE, the focus was on tumour excavation following submucosal injection and incision, with subsequent haemostasis and defect closure (Figure 1). EFTR, on the other hand, involved dissection extending to the serosal layer, inducing a controlled perforation, with complete tumour removal including parts of the muscularis propria and serosa, followed by closure using nylon rope sutures or endoclips (Figure 2). The essential steps of the STER procedure comprised: (1) submucosal injection approximately 5 cm proximal to the lesion to facilitate dissection, (2) creating a longitudinal incision (2 cm) to establish an entry point for tunnelling, (3) developing a tunnel between the layers of mucosal and MP, (4) precise dissection and removal of the tumour along its margins using endoknives, (5) coagulation of visible blood vessels within the tunnel to prevent bleeding, and (6) closure of the entry incision using multiple endoclips (Figure 3).
Figure 1.
Endoscopic submucosal excavation procedure. (A) Endoscopic view of a subepithelial tumour in the gastric body. (B) Marking of the lesion. (C) Creation of a fluid cushion through submucosal injection. (D) Incision along the marked mucosal points. (E) Dissection around the tumour alongside the muscularis propria. (F) The artificial ulcer after tumour retrieval. (G) Ulcer closure with several endoclips. (H) The resected specimen.
Figure 2.
Endoscopic full-thickness resection procedure. (A) Endoscopic view of a subepithelial tumour situated in the fundus. (B) Creation of a fluid cushion through submucosal injection. (C) Precut incision on the mucosal layer. (D) Tumour exposure. (E) Full-thickness resection causing intentional perforation. (F) The full-thickness defect after tumour removal. (G) Defect closure using several endoclips. (H) The resected specimen.
Figure 3.
Submucosal tunnelling endoscopic resection procedure. (A) Endoscopic view of a subepithelial tumour located in the cardia. (B) Creation of the tunnel entry point. (C) Establishment of a submucosal tunnel between the mucosal and muscularis propria layers. (D) Dissection along the tumour’s margin within the tunnel. (E) Removal of the tumour from the tunnel using a snare. (F) The defect after tumour removal. (G) Closure of tunnel entry with clips. (H) The resected specimen.
Postoperative management
After the procedure, patients were instructed to fast for a minimum of 48 h, followed by a gradual transition from a liquid diet to regular meals. Postoperative care was standardized, involving the routine administration of proton pump inhibitors (PPIs) for all patients, with prophylactic antibiotics provided for those with MP injuries. For patients who underwent EFTR, gastrointestinal decompression was performed according to established protocols. Continuous monitoring of clinical signs and abdominal symptoms was crucial for the early identification of complications such as infections, delayed haemorrhage, or perforation. When necessary, prompt diagnostic evaluations using abdominal CT or endoscopy were initiated to address any complications.
Clinical data collection
A retrospective review of detailed clinical data was performed using the hospital’s electronic medical record system. The collected data encompassed demographic information (e.g. age and sex), presenting symptoms, tumour characteristics, procedural outcomes (such as complete resection and duration), complications, recurrence rates, and follow-up information. Pathological assessments, comprising sectioning, haematoxylin and eosin staining, as well as immunohistochemical analysis, were independently performed by two pathologists. In instances of disagreement, a third pathologist was consulted to reach a consensus. Tumour size (maximum diameter) was documented, and GISTs were classified based on the modified National Institutes of Health (NIH) risk criteria [3,17].
Definitions
Complete resection was defined as the en bloc removal of a tumour with both horizontal and vertical margins confirmed to be tumour-free. Bleeding was defined as hematemesis and/or melena, or a drop in haemoglobin of more than 2 g/dL. In EFTR procedures, perforation was generally expected and only recorded as a complication if surgical intervention was required. In contrast, all perforations during ESE and STER procedures were classified as complications. Fever was defined as a body temperature exceeding 37.3 °C. The severity of adverse events was assessed using the American Society for Gastrointestinal Endoscopy (ASGE) severity grading system [18]. Tumour shape was categorized according to established standards [10,19]. Recurrence was defined as the reappearance of SETs at or near the site of the primary resection more than three months after ER, whereas detection within three months was considered residual disease [20].
Follow up
Approximately 3–6 months after endoscopic treatment, it was recommended to undergo endoscopy and abdominal CT scans to assess for any residual tumours or recurrence. Thereafter, periodic surveillance was advised biennially. For patients with GISTs, follow-up assessments were suggested as previously outlined [21]. Follow-up duration was defined as the period from completion of ER to last endoscopic or CT evaluation. All patients were monitored until October 31, 2023.
Statistical analysis
Data analysis was performed using SPSS version 22.0 (IBM Corp., Armonk, NY). Continuous data were presented as medians with corresponding ranges, while categorical variables were expressed as frequencies and percentages. Multivariate logistic regression analysis was employed to calculate the odds ratio (OR) and 95% confidence interval (CI) to identify risk factors for incomplete resection, perioperative bleeding, and perforation. A two-tailed P-value of less than 0.05 was considered statistically significant.
Results
Baseline characteristics
Initially, 519 patients with 547 gastric SETs met the inclusion criteria for this study. After excluding 36 patients due to synchronous lesions, the cohort consisted of 483 patients, each with a single SET. The median age was 52.0 years, with a female-to-male ratio of 1.9:1. A total of 174 patients (36.0%) had at least one comorbidity, the most common being hypertension (13.0%), hepatitis (6.0%), and diabetes mellitus (4.3%). Chronic kidney disease (0.2%) and liver cirrhosis (0.4%) were relatively uncommon. Additionally, 3.1% of patients were on antithrombotic therapy. Regarding lifestyle factors, alcohol consumption and smoking were reported in 19 (3.9%) and 57 (11.8%) patients, respectively. Nearly half of the patients were asymptomatic. Among symptomatic individuals, the most frequently reported complaint was abdominal pain (31.5%, 152/483), followed by abdominal discomfort (18.4%, 89/483) and gastrointestinal bleeding (0.8%, 4/483). The clinical characteristics of patients according to different endoscopic procedures are summarized in Table 1.
Table 1.
Demographic characteristics of 483 patients with gastric SETs among different endoscopic procedures.
| Total | ESE | EFTR | STER | ESD | |
|---|---|---|---|---|---|
| Number | 483 | 300 | 116 | 18 | 49 |
| Age, median (range), years | 52 (22–80) | 51 (23–80) | 55 (22–74) | 43.5 (22–73) | 43 (23–77) |
| Sex, n (%) | |||||
| Male | 165 (34.2) | 89 (29.7) | 51 (44.0) | 4 (22.2) | 21 (42.9) |
| Female | 318 (65.8) | 211 (70.3) | 65 (56.0) | 14 (77.8) | 28 (57.1) |
| Concomitance diseases, n (%) | 174 (36.0) | 107 (35.7) | 45 (38.8) | 6 (33.3) | 16 (32.7) |
| Hypertension | 63 (13.0) | 37 (12.3) | 15 (12.9) | 5 (27.8) | 6 (12.2) |
| Hepatitis B | 29 (6.0) | 16 (5.3) | 8 (6.9) | 1 (5.6) | 4 (8.2) |
| Diabetes mellitus | 21 (4.3) | 8 (2.7) | 9 (7.8) | 1 (5.6) | 3 (6.1) |
| Peptic ulcer | 20 (4.1) | 16 (5.3) | 4 (3.4) | 0 (0.0) | 0 (0.0) |
| Chronic kidney disease | 1 (0.2) | 1 (0.3) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Liver cirrhosis | 2 (0.4) | 1 (0.3) | 0 (0.0) | 1 (5.6) | 0 (0.0) |
| Respiratory diseases | 16 (3.3) | 10 (3.3) | 5 (4.3) | 1 (5.6) | 0 (0.0) |
| Heart diseases | 12 (2.5) | 6 (2.0) | 4 (3.4) | 0 (0.0) | 2 (4.1) |
| Others | 43 (8.9) | 28 (9.3) | 9 (7.8) | 0 (0.0) | 6 (12.2) |
| Antithrombotic agents, n (%) | 15 (3.1) | 10 (3.3) | 3 (2.6) | 0 (0.0) | 2 (4.1) |
| Drinking, n (%) | 19 (3.9) | 9 (3.0) | 6 (5.2) | 0 (0.0) | 4 (8.2) |
| Smoking, n (%) | 57 (11.8) | 26 (8.7) | 22 (19.0) | 2 (11.1) | 7 (14.3) |
| Symptoms, n (%) | |||||
| Asymptomatic | 238 (49.3) | 141 (47.0) | 60 (51.7) | 10 (55.6) | 27 (55.1) |
| Abdominal pain | 152 (31.5) | 104 (34.7) | 31 (26.7) | 3 (16.7) | 14 (28.6) |
| Abdominal discomfort | 89 (18.4) | 52 (17.3) | 24 (20.7) | 5 (27.8) | 8 (16.3) |
| Bleeding | 4 (0.8) | 3 (1.0) | 1 (0.9) | 0 (0.0) | 0 (0.0) |
SETs, subepithelial tumours; ESE, endoscopic submucosal excavation; EFTR, endoscopic full-thickness resection; STER, submucosal tunnelling endoscopic resection; ESD, endoscopic submucosal dissection.
Tumour features
As shown in Table 2, the median tumour size was 1.2 cm, with a range from 0.5 to 6.0 cm. Predominantly, tumours were located in the upper and middle third of the stomach (44.3%, 214/483), followed by the lower third of the stomach (11.4%, 55/483). In terms of growth characteristics, 94.0% (454/483) showed intraluminal growth, while 6.0% (29/483) exhibited primarily extraluminal growth. Surface erosion or ulceration was observed in only 9 tumours. The majority of gastric SETs were regular in shape (81.4%, 393/483), with the remaining being irregular (18.6%, 90/483). EUS findings before procedures indicated a homogeneous echo pattern in 76.2% (368/483) of gastric SETs, while 14.3% (69/483) exhibited an inhomogeneous echo pattern. Most SETs originated from the MP (85.9%, 415/483), followed by the submucosa (12.0%, 58/483) and muscularis mucosa (2.1%, 10/483). Pathological analysis revealed that the majority of gastric SETs were GISTs (48.2%, 233/483), followed by leiomyomas (32.5%, 157/483), with the remaining 19.3% consisting of various other types including ectopic pancreas (n = 35), neuroendocrine tumours (n = 13), and others. According to the NIH risk classification, the majority of GISTs were categorized as very low risk (81.5%, 190/233), with low risk accounting for 11.2% (26/233), intermediate risk for 4.3% (10/233), and high risk for 3.0% (7/233).
Table 2.
Pathologic and clinical outcomes among different endoscopic procedures.
| Total | ESE | EFTR | STER | ESD | |
|---|---|---|---|---|---|
| Number | 483 | 300 | 116 | 18 | 49 |
| Size, cma | 1.2 (0.5–6.0) | 1.0 (0.5–6.0) | 1.3 (0.6–6.0) | 2.0 (1.2–6.0) | 1.1 (0.6–3.0) |
| Tumour location, n (%) | |||||
| Upper third | 214 (44.3) | 123 (41.0) | 66 (56.9) | 18 (100.0) | 7 (14.3) |
| Middle third | 214 (44.3) | 158 (52.7) | 43 (37.1) | 0 (0.0) | 13 (26.5) |
| Lower third | 55 (11.4) | 19 (6.3) | 7 (6.0) | 0 (0.0) | 29 (59.2) |
| Tumour growth pattern, n (%) | |||||
| Intraluminal growth | 454 (94.0) | 292 (97.3) | 97 (83.6) | 16 (88.9) | 49 (100.0) |
| Extraluminal growth | 29 (6.0) | 8 (2.7) | 19 (16.4) | 2 (11.1) | 0 (0.0) |
| Shape, n (%) | |||||
| Regular | 393 (81.4) | 246 (82.0) | 90 (77.6) | 10 (55.6) | 47 (95.9) |
| Irregular | 90 (18.6) | 54 (18.0) | 26 (22.4) | 8 (44.4) | 2 (4.1) |
| Erosion or ulceration on the surface, n (%) | 9 (1.9) | 1 (0.3) | 1 (0.9) | 3 (16.7) | 4 (8.2) |
| EUS characteristics, n (%)b | |||||
| Homogeneous echo | 368 (76.2) | 243 (81.0) | 92 (79.3) | 14 (77.8) | 19 (38.8) |
| Inhomogeneous echo | 69 (14.3) | 32 (10.7) | 14 (12.1) | 2 (11.1) | 21 (42.9) |
| Layer, n (%) | |||||
| Muscularis mucosa | 10 (2.1) | 1 (0.3) | 0 (0.0) | 2 (11.1) | 7 (14.3) |
| Submucosa | 58 (12.0) | 13 (4.3) | 2 (1.7) | 1 (5.6) | 42 (85.7) |
| Muscularis propria | 415 (85.9) | 286 (95.3) | 114 (98.3) | 15 (83.3) | 0 (0.0) |
| Tumour pathology, n (%) | |||||
| GIST | 233 (48.2) | 127 (42.3) | 103 (88.8) | 3 (16.7) | 0 (0.0) |
| Leiomyoma | 157 (32.5) | 139 (46.3) | 2 (1.7) | 14 (77.8) | 2 (4.1) |
| Others | 93 (19.3) | 34 (11.3) | 11 (9.5) | 1 (5.6) | 47 (95.9) |
| Complete resection, n (%) | 453 (93.8) | 288 (96.0) | 107 (92.2) | 14 (77.8) | 44 (89.8) |
| Procedure time, mina | 45 (8–550) | 40 (8–314) | 64 (10–550) | 60 (30–294) | 30 (10–177) |
| Postoperative stay, daysa | 5 (1–27) | 5 (1–27) | 6 (3–17) | 6 (3–19) | 4 (1–11) |
| Estimated blood loss, mLa | 0 (0–400) | 0 (0–200) | 0 (0–400) | 0 (0–20) | 0 (0–20) |
| Complication, n (%) | 33 (6.8) | 16 (5.3) | 7 (6.0) | 3 (16.7) | 7 (14.3) |
| Fever | 20 (4.1) | 10 (3.3) | 4 (3.4) | 2 (11.1) | 4 (8.2) |
| Perioperative perforation | 8 (1.7) | 3 (1.0) | 2 (1.7) | 3 (16.7) | 0 (0.0) |
| Perioperative bleeding | 10 (2.1) | 4 (1.3) | 3 (2.6) | 0 (0.0) | 3 (6.1) |
| Follow-up period, monthsa,c | 7 (2–79) | 7 (2–74) | 6 (2–79) | 6 (2–34) | 7 (2–48) |
ESE, endoscopic submucosal excavation; EFTR, endoscopic full-thickness resection; STER, submucosal tunnelling endoscopic resection; ESD, endoscopic submucosal dissection; GIST, gastrointestinal stromal tumour.
aMedian (range).
bEvaluated in 437 patients.
cEvaluated in 448 patients.
Treatment and clinical outcomes
Table 2 further delineates the clinical outcomes for all SETs across various endoscopic procedures. In this study, the majority of patients underwent ESE (62.1%, 300/483), followed by EFTR (24.0%, 116/483), ESD (10.1%, 49/483), and STER (3.7%, 18/483). The median duration of procedures was 45 min, ranging from 8 to 550 min, and the median postoperative hospital stay was 5 days (range: 1–27 days). Estimated blood loss during procedures varied from 0 to 400 mL.
Complete resection was achieved in 453 cases (93.8%), with the highest rates observed in ESE (96.0%), followed by EFTR (92.2%), ESD (89.8%), and STER (77.8%). Adverse events occurred in 33 patients (6.8%), including fever (n = 20), perioperative bleeding (n = 10, involving 4 cases treated with ESE, 3 with EFTR, and 3 with ESD), and perforation (n = 8, distributed among 3 cases treated with ESE, 3 with STER, and 2 with EFTR), with some patients experiencing multiple complications. Four patients required additional surgical interventions as a result of major complications.
The follow-up period ranged between 2 and 79 months, with a median duration of 7 months. During this period, no residual lesions were detected, and recurrence was observed in only one patient with GIST, which occurred two years after ER. However, it is noteworthy that thirty-five patients were lost to follow-up.
Risk factors associated with incomplete resection and complications
Multivariate logistic regression analysis identified significant risk factors for incomplete resection, perioperative bleeding, and perforation. Tumour size was a significant predictor of incomplete resection, with OR of 1.605 (95% CI: 1.066–2.416, p = 0.023). Larger tumours were also associated with a higher risk of perioperative bleeding (OR = 2.004, 95% CI: 1.099–3.653, p = 0.023) and perforation (OR = 3.476, 95% CI: 1.830–6.602, p < 0.001). Moreover, an irregular tumour shape markedly increased the likelihood of incomplete resection (OR = 10.771, 95% CI: 3.452–33.605, p < 0.001). Conversely, SETs located in the middle third (OR = 0.068, 95% CI: 0.014–0.341, p = 0.001) and upper third (OR = 0.211, 95% CI: 0.051–0.884, p = 0.033) of the stomach substantially reduced the risk of incomplete resection (Table 3).
Table 3.
Multivariate analysis for incomplete resection, perioperative bleeding, and perioperative perforation during endoscopic resection procedure.
| Incomplete resection |
Perioperative bleeding |
Perioperative perforation |
||||
|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Age, years | – | 0.602 | – | 0.785 | – | 0.353 |
| Sex (male, female) | – | 0.357 | – | 0.359 | – | 0.584 |
| Size, cm | 1.605 (1.066–2.416) | 0.023 | 2.004 (1.099–3.653) | 0.023 | 3.476 (1.830–6.602) | <0.001 |
| Tumour growth pattern (intraluminal, extraluminal) |
– | 0.098 | – | 0.998 | – | 0.998 |
| Tumour location | – | 0.005 | – | 0.992 | – | 0.865 |
| Lower third | 1 | – | – | – | – | – |
| Middle third | 0.068 (0.014–0.341) | 0.001 | – | – | – | – |
| Upper third | 0.211 (0.051–0.884) | 0.033 | – | – | – | – |
| Erosion or ulceration on the surface (yes or no) | – | 0.185 | – | 0.999 | – | 0.999 |
| EUS characteristics (inhomogeneous or homogeneous echo) | – | 0.701 | – | 0.895 | – | 0.510 |
| Procedure type (ESE, EFTR, STER, ESD) | – | 0.217 | – | 0.737 | – | 0.451 |
| Tumour pathology (Leiomyoma + others, GIST) | – | 0.443 | – | 0.956 | – | 0.385 |
| Chronic disease (yes or no) | – | 0.592 | – | 0.315 | – | 0.128 |
| Shape (regular, irregular) | 10.771 (3.452–33.605) | <0.001 | – | 0.633 | – | 0.484 |
| Layer (muscularis mucasa, submucosa, MP) | – | 1.000 | – | 0.416 | – | 0.254 |
OR, odds ratio; CI, confidence interval; ESE, endoscopic submucosal excavation; EFTR, endoscopic full-thickness resection; STER, submucosal tunnelling endoscopic resection; ESD, endoscopic submucosal dissection; GIST, gastrointestinal stromal tumour; MP, muscularis propria.
Discussion
Gastric SETs typically present as protruding lesions within the gastric cavity, often covered by an intact mucosal surface. While many SETs, including leiomyomas, ectopic pancreas, and lipomas, are benign, a subset—such as GISTs, neuroendocrine tumours, and glomus tumours—may be malignant or premalignant [22]. Previous studies have demonstrated that EUS characteristics lack specificity in differentiating individual SETs, thereby limiting their diagnostic accuracy. Additionally, routine biopsy and EUS-guided fine-needle aspiration (FNA) often fail to provide a definitive preoperative diagnosis due to insufficient tissue sampling and other technical limitations [5,7]. Although periodic surveillance remains the primary recommendation [3–5,23], the potential for malignancy can cause significant psychological distress on patients. Moreover, advancements in endoscopic techniques have increasingly influenced both physicians and patients to opt for proactive lesion removal upon detection [6,7].
Despite ongoing debate regarding the specific indications for endoscopic treatment of SETs, ER has gained significant recognition as an effective technique for managing upper gastrointestinal SETs [20]. Procedures such as ESD, ESE, STER, and EFTR have shown high rates of complete resection and minimal complications in the treatment of gastric SETs [2,23,24]. In line with previous findings, our study demonstrated a successful complete resection rate exceeding 93.0%, with a complication rate of just 6.8%. Remarkably, there were no procedure-related deaths, and recurrence was observed in only one GIST patient. Throughout the follow-up period, all participants maintained their pre-procedure quality of life. These outcomes underscore the efficacy and safety of ER for gastric SETs, reflecting the rigorous adherence to procedural criteria and the proficiency of our medical team.
Complete resection serves as a critical benchmark for evaluating the clinical efficacy and formulating appropriate follow-up strategies in the management of gastric SETs. Our study achieved a satisfactory complete resection rate of 93.8%. Among the various techniques utilized, ESE demonstrated the highest complete resection rate of 96.0%, while STER exhibited the lowest rate of 77.8%. Discrepancies in these rates could potentially arise from differences in sample size and tumour location, as evidenced in existing literature [10,25,26]. Compared to SETs in the lower third of the stomach, those in other gastric regions are more likely to be completely resected due to several factors. The distal stomach has limited maneuvering space near the pylorus, making endoscopic manipulation more challenging. Additionally, tumours in this region exhibit greater mobility, complicating en bloc resection. Furthermore, the thin submucosal layer and rich vasculature increase the risk of incomplete resection or perforation. In contrast, SETs in the gastric body and fundus have a more favourable anatomical structure, allowing for better visualization, stable dissection, and higher complete resection rates. Importantly, we identified irregular tumour shape and larger tumour size as significant predictors of incomplete resection. These results are consistent with prior studies that have emphasized the role of irregular morphology and larger tumour diameters in contributing to incomplete resection of SETs [19,21,27]. Furthermore, previous studies have indicated that tumour size and shape are closely linked to the incidence of clinical complications [28,29], underscoring the importance of integrating tumour dimensions and morphology into treatment planning for gastric SETs.
Ensuring safety in ER remains a primary concern, despite the overall low complication rates [14,30]. Haemorrhage poses a considerable risk throughout endoscopic procedures and post-procedure [31]. While minor bleeding is relatively common, it is generally manageable through immediate haemostatic measures using endoknife tip or forceps. Effective haemostasis hinges on maintaining a clear endoscopic field and carefully dissecting to pinpoint major blood vessels. If endoscopic methods prove insufficient, surgical intervention may be required. In this study, five patients experienced significant intraoperative bleeding due to vascular injuries; two of these were managed endoscopically while three required surgical intervention. Additionally, postoperative bleeding occurred in five patients; four were successfully treated endoscopically, and one case was managed conservatively. All instances of bleeding were ultimately resolved. Perforation during ER procedures may result in the extravasation of blood and fluids into the abdominal cavity, thereby heightening the risk of severe complications, including peritonitis and the development of abdominal abscesses. Such perioperative perforations are deemed serious complications, associated with substantial morbidity and increased mortality rates [20,32]. Recent studies have reported comparably low rates of perforation in both ESE and STER [29,33]. However, intentional perforations during EFTR are necessary for complete lesion removal. To minimize these risks, several techniques have been developed, such as the thread-traction method to improve visualization, and the use of nylon loops and metallic endoclips for secure closure. Advanced devices like the OverStitch system and over-the-scope clip (OTSC) system have also been introduced to simplify the closure process [34,35]. In our cohort, three individuals necessitated surgical intervention due to perforation during ESE and EFTR, whereas five were effectively managed with endoscopic techniques during ESE and STER. Among them, two patients experiencing both bleeding and perforation underwent surgical intervention. Our multivariate analysis notably revealed that larger tumour size increased the risk of perioperative bleeding and perforation, consistent with previous research [20,36]. Optimal management of gastric SETs necessitates meticulous patient selection and the development of personalized treatment strategies. Customizing therapeutic approaches according to both patient profiles and tumour characteristics enables clinicians to optimize clinical outcomes and reduce procedural risks.
This study offers important insights into the efficacy and safety of advanced endoscopic methods for treating gastric SETs arising from different layers of the gastric wall. The large cohort of 483 patients allows for robust analysis of risk factors for complications and incomplete resection. Our detailed evaluation of tumour characteristics, such as size, shape, and location, provides practical information for personalized patient management. The real-world nature of our data further enhances clinical relevance, offering insights applicable to routine practice. Nevertheless, certain limitations must be considered. Firstly, despite the relatively large sample size, the retrospective design, variability in follow-up duration, and some degree of loss to follow-up may introduce potential biases. Secondly, as the procedures were performed by highly experienced endoscopists in a single-centre setting, the generalizability of our results may be limited. Thirdly, the median follow-up period of 7 months is relatively short, particularly for assessing the long-term outcomes of malignant or potentially malignant tumours such as GISTs. This limitation may affect the ability to evaluate late recurrence or delayed disease progression, underscoring the need for longer-term surveillance. Finally, the study included various endoscopic procedures with differing sample sizes, and the combined analysis approach may have overlooked procedure-specific risk factors due to limited statistical power for individual analyses. To further validate the efficacy and safety of these techniques, large-scale, multicentre, prospective studies with standardized follow-up protocols and extended follow-up durations are warranted.
Conclusions
ER is an effective and generally safe therapeutic approach for gastric SETs when performed by experienced endoscopists. A larger tumour size and irregular shape are significant risk factors that can lead to adverse outcomes during ER procedures, underscoring the importance of meticulous preoperative assessment and careful selection of the most appropriate resection techniques.
Funding Statement
This work was supported by the Fujian Provincial Natural Science Foundation (2022J05308) and the Medical and Health Guiding Project of Xiamen (3502Z20224ZD1009).
Authors contributions
Ji Li: Funding acquisition, Investigation, Project administration, Writing – original draft; Dong Xu: Data curation, Formal analysis, Investigation, Resources; Weifeng Huang: Data curation; Xiaoyi Lei: Data curation; Yanqing Wang: Conceptualization, Supervision, Writing – review & editing; Jinyan Zhang: Funding acquisition, Project administration, Writing – review & editing.
Ethical approval
This retrospective study was conducted in accordance with the ethical standards of the Declaration of Helsinki and was granted an exemption from ethical review by the ethics committee of the First Affiliated Hospital of Xiamen University.
Consent form
The written informed consent was waived due to the retrospective setting by the Institutional Review Boards of the First Affiliated Hospital of Xiamen University.
Disclosure statement
No potential conflict of interest was reported by the authors.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1.Nishida T, Kawai N, Yamaguchi S, et al. Submucosal tumors: comprehensive guide for the diagnosis and therapy of gastrointestinal submucosal tumors. Dig Endosc. 2013;25(5):479–489. doi: 10.1111/den.12149. [DOI] [PubMed] [Google Scholar]
- 2.Shichijo S, Abe N, Takeuchi H, et al. Endoscopic resection for gastric submucosal tumors: Japanese multicenter retrospective study. Dig Endosc. 2023;35(2):206–215. doi: 10.1111/den.14446. [DOI] [PubMed] [Google Scholar]
- 3.Li J, Ye Y, Wang J, et al. Chinese consensus guidelines for diagnosis and management of gastrointestinal stromal tumor. Chin J Cancer Res. 2017;29(4):281–293. doi: 10.21147/j.issn.1000-9604.2017.04.01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Serrano C, Martín-Broto J, Asencio-Pascual JM, et al. 2023 GEIS Guidelines for gastrointestinal stromal tumors. Ther Adv Med Oncol. 2023;15:17588359231192388. doi: 10.1177/17588359231192388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Deprez PH, Moons LMG, OʼToole D, et al. Endoscopic management of subepithelial lesions including neuroendocrine neoplasms: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy. 2022;54(4):412–429. doi: 10.1055/a-1751-5742. [DOI] [PubMed] [Google Scholar]
- 6.Zhu L, Khan S, Hui Y, et al. Treatment recommendations for small gastric gastrointestinal stromal tumors: positive endoscopic resection. Scand J Gastroenterol. 2019;54(3):297–302. doi: 10.1080/00365521.2019.1578405. [DOI] [PubMed] [Google Scholar]
- 7.Hu J, Sun X, Ge N, et al. The necessarity of treatment for small gastric subepithelial tumors (1–2 cm) originating from muscularis propria: an analysis of 972 tumors. BMC Gastroenterol. 2022;22(1):182. doi: 10.1186/s12876-022-02256-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kim SY, Kim KO.. Management of gastric subepithelial tumors: the role of endoscopy. World J Gastrointest Endosc. 2016;8(11):418–424. doi: 10.4253/wjge.v8.i11.418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lu J, Jiao T, Zheng M, et al. Endoscopic resection of submucosal tumors in muscularis propria: the choice between direct excavation and tunneling resection. Surg Endosc. 2014;28(12):3401–3407. doi: 10.1007/s00464-014-3610-y. [DOI] [PubMed] [Google Scholar]
- 10.Tan Y, Zhou B, Zhang S, et al. Submucosal tunneling endoscopic resection for gastric submucosal tumors: a comparison between cardia and non-cardia location. J Gastrointest Surg. 2019;23(11):2129–2135. doi: 10.1007/s11605-019-04182-4. [DOI] [PubMed] [Google Scholar]
- 11.Abe N, Takeuchi H, Ohki A, et al. Comparison between endoscopic and laparoscopic removal of gastric submucosal tumor. Dig Endosc. 2018;30(S1):7–16. doi: 10.1111/den.13010. [DOI] [PubMed] [Google Scholar]
- 12.Li J, Chen JH, Huang WF, et al. Gastric schwannoma: a retrospective analysis of clinical characteristics, treatments, and outcomes. Asian J Surg. 2024;47(1):407–412. doi: 10.1016/j.asjsur.2023.09.042. [DOI] [PubMed] [Google Scholar]
- 13.Yu C, Liao G, Fan C, et al. Long-term outcomes of endoscopic resection of gastric GISTs. Surg Endosc. 2017;31(11):4799–4804. doi: 10.1007/s00464-017-5557-2. [DOI] [PubMed] [Google Scholar]
- 14.Andalib I, Yeoun D, Reddy R, et al. Endoscopic resection of gastric gastrointestinal stromal tumors originating from the muscularis propria layer in North America: methods and feasibility data. Surg Endosc. 2018;32(4):1787–1792. doi: 10.1007/s00464-017-5862-9. [DOI] [PubMed] [Google Scholar]
- 15.Zhai YQ, Chai NL, Li HK, et al. Endoscopic submucosal excavation and endoscopic full-thickness resection for gastric schwannoma: five-year experience from a large tertiary center in China. Surg Endosc. 2020;34(11):4943–4949. doi: 10.1007/s00464-019-07285-w. [DOI] [PubMed] [Google Scholar]
- 16.Dalal I, Andalib I.. Advances in endoscopic resection: a review of endoscopic submucosal dissection (ESD), endoscopic full thickness resection (EFTR) and submucosal tunneling endoscopic resection (STER). Transl Gastroenterol Hepatol. 2022;7:19. doi: 10.21037/tgh-2020-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Joensuu H. Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Hum Pathol. 2008;39(10):1411–1419. doi: 10.1016/j.humpath.2008.06.025. [DOI] [PubMed] [Google Scholar]
- 18.Cotton PB, Eisen GM, Aabakken L, et al. A lexicon for endoscopic adverse events: report of an ASGE workshop. Gastrointest Endosc. 2010;71(3):446–454. doi: 10.1016/j.gie.2009.10.027. [DOI] [PubMed] [Google Scholar]
- 19.Du C, Chai N, Linghu E, et al. Clinical outcomes of endoscopic resection for the treatment of gastric gastrointestinal stromal tumors originating from the muscularis propria: a 7-year experience from a large tertiary center in China. Surg Endosc. 2022;36(2):1544–1553. doi: 10.1007/s00464-021-08443-9. [DOI] [PubMed] [Google Scholar]
- 20.Ye LP, Zhang Y, Luo DH, et al. Safety of endoscopic resection for upper gastrointestinal subepithelial tumors originating from the muscularis propria layer: an analysis of 733 tumors. Am J Gastroenterol. 2016;111(6):788–796. doi: 10.1038/ajg.2015.426. [DOI] [PubMed] [Google Scholar]
- 21.Li J, Xu D, Huang WF, et al. Efficacy and safety of endoscopic resection for gastric gastrointestinal stromal tumors originating from the muscularis propria. Dig Dis Sci. 2024;69(6):2184–2192. doi: 10.1007/s10620-024-08359-z. [DOI] [PubMed] [Google Scholar]
- 22.Cho JW, Korean ESD Study Group . Current guidelines in the management of upper gastrointestinal subepithelial tumors. Clin Endosc. 2016;49(3):235–240. doi: 10.5946/ce.2015.096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kim CG. Endoscopic treatment for gastric subepithelial tumor. J Gastric Cancer. 2024;24(1):122–134. doi: 10.5230/jgc.2024.24.e11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Godat S, Robert M, Caillol F, et al. Efficiency and safety of endoscopic resection in the management of subepithelial lesions of the stomach. United European Gastroenterol J. 2016;4(2):250–256. doi: 10.1177/2050640615604774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wang H, Tan Y, Zhou Y, et al. Submucosal tunneling endoscopic resection for upper gastrointestinal submucosal tumors originating from the muscularis propria layer. Eur J Gastroenterol Hepatol. 2015;27(7):776–780. doi: 10.1097/MEG.0000000000000394. [DOI] [PubMed] [Google Scholar]
- 26.Inoue H, Ikeda H, Hosoya T, et al. Submucosal endoscopic tumor resection for subepithelial tumors in the esophagus and cardia. Endoscopy. 2012;44(3):225–230. doi: 10.1055/s-0031-1291659. [DOI] [PubMed] [Google Scholar]
- 27.Du C, Ma L, Chai N, et al. Factors affecting the effectiveness and safety of submucosal tunneling endoscopic resection for esophageal submucosal tumors originating from the muscularis propria layer. Surg Endosc. 2018;32(3):1255–1264. doi: 10.1007/s00464-017-5800-x. [DOI] [PubMed] [Google Scholar]
- 28.Chen Q, Yu M, Lei Y, et al. Efficacy and safety of endoscopic submucosal dissection for large gastric stromal tumors. Clin Res Hepatol Gastroenterol. 2020;44(1):90–100. doi: 10.1016/j.clinre.2019.03.004. [DOI] [PubMed] [Google Scholar]
- 29.Xu HW, Zhao Q, Yu SX, et al. Comparison of different endoscopic resection techniques for submucosal tumors originating from muscularis propria at the esophagogastric junction. BMC Gastroenterol. 2019;19(1):174. doi: 10.1186/s12876-019-1099-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lv XH, Wang CH, Xie Y.. Efficacy and safety of submucosal tunneling endoscopic resection for upper gastrointestinal submucosal tumors: a systematic review and meta-analysis. Surg Endosc. 2017;31(1):49–63. doi: 10.1007/s00464-016-4978-7. [DOI] [PubMed] [Google Scholar]
- 31.Yamamoto Y, Uedo N, Abe N, et al. Current status and feasibility of endoscopic full-thickness resection in Japan: results of a questionnaire survey. Dig Endosc. 2018;30(S1):2–6. doi: 10.1111/den.13045. [DOI] [PubMed] [Google Scholar]
- 32.Kang DH, Ryu DG, Choi CW, et al. Clinical outcomes of iatrogenic upper gastrointestinal endoscopic perforation: a 10-year study. BMC Gastroenterol. 2019;19(1):218. doi: 10.1186/s12876-019-1139-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ponte Neto FL, de Moura DTH, Sagae VMT, et al. Endoscopic resection of esophageal and gastric submucosal tumors from the muscularis propria layer: submucosal tunneling endoscopic resection versus endoscopic submucosal excavation: a systematic review and meta-analysis. Surg Endosc. 2021;35(12):6413–6426. doi: 10.1007/s00464-021-08659-9. [DOI] [PubMed] [Google Scholar]
- 34.Zhai YQ, Chai NL, Zhang WG, et al. Endoscopic versus surgical resection in the management of gastric schwannomas. Surg Endosc. 2021;35(11):6132–6138. doi: 10.1007/s00464-020-08108-z. [DOI] [PubMed] [Google Scholar]
- 35.Liu S, Zhou X, Yao Y, et al. Resection of the gastric submucosal tumor (G-SMT) originating from the muscularis propria layer: comparison of efficacy, patients’ tolerability, and clinical outcomes between endoscopic full-thickness resection and surgical resection. Surg Endosc. 2020;34(9):4053–4064. doi: 10.1007/s00464-019-07311-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ni M, Tang D, Ren W, et al. Risk factors of perforation in gastric stromal tumors during endoscopic resection: a retrospective case-control study. Gastric Cancer. 2023;26(4):590–603. doi: 10.1007/s10120-023-01391-4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.