A pilot study on the application and efficacy of mucosal traction-assisted clip closure in endoscopic full-thickness resection for gastric submucosal tumors

Abstract

Endoscopic full-thickness resection (EFTR) for gastric submucosal tumors (SMTs) presents a technical challenge for defect closure. This study aims to evaluate the feasibility and efficacy of a novel mucosal traction-assisted clip closure (MTCC) technique compared to conventional clip closure (CCC). In this multi-center retrospective study of 218 patients undergoing EFTR for gastric SMTs at seven hospitals (June 2020–June 2025), patients were divided into MTCC (n = 32) and CCC (n = 186) groups. To minimize selection bias, 1:2 propensity score matching (PSM) yielded a matched cohort of 96 patients (MTCC, n = 32; CCC, n = 64). Primary outcomes were defect closure success rate and closure time. Secondary outcomes included postoperative complications, fasting time, and hospital stay. Subgroup analyses were performed by defect diameter (≤ 2.5 cm vs. >2.5 cm) and tumor location (gastric fundus). Defect closure was successfully achieved in all patients in the MTCC group. After PSM, baseline characteristics were well-balanced between the two groups. The MTCC group continued to demonstrate superior closure efficiency compared to the CCC group, with a significantly shorter median defect closure time (7.0 min vs. 17.0 min, P < 0.001) and fewer titanium clips used (8.0 vs. 12.0, P < 0.001). The median total procedure time was also significantly reduced in the MTCC group (47.0 min vs. 53.0 min, P = 0.027). Postoperatively, patients in the MTCC group had a shorter median fasting time (2.0 days vs. 3.0 days, P = 0.001), while complication rates remained comparable between groups. The advantages of MTCC were consistently observed across subgroups stratified by defect size and tumor location. After adjusting for potential confounders using PSM, the MTCC technique appears to be a feasible method for managing post-EFTR defects. In this cohort, it was associated with significantly enhanced closure efficiency and accelerated postoperative recovery.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-51268-6.

Introduction

Gastric submucosal tumors (SMTs) represent a heterogeneous group of mesenchymal lesions. Among these, gastrointestinal stromal tumors (GISTs) are the most prevalent and clinically significant due to their malignant potential^1–3; other types, such as leiomyomas, schwannomas, glomus tumors, and inflammatory fibroid polyps, present distinct diagnostic and therapeutic challenges^4–6. Widespread endoscopic screening has markedly increased the incidental detection of these lesions, particularly those smaller than 20 mm, which now constitute approximately 90% of all gastric SMTs⁷. Although many small SMTs are benign, a substantial proportion are GISTs that can exhibit malignant behavior even at a very small size, necessitating histopathological confirmation for accurate risk stratification^7,8.

Consequently, the management paradigm has shifted decisively from open surgery toward minimally invasive endoscopic techniques, prioritizing organ preservation and faster patient recovery. Endoscopic full-thickness resection (EFTR) has emerged as an effective method for the complete removal of tumors involving the entire gastric wall layer^1,9, making it particularly suitable for extraluminal SMTs and those located in challenging anatomical sites like the esophagogastric junction^9,10. However, the critical challenge of this procedure lies in securing the resultant full-thickness defect; inadequate closure can lead to fatal complications such as perforation or fistula formation^1,11. While conventional clip closure (CCC) remains the standard technique, it is often time-consuming, inefficient for large or irregular defects, and can be less cost-effective¹, especially for defects exceeding 2.5 cm or those situated in the gastric fundus.

Recent advancements aim to address these limitations through both technological and strategic innovation. On the technological front, devices like the over-the-scope clip (OTSC) system provide a more robust closure for large defects by capturing a greater volume of tissue^11,12, while novel devices such as the through-the-scope twin clip (TTS-TC) are designed to expedite the closure process itself¹³. Beyond device-centric improvements, strategic procedural techniques offer another promising avenue. Traction methods, for instance, have been revolutionary in related fields like endoscopic submucosal dissection (ESD) by dramatically improving surgical exposure and facilitating resection¹⁴. This success suggests that applying an analogous traction-assist principle specifically to the defect closure phase of EFTR—by optimizing tissue alignment and approximation before final clipping—could enhance both the efficiency and reliability of closure, without compromising procedural safety.

Notably, several recent studies have explored traction-assisted strategies. Broadly, these can be categorized into two conceptual approaches:¹ device-assisted traction, which relies on external tools such as dental floss, snares, or magnets^15–17; and² intrinsic tissue-based traction, which leverages the patient’s own preserved tissue—most commonly a mucosal flap—as an anchor without auxiliary devices. Gong et al.¹⁵ described an “internal traction-assisted suspended closure” (ITASC) technique for GI defect closure in a pilot study of 8 patients, demonstrating its feasibility with a median closing time of 13 min. Li et al.¹⁶ reported the efficacy of dental floss and hemoclip as a traction method for EFTR of submucosal tumors in the gastric fundus. However, these studies were limited by small sample sizes, single-center designs, or a focus on the resection rather than the closure phase. More recently, Nomura et al.¹⁸ described an underwater internal traction-assisted reopenable clip-over-the-line method for closing gastric full-thickness defects, further supporting the principle that strategic internal traction can facilitate defect closure.

To our knowledge, this is the first multi-center study to systematically evaluate a closure-phase, intrinsic tissue-based traction technique—using only a preserved mucosal flap and standard reopenable clips—specifically for defect closure after EFTR for gastric SMTs. Unlike device-assisted methods, our mucosal traction-assisted clip closure (MTCC) technique requires no additional equipment or assistant coordination, and it uniquely targets closure (not resection) by converting a round defect into a linear configuration via stepwise clip placement from the traction root. By analyzing a substantial retrospective cohort, we aimed to provide robust comparative evidence on the feasibility, efficacy, and safety of MTCC, to determine if it represents a superior closure strategy that enhances the overall therapeutic efficiency of EFTR and contributes to accelerated patient recovery.

Materials and methods

Patients

This retrospective, multi-center study analyzed data from patients with gastric SMTs who underwent endoscopic treatment at seven participating institutions between June 2020 and June 2025. The collaborating centers were: Suzhou BenQ Medical Center, Wuxi Hospital of Traditional Chinese Medicine, the Affiliated Zhangjiagang Hospital of Soochow University, Changshu Hospital Affiliated to Soochow University, The First Affiliated Hospital of Soochow University, Kunshan First People’s Hospital, and Nanjing First Hospital. Patients were included if they met the following criteria:¹ diagnosis of a gastric SMT managed with endoscopic resection;² normal preoperative laboratory results, including complete blood count and coagulation profile;³ tumor diameter < 5.0 cm. Exclusion criteria comprised:¹ procedures performed via endoscopic submucosal dissection (ESD) or submucosal tunneling endoscopic resection (STER);² history of prior gastric surgery;³ intraoperative wound closure using nylon loops or over-the-scope clips (OTSC);⁴ conversion to open surgery due to intraoperative complications such as uncontrolled hemorrhage or technical unresectability;⁵ incomplete clinical or pathological records. A flowchart detailing patient enrollment is provided in Fig. 1. According to the technique used for post-resection defect closure, patients were categorized into two groups: the MTCC group and the CCC group. All endoscopic interventions were carried out by certified endoscopists with substantial expertise in therapeutic endoscopy. The study protocol received ethical approval from the Institutional Review Board of Changshu Hospital Affiliated to Soochow University and other involved institutions (Approval No: L2026011). The investigation was conducted in compliance with relevant ethical guidelines and regulatory standards. Written informed consent was obtained from all patients prior to the endoscopic procedure, which included permission for the use of their anonymized clinical data for research purposes, including retrospective analyses.

Fig. 1.

The study flow chart. After initial enrollment of 218 patients (MTCC = 32, CCC = 186), a 1:2 propensity score matching was performed to balance baseline characteristics, resulting in a matched cohort of 96 patients (MTCC = 32, CCC = 64) for the primary analysis.

Endoscopic procedures and equipment

In this study, EFTR was the standard surgical approach for all enrolled cases. The principal intergroup variable was the technique adopted for defect closure, which was classified into two distinct modalities: MTCC and CCC. The procedural protocol for the CCC group conformed to established methods documented in prior literature¹⁹. The operative sequence for the MTCC group consisted of the following steps:¹ lesion identification via white-light endoscopy;² circumferential submucosal injection around the tumor;³ a partial (approximately three-quarter) circumferential incision of the mucosa overlying the lesion to prepare for subsequent traction;⁴ meticulous dissection within the plane between the mucosal layer and the tumor to achieve complete separation and full exposure of the lesion;⁵ continuation of dissection along the tumor periphery, with a switch to an insulated-tip (IT) knife upon intentional entry into the gastric cavity (i.e., creation of a full-thickness defect) to finalize the resection;⁶ grasping of the prepared mucosal flap with titanium clips, accompanied by controlled suction to reduce intragastric tension;⁷ strategic adjustment of the traction vector applied through the clips to reconfigure the residual defect into a near-linear shape;⁸ once optimal tissue apposition was achieved, repositioning the clips to anchor the mobilized mucosal edge to the adjacent gastric muscular layer (a maneuver facilitated by the prior reduction in luminal pressure);⁹ sequential application of additional titanium clips from the base of the traction point to complete the final closure of the wound. Figure 2 illustrates a representative application of the MTCC technique for a submucosal tumor located in the gastric fundus, while Fig. 3 depicts its use for a lesion situated on the anterior wall of the gastric angle. The schematic diagrams of the surgical procedure in the MTCC group are shown in Supplementary Fig. 1.

Fig. 2.

Closing the defect after full-thickness resection of gastric SMTs in the fundus using mucosal traction-assisted clip closure technique. (A) Gastric SMTs located in the fundus; (B) Submucosal injection to elevate the mucosal layer covering the tumor; (C) Incising approximately three-quarters of the mucosal circumference followed by dissection between the tumor and the mucosal layer to separate them; (D) Stepwise full-thickness dissection along the tumor margin; (E) Further full-thickness dissection of the tumor until complete resection; (F) Defect after full-thickness resection; (G) The mucosal flap (circled) is grasped with a titanium clip to provide traction, which linearizes the defect; no external traction device is used; (H) Placing clips sequentially starting from the root of the mucosal traction to gradually close the defect; (I) Complete closure of the defect; (J) The fully resected SMT specimen.

Fig. 3.

Closing the defect after full-thickness resection of gastric SMTs in the anterior wall of the gastric angulus using mucosal traction-assisted clip closure technique. (A) Gastric SMTs located in the anterior wall of the gastric angulus; (B) Submucosal injection to elevate the mucosal layer covering the tumor; (C) Incising the mucosal layer over the tumor followed by stepwise dissection along the tumor margin; (D) Full-thickness resection after complete exposure of the tumor; (E) Defect after tumor resection; (F) The mucosal flap (circled) is grasped with a titanium clip to provide traction, which linearizes the defect; no external traction device is used; (G) Placing clips sequentially starting from the root of the mucosal traction; (H) Placing clips sequentially to gradually close the defect; (I) Complete closure of the defect; (J) The fully resected SMT specimen.

All interventions were performed under general endotracheal anesthesia with patients maintained in the left lateral decubitus position throughout the entire procedure; no intraoperative positional changes were performed. A single-channel video gastroscope (models: Olympus GIF-Q260J or Sonoscape EG-550), fitted with a distal transparent cap, was employed for all procedures. No multibending endoscopes (e.g., Olympus GIF-2TQ240M) or other specialized scopes were used in any of the participating centers during the study period. Carbon dioxide (CO₂) insufflation was used for pneumoperitoneum. Tissue dissection was carried out using specialized electrosurgical knives (such as the Olympus KD-650 L dual knife, KD-611 L insulated-tip knife, or the Vedkang Kunpeng knife) connected to and powered by an ERBE VIO 200D electrosurgical generator. Standard ancillary endoscopic equipment, including submucosal injection needles, titanium clips, and hot biopsy forceps, were available and utilized as needed. All participating centers used the same model of reopenable titanium clips (model AG-51042-1950-135-16, Anjieisi Medical Device Co., Hangzhou, China), which have a maximum opening width of 16 mm when fully deployed. These clips allow repeated opening and closing before final deployment, which is essential for the “hold-and-drag” maneuver in the MTCC technique. To ensure successful traction and fixation, the following steps are recommended: (i) the reopenable clip is used to grasp the preserved mucosal flap under direct visualization; (ii) if the initial grasp or traction vector is suboptimal, the clip is reopened and repositioned to a more suitable location on the flap without discarding the clip; (iii) gentle traction is then applied by carefully withdrawing or retroflexing the endoscope while the clip maintains grasp on the flap, which helps linearize the defect; (iv) once satisfactory tissue apposition is achieved, the clip is finally deployed. The use of reopenable clips significantly enhances the technical reproducibility and clinical applicability of the MTCC technique.

Technique selection and assignment

This was a retrospective, non-randomized study. Assignment of patients to either the MTCC or CCC group was not protocol-driven but reflected real-world clinical practice across the seven participating centers between June 2020 and June 2025. Initially, all centers performed CCC as the standard closure method. Following technical standardization and team-based training on the MTCC technique (starting in 2024), MTCC was selectively introduced. The choice of closure method was influenced by a combination of factors, including the operator’s proficiency with the novel technique, the adequacy of the mucosal flap created during resection, and the perceived feasibility of traction-assisted linearization of the defect. As a result, temporal trends, center-specific experience, and operator preference all contributed to the assignment. Assignment of patients to the MTCC or CCC group was determined by reviewing the detailed operative reports from each center, which routinely documented the specific defect closure technique employed.

Postoperative management

After removal, the tumor was stored in an airtight container with 10% formalin and transferred to the pathology lab for paraffin embedding, sectioning, and immunohistochemical analysis targeting markers such as CD117, CD34, and DOG-1. Postoperatively, all patients received gastrointestinal decompression and were kept nil by mouth for 24 to 48 h, after which the decompression tube was withdrawn according to individual recovery. Intravenous therapy comprised proton pump inhibitors, hemostatics, fluid resuscitation, and prophylactic second-generation cephalosporins. Close surveillance was maintained for potential complications, including hemorrhage, perforation, or infection, with interventions applied as clinically indicated.

Data collection

The variables analyzed in this study covered patient demographics, clinical characteristics, intraoperative details, and postoperative outcomes. Demographic and clinical variables included sex, age, body mass index (BMI), presenting symptoms (such as abdominal discomfort, gastrointestinal bleeding, or asymptomatic status), and medical history (including hypertension, diabetes mellitus, and coronary artery disease). Tumor-related characteristics comprised location, size, shape, and depth of invasion. Intraoperative parameters recorded were defect diameter, total procedure duration, endoscopic resection time, time required for defect closure, and the number of titanium clips utilized. Postoperative evaluation included pathological diagnosis, incidence of complications (including bleeding, perforation, and infection), duration of postoperative fasting, and total postoperative hospital stay.

Definitions

Tumor location, size, shape, and invasion depth were primarily evaluated via preoperative endoscopic ultrasound (EUS). Satisfactory closure was defined as complete endoscopic closure of the defect without evidence of abdominal abscess or subsequent perforation¹⁵. Defect diameter measurement: The maximal diameter of the full-thickness defect was measured immediately after tumor resection and before any traction or clip placement (i.e., at the native, untracted state). Measurements were performed endoscopically using the opening width of a standard titanium clip (16 mm, model AG-51042-1950-135-16, Anjieisi Medical Device Co., Hangzhou, China) — the same clip model used for defect closure in all participating centers — as a real-time reference. The maximal diameter was defined as the longest linear distance between two opposing edges of the defect. This measurement protocol was standardized across all seven participating centers through a pre-study training session for all endoscopists. Of note, the measured defect diameter is consistently larger than the preoperative tumor diameter for two reasons:¹ EFTR requires en bloc resection with a clear negative margin (typically 3–5 mm of surrounding healthy tissue beyond the tumor capsule);² after full-thickness resection involving all gastric wall layers (mucosa, submucosa, muscularis propria, and serosa), the defect may “spring open” due to retraction of the muscular and serosal layers, which is a well-recognized phenomenon in the EFTR literature^1,9. Total procedure time was defined as the period from initial submucosal injection to mucosal elevation until final defect closure. Endoscopic resection time spanned from the start of submucosal injection for mucosal lifting to the complete removal of the tumor. Defect closure time was recorded from the placement of the first titanium clip to secure the defect until the final clip was deployed (for the MTCC group, timing commenced upon initiating mucosal traction with clips). Postoperative bleeding was characterized by clinically significant hematemesis or melena following endoscopic intervention, accompanied by a hemoglobin drop of ≥ 2 g/dL, alterations in vital signs, and endoscopic verification of hemorrhage originating from the mucosal defect or gastric site²⁰. Postoperative perforation refers to the occurrence of peritoneal irritation signs or imaging-confirmed subdiaphragmatic free air (via abdominal X-ray or CT) during the recovery period after the tumor has been resected and the intraoperative perforation has been endoscopically closed²¹. Postoperative infection was identified when preoperative white blood cell count, neutrophil percentage, C-reactive protein (CRP), and body temperature were within normal limits, but postoperative presentation included fever (37.3 °C), leukocytosis (10 × 10⁹/L), neutrophilia (70%), elevated CRP (10 mg/L), or positive microbiological findings²².

Follow-up

All patients were required to undergo endoscopic evaluation at 6 and 12 months after surgery. The follow-up included EUS to assess mucosal healing and to identify any residual or recurrent tumors. If no clinical symptoms or signs emerged, further endoscopic surveillance was reduced to a yearly schedule. For metastasis monitoring, patients received annual imaging with contrast-enhanced computed tomography (CT) of the abdomen and chest to evaluate possible lymph node involvement or distant organ metastasis.

Sample size calculation

The required sample size for this retrospective comparative study was not determined prospectively by a formal power calculation, as it was designed based on available clinical data from multiple centers over a defined period (June 2020 to June 2025). The study included all consecutive patients who met the eligibility criteria during this timeframe, resulting in a total cohort of 218 patients (MTCC: n = 32; CCC: n = 186). Based on the observed difference in median closure time (7.0 min vs. 19.0 min) with narrow interquartile ranges, the 95% confidence interval for the difference in medians was estimated to exclude clinically negligible effects. The consistency of statistically significant and clinically relevant findings across all primary, secondary, and subgroup analyses, combined with the large effect sizes observed (Cohen’s d > 1.0), supports that the enrolled sample size was sufficient to detect meaningful differences in the primary outcome between the MTCC and CCC groups. Nevertheless, the unequal group sizes (32 vs. 186) represent a limitation, and the findings should be interpreted in this context.

Statistical analysis

Categorical variables are presented as frequencies and percentages, with group comparisons conducted using the chi-square test or Fisher’s exact test when expected cell counts were less than 5. For continuous data that did not follow a normal distribution, medians and interquartile ranges (IQR) were reported, and the Mann-Whitney U test was used for inter-group analysis. To minimize potential selection bias and confounding due to the retrospective design, propensity score matching (PSM) was performed. A logistic regression model was used to estimate propensity scores based on baseline characteristics, including sex, age, BMI, primary symptoms, underlying diseases, tumor location, tumor size, tumor shape, and depth of invasion. Patients in the MTCC group were matched 1:2 to patients in the CCC group using a caliper width of 0.1 times the standard deviation of the logit of the propensity score. Higher ratios (e.g., 1:3 or 1:4) were considered but not feasible due to the limited number of eligible controls within the prespecified caliper width (0.1 × standard deviation (SD) of the logit of the propensity score) for several MTCC cases. Forcing higher ratios would have required either widening the caliper (risking poorer covariate balance) or excluding some MTCC cases. The 1:2 ratio allowed retention of all 32 MTCC patients while maintaining excellent balance, as reflected by standardized mean differences (SMDs) below 0.1 for most covariates. SMDs were calculated to assess balance after matching, with an SMD < 0.1 indicating adequate balance. All analyses were performed on both the original cohort and the matched cohort, with the latter serving as the primary analysis for outcome comparisons. Statistical significance was defined as a two-sided p-value < 0.05. All analyses were performed using R software (version 4.1.0).

Results

Baseline characteristics

A total of 218 patients who underwent EFTR for gastric SMTs were included, with 32 patients in the MTCC group and 186 patients in the CCC group. Before matching, baseline characteristics were generally comparable, although the MTCC group had a numerically higher proportion of female patients (50.0% vs. 36.0%) and asymptomatic individuals (34.4% vs. 19.9%), while the CCC group had a slightly higher median age (54.0 vs. 49.0 years) (Table 1). To address potential selection bias and confounding, 1:2 PSM was successfully performed, resulting in a matched cohort of 96 patients (MTCC: n = 32; CCC: n = 64). After matching, baseline characteristics were generally well-balanced between the two groups, with the majority of standardized mean differences (SMDs) below 0.1. Although slight imbalances were observed for BMI < 18.5 kg/m² (SMD = -0.160) and underlying diseases (SMD = 0.258), the remaining covariates achieved good balance (SMD < 0.1) (Table 1).

Table 1.

Comparison of preoperative characteristics between the two groups before and after PSM.

Characteristics	Overall (n = 218)	Before PSM			After PSM
		MTCC group (n = 32)	CCC group (n = 186)	P-value	MTCC group (n = 32)	CCC group (n = 64)	SMD
Sex, n (%)				0.133
Male	135 (61.9%)	16 (50.0%)	119 (64.0%)		16 (50.0%)	30 (46.9%)	0.063
Female	83 (38.1%)	16 (50.0%)	67 (36.0%)		16 (50.0%)	34 (53.1%)	-0.063
Age, years, median (IQR)	53.0 (44.0, 62.0)	49.0 (44.0, 59.0)	54.0 (44.0, 62.0)	0.319	49.0 (44.0, 59.0)	51.0 (41.0, 60.0)	0.066
BMI category, kg/m², n (%)				0.427
˂18.5	33 (15.1%)	6 (18.8%)	27 (14.5%)		6 (18.8%)	16 (25.0%)	-0.160
18.5–23.9	110 (50.5%)	13 (40.6%)	97 (52.2%)		13 (40.6%)	25 (39.1%)	0.032
≥ 24.0	75 (34.4%)	13 (40.6%)	62 (33.3%)		13 (40.6%)	23 (35.9%)	0.095
Primary symptom, n (%)				0.076
Abdominal discomfort	162 (74.3%)	19 (59.4%)	143 (76.9%)		19 (59.4%)	42 (65.6%)	-0.127
Asymptomatic	48 (22.0%)	11 (34.4%)	37 (19.9%)		11 (34.4%)	20 (31.3%)	0.066
Hemorrhage	8 (3.7%)	2 (6.3%)	6 (3.2%)		2 (6.3%)	2 (3.1%)	0.129
Underlying diseases, n (%)				0.150
Yes	59 (27.1%)	12 (37.5%)	47 (25.3%)		12 (37.5%)	16 (25.0%)	0.258
No	159 (72.9%)	20 (62.5%)	139 (74.7%)		20 (62.5%)	48 (75.0%)	-0.258
Tumor location, n (%)				0.854
Gastric fundus	152 (69.7%)	22 (68.8%)	130 (69.9%)		22 (68.8%)	42 (65.6%)	0.067
Gastric body/angulus	50 (22.9%)	7 (21.9%)	43 (23.1%)		7 (21.9%)	14 (21.9%)	0.000
Gastric antrum	16 (7.3%)	3 (9.4%)	13 (7.0%)		3 (9.4%)	8 (12.5%)	-0.107
Tumor size, cm, n (%)				0.229
˂2.0	158 (72.5%)	26 (81.3%)	132 (71.0%)		26 (81.3%)	53 (82.8%)	-0.040
≥ 2.0	60 (27.5%)	6 (18.8%)	54 (29.0%)		6 (18.8%)	11 (17.2%)	0.040
Tumor shape, n (%)				0.467
Regular	203 (93.1%)	29 (90.6%)	174 (93.5%)		29 (90.6%)	57 (89.1%)	0.054
Irregular	15 (6.9%)	3 (9.4%)	12 (6.5%)		3 (9.4%)	7 (10.9%)	-0.054
Invasion depth, n (%)				0.796
MP	174 (79.8%)	25 (78.1%)	149 (80.1%)		25 (78.1%)	53 (82.8%)	-0.113
MP-ex	44 (20.2%)	7 (21.9%)	37 (19.9%)		7 (21.9%)	11 (17.2%)	0.113

MTCC: mucosal traction-assisted clip closure; CCC: conventional clip closure; BMI: body mass index; MP: muscularis propria; MP-ex: muscularis propria with exophytic growth; IQR: interquartile range; PSM: propensity score matching; SMD: standardized mean difference.

Intraoperative and postoperative outcomes

Defect closure was successfully achieved in all patients in the MTCC group. However, according to the predefined definition (complete endoscopic closure without subsequent abdominal abscess or perforation), the technical closure success rate in the original cohort was 100% (32/32) in the MTCC group and 98.9% (184/186) in the CCC group. After PSM, the success rate remained 100% (32/32) in the MTCC group and was 98.4% (63/64) in the CCC group. In the original CCC group, both failures were attributed to a postoperative perforation that occurred on day 2 after the procedure and was successfully managed with conservative treatment (nasogastric decompression, intravenous antibiotics, and parenteral nutrition) without surgical intervention. As shown in Table 2 (After PSM column), the median defect diameter was comparable between the MTCC and CCC groups in the matched cohort (2.8 cm vs. 2.3 cm, P = 0.228). The MTCC group demonstrated superior closure efficiency: the median total procedure time was significantly shorter in the MTCC group (47.0 min vs. 53.0 min, P = 0.027), and the median defect closure time was markedly reduced (7.0 min vs. 17.0 min, P < 0.001). The median endoscopic resection time showed a trend towards being longer in the MTCC group but did not reach statistical significance after matching (40.0 min vs. 37.0 min, P = 0.063). Importantly, the MTCC group required significantly fewer titanium clips to achieve closure (median 8.0 vs. 12.0, P < 0.001). The difference in operative time between the MTCC and CCC groups is shown in Fig. 4A, and the difference in titanium clip usage is presented in Fig. 4B.

Table 2.

Comparison of intraoperative outcomes between the two groups before and after PSM.

Variables	Overall (n = 218)	Before PSM			After PSM
		MTCC group (n = 32)	CCC group (n = 186)	P-value	MTCC group (n = 32)	CCC group (n = 64)	P-value
Defect diameter, cm, median (IQR)	2.5 (2.0, 3.5)	2.8 (2.0, 3.5)	2.5 (1.5, 3.5)	0.329	2.8 (2.0, 3.5)	2.3 (1.5, 3.5)	0.228
Total procedure time, minutes, median (IQR)	52.0 (42.0, 64.0)	47.0 (40.0, 60.0)	52.0 (43.0, 65.0)	0.015	47.0 (40.0, 60.0)	53.0 (44.0, 65.0)	0.027*
Endoscopic resection time, minutes, median (IQR)	36.0 (26.0, 45.0)	40.0 (30.0, 53.0)	36.0 (24.0, 45.0)	0.027	40.0 (30.0, 53.0)	37.0 (26.0, 46.0)	0.063
Defect closure time, minutes, median (IQR)	17.0 (10.0, 25.0)	7.0 (6.0, 9.0)	19.0 (13.0, 25.0)	< 0.001	7.0 (6.0, 9.0)	17.0 (15.0, 25.0)	< 0.001*
Number of titanium clips, median (IQR)	12.0 (10.0, 14.0)	8.0 (7.0, 9.0)	12.0 (11.0, 14.0)	< 0.001	8.0 (7.0, 9.0)	12.0 (11.0, 14.0)	< 0.001*
Technical closure success rate, n (%)	216 (99.1%)	32 (100.0%)	184 (98.9%)	> 0.999	32 (100.0%)	63 (98.4%)	> 0.999

MTCC: mucosal traction-assisted clip closure; CCC: conventional clip closure; IQR: interquartile range; PSM: propensity score matching.

Fig. 4.

Comparison of intraoperative and postoperative outcomes between the mucosal traction-assisted clip closure (MTCC) and conventional clip closure (CCC) groups. (A) Operation time; (B) Number of clips used; (C) Postoperative fasting and hospital stay days; (D) Postoperative complications.

Postoperative outcomes in the matched cohort are summarized in Table 3 (After PSM column). There were no significant differences between the two groups in pathological findings or in the incidence of postoperative bleeding, perforation, or fever (all P > 0.05). However, the median postoperative fasting time remained significantly shorter in the MTCC group (2.0 days vs. 3.0 days, P = 0.001). The median postoperative hospital stay was similar between the groups (4.0 days in both groups, P = 0.184). Figure 4C displays the differences in postoperative fasting duration and hospital stay between the MTCC and CCC groups, whereas Fig. 4D presents the variation in postoperative complication rates between the groups.

Table 3.

Comparison of postoperative outcomes between the two groups before and after PSM.

Variables	Overall (n = 218)	Before PSM			After PSM
		MTCC group (n = 32)	CCC group (n = 186)	P-value	MTCC group (n = 32)	CCC group (n = 64)	P-value
Pathology, n (%)				0.185			0.083
GIST	169 (77.5%)	24 (75.0%)	145 (78.0%)		24 (75.0%)	45 (70.3%)
Leiomyoma	41 (18.8%)	5 (15.6%)	36 (19.4%)		5 (15.6%)	18 (28.1%)
Others	8 (3.7%)	3 (9.4%)	5 (2.7%)		3 (9.4%)	1 (1.6%)
Postoperative bleeding, n (%)				> 0.999			> 0.999
Yes	4 (1.8%)	0 (0.0%)	4 (2.2%)		0 (0.0%)	0 (0.0%)
No	214 (98.2%)	32 (100.0%)	182 (97.8%)		32 (100.0%)	64 (100.0%)
Postoperative perforation, n (%)				> 0.999			> 0.999
Yes	2 (0.9%)	0 (0.0%)	2 (1.1%)		0 (0.0%)	1 (1.6%)
No	216 (99.1%)	32 (100.0%)	184 (98.9%)		32 (100.0%)	63 (98.4%)
Postoperative fever, n (%)				0.755			> 0.999
Yes	23 (10.6%)	4 (12.5%)	19 (10.2%)		4 (12.5%)	7 (10.9%)
No	195 (89.4%)	28 (87.5%)	167 (89.8%)		28 (87.5%)	57 (89.1%)
Postoperative fasting, days, median (IQR)	3.0 (2.0, 3.0)	2.0 (2.0, 3.0)	3.0 (2.0, 3.0)	< 0.001	2.0 (2.0, 3.0)	3.0 (2.0, 3.0)	0.001
Postoperative hospital stay, days, median (IQR)	4.0 (3.0, 5.0)	4.0 (3.5, 4.0)	4.0 (3.0, 5.0)	0.392	4.0 (3.5, 4.0)	4.0 (3.0, 5.0)	0.184

MTCC: mucosal traction-assisted clip closure; CCC: conventional clip closure; IQR: interquartile range; “Others” includes glomus tumor and ectopic pancreas; PSM: propensity score matching.

Subgroup analyses

The advantages of MTCC were consistently observed when patients were stratified by defect diameter (Tables 4 and 5). In patients with defect diameter ≤ 2.5 cm, the MTCC group demonstrated a significantly shorter total procedure time (48.0 vs. 55.0 min, P = 0.038) and defect closure time (7.0 vs. 19.0 min, P < 0.001), while requiring fewer titanium clips (8.0 vs. 12.0, P < 0.001) compared to the CCC group, with no significant differences in postoperative complications. Similarly, in patients with defect diameter > 2.5 cm, the MTCC group exhibited a shorter defect closure time (7.0 vs. 18.0 min, P < 0.001), used fewer clips (8.5 vs. 13.0, P < 0.001), and had a reduced postoperative fasting time (2.0 vs. 3.0 days, P < 0.001), while maintaining comparable safety outcomes. However, the small subgroup sample sizes (n = 16 per MTCC subgroup) limited the statistical power for detecting differences in rare adverse events, and the comparable complication rates should be interpreted with caution. These results indicate that the efficacy of MTCC in improving closure efficiency is independent of defect size. For tumors located in the gastric fundus (Table 6), which represented the majority of cases (n = 152), the MTCC group (n = 22) maintained significant advantages over the CCC group (n = 130). These included a shorter total procedure time (46.0 min vs. 54.0 min, P = 0.036), a markedly reduced defect closure time (7.0 min vs. 19.0 min, P < 0.001), lower clip usage (8.5 vs. 12.0, P < 0.001), and a shorter postoperative fasting period (2.0 days vs. 3.0 days, P < 0.001). Rates of postoperative complications were not increased with the MTCC technique. However, the small subgroup sample sizes limited the statistical power for detecting differences in rare adverse events, and the comparable complication rates should be interpreted with caution. These subgroup findings are exploratory and hypothesis-generating rather than confirmatory.

Table 4.

Comparison of intraoperative and postoperative outcomes between the two groups in patients with defect diameter ≤ 2.5 cm.

Variables	MTCC group (n = 16)	CCC group (n = 108)	P-value
Total procedure time, minutes, median (IQR)	48.0 (37.0, 60.0)	55.0 (43.0, 65.0)	0.038
Endoscopic resection time, minutes, median (IQR)	40.0 (29.0, 53.0)	37.0 (24.0, 45.0)	0.202
Defect closure time, minutes, median (IQR)	7.0 (6.0, 9.0)	19.0 (14.0, 25.0)	< 0.001
Number of titanium clips, median (IQR)	8.0 (7.0, 9.0)	12.0 (11.0, 14.0)	< 0.001
Postoperative bleeding, n (%)			> 0.999
Yes	0 (0.0%)	3 (2.8%)
No	16 (100.0%)	105 (97.2%)
Postoperative perforation, n (%)			> 0.999
Yes	0 (0.0%)	1 (0.9%)
No	16 (100.0%)	107 (99.1%)
Postoperative fever, n (%)			> 0.999
Yes	1 (6.3%)	9 (8.3%)
No	15 (93.8%)	99 (91.7%)
Postoperative fasting, days, median (IQR)	2.5 (2.0, 3.0)	3.0 (2.0, 3.0)	0.054
Postoperative hospital stay, days, median (IQR)	4.0 (3.0, 4.0)	4.0 (3.0, 5.0)	0.202

MTCC: mucosal traction-assisted clip closure; CCC: conventional clip closure; IQR: interquartile range.

Table 5.

Comparison of intraoperative and postoperative outcomes between the two groups in patients with defect diameter > 2.5 cm.

Variables	MTCC group (n = 16)	CCC group (n = 78)	P-value
Total procedure time, minutes, median (IQR)	46.0 (40.0, 59.0)	51.0 (43.0, 64.0)	0.208
Endoscopic resection time, minutes, median (IQR)	40.0 (33.0, 52.0)	35.0 (24.0, 43.0)	0.049
Defect closure time, minutes, median (IQR)	7.0 (6.0, 9.0)	18.0 (12.0, 25.0)	< 0.001
Number of titanium clips, median (IQR)	8.5 (7.5, 9.0)	13.0 (11.0, 14.0)	< 0.001
Postoperative bleeding, n (%)			> 0.999
Yes	0 (0.0%)	1 (1.3%)
No	16 (100.0%)	77 (98.7%)
Postoperative perforation, n (%)			> 0.999
Yes	0 (0.0%)	1 (1.3%)
No	16 (100.0%)	77 (98.7%)
Postoperative fever, n (%)			0.690
Yes	3 (18.8%)	10 (12.8%)
No	13 (81.3%)	68 (87.2%)
Postoperative fasting, days, median (IQR)	2.0 (2.0, 2.0)	3.0 (2.0, 3.0)	< 0.001
Postoperative hospital stay, days, median (IQR)	4.0 (4.0, 4.0)	4.0 (3.0, 4.0)	0.861

MTCC: mucosal traction-assisted clip closure; CCC: conventional clip closure; IQR: interquartile range.

Table 6.

Comparison of intraoperative and postoperative outcomes between the two groups in patients with tumors located at the gastric fundus.

Variables	MTCC group (n = 22)	CCC group (n = 130)	P-value
Total procedure time, minutes, median (IQR)	46.0 (39.0, 60.0)	54.0 (43.0, 65.0)	0.036
Endoscopic resection time, minutes, median (IQR)	40.0 (30.0, 54.0)	37.0 (24.0, 45.0)	0.089
Defect closure time, minutes, median (IQR)	7.0 (5.0, 9.0)	19.0 (13.0, 25.0)	< 0.001
Number of titanium clips, median (IQR)	8.5 (8.0, 9.0)	12.0 (11.0, 14.0)	< 0.001
Postoperative bleeding, n (%)			> 0.999
Yes	0 (0.0%)	4 (3.1%)
No	22 (100.0%)	126 (96.9%)
Postoperative perforation, n (%)			> 0.999
Yes	0 (0.0%)	1 (0.8%)
No	22 (100.0%)	129 (99.2%)
Postoperative fever, n (%)			0.740
Yes	3 (13.6%)	16 (12.3%)
No	19 (86.4%)	114 (87.7%)
Postoperative fasting, days, median (IQR)	2.0 (2.0, 3.0)	3.0 (2.0, 3.0)	< 0.001
Postoperative hospital stay, days, median (IQR)	4.0 (4.0, 4.0)	4.0 (3.0, 5.0)	0.337

MTCC: mucosal traction-assisted clip closure; CCC: conventional clip closure; IQR: interquartile range.

Discussion

The findings of this multi-center retrospective study suggest that the novel MTCC technique significantly enhances the efficiency of defect management following EFTR for gastric SMTs. In this matched cohort, we observed no increase in the rates of postoperative bleeding, perforation, or fever compared to CCC. Unlike previous reports that focused on traction during resection or used additional devices (e.g., dental floss, snares), our MTCC technique uniquely targets the closure phase using only endoscopic clips and the patient’s own mucosal flap, and is evaluated in a propensity-matched multi-center cohort. The MTCC group achieved a 100% success rate for defect closure, a markedly shorter median closure time of 7.0 min, and required fewer titanium clips (median 8.0) compared to CCC. These procedural efficiencies translated into a clinically meaningful reduction in postoperative fasting time, suggesting faster wound sealing and recovery. Crucially, these benefits were observed across subgroups stratified by defect size and were particularly evident for tumors located in the challenging gastric fundus, indicating the broad applicability and robustness of the technique.

The core innovation of the MTCC technique lies in the synergistic combination of two complementary technical principles: maximal preservation of the mucosal flap and strategic application of traction, which together transform a challenging closure into a rapid, controlled procedure. First, maximal preservation of the mucosal flap is the foundational element. In conventional CCC, the mucosal edge is often circumferentially incised to facilitate tumor exposure, leaving limited tissue for clip anchorage. In contrast, MTCC deliberately preserves a broad-based flap by performing only a partial (approximately three-quarter) circumferential incision and meticulous dissection between the mucosa and tumor. This flap provides a robust, well-vascularized anchor without which effective traction would be impossible. Importantly, flap formation is integrated into the natural sequence of tumor exposure, not an extra step; thus, the median endoscopic resection time was only numerically longer (40.0 vs. 37.0 min, P = 0.063), representing a modest trade-off for dramatically faster closure (7.0 vs. 17.0 min, P < 0.001) and reduced clip usage. Second, strategic traction applied through this flap is equally indispensable. Once grasped, the clip applies traction to reduce tension at defect edges, convert a round or irregular defect into a near-linear configuration, and improve visualization for precise sequential clip placement.

To contextualize MTCC within the broader landscape of traction-assisted endoscopic closure, it is useful to distinguish two general conceptual approaches. The first category, device-assisted traction, relies on external tools such as dental floss, snares, through-the-scope graspers, or magnetic anchors to manipulate tissue; these methods often require additional equipment, dedicated device passage, or assistant coordination, which can add procedural complexity^15–17. The second category, intrinsic tissue-based traction, leverages the patient’s own preserved tissue—most commonly a mucosal flap—as an anchor, applying traction solely through standard endoscopic clips and scope manipulation without auxiliary devices. MTCC belongs firmly to this second category. A direct conceptual comparison between MTCC and the previously reported ITASC by Gong et al.¹⁵ is warranted, as both techniques share the core principle of using internal traction without external devices. However, key differences exist:¹ ITASC relies on residual mucosal tissue after resection, whereas MTCC deliberately preserves a maximally broad-based mucosal flap during the resection phase;² ITASC uses standard clips without emphasizing reopenable capability, while MTCC specifically employs reopenable titanium clips that allow repeated grasp-and-release adjustments to optimize the traction vector before final deployment;³ ITASC achieves closure by suspending the defect edges, whereas MTCC employs a stepwise sequential closure starting from the traction root, systematically converting a round or irregular defect into a linear configuration. These differences translate into practical advantages: MTCC requires only the same clips already used for closure, no additional device passage, and the flap provides a robust, reproducible anchor even in challenging locations such as the gastric fundus. Its unique features compared to previously reported intrinsic traction techniques include:¹ deliberate maximal preservation of a broad-based mucosal flap during the resection phase rather than relying on residual incidental tissue;² use of reopenable titanium clips that allow repeated grasp-and-release adjustments to optimize the traction vector before final deployment; and³ a stepwise sequential closure starting from the traction root, which systematically converts a round or irregular defect into a linear configuration. This combination minimizes device requirements (only standard clips and a conventional endoscope), reduces procedural complexity (no assistant-needed traction or device exchange), and maintains broad applicability across different defect sizes and anatomical locations, as demonstrated in our subgroup analyses. Thus, while MTCC shares the core principle of internal traction with other reported methods, its specific mechanism—preserved flap as anchor + reopenable clip-mediated vector control + root-to-periphery sequential closure—offers a distinct, low-device, and easily adoptable technical refinement for post-EFTR defect closure.

The principle of traction to improve endoscopic procedure efficiency is well-supported in the literature. For instance, in ESD for gastric neoplasms, the use of direction-selectable traction devices has been shown to significantly reduce procedure time and increase dissection speed compared to unidirectional methods²³. Similarly, magnetic anchor-guided techniques and magnetic hydrogel-assisted ESD have demonstrated accelerated dissection speeds in experimental and clinical settings by providing stable and adjustable retraction²⁴. Our results align with this paradigm, extending the benefits of traction-assisted surgery specifically to the critical step of full-thickness defect closure, which is often the most time-consuming and technically demanding part of EFTR.

The significant reduction in defect closure time from a median of 19.0 min with CCC to 7.0 min with MTCC represents a substantial advancement. This efficiency gain is noteworthy when contextualized within the broader landscape of EFTR procedure times reported in other studies. A large retrospective study on EFTR for upper GISTs reported an average wound closure time of 19.50 min, which is nearly identical to the CCC group in our study, underscoring the representativeness of our control cohort¹. Another technical analysis of EFTR for gastric SMTs noted that advanced closure methods, while sometimes prolonging the closure step initially, could lead to improved overall outcomes such as shorter fasting periods²⁵. In our study, the MTCC technique not only expedited closure but also contributed to a reduction in the total procedure time, despite a slightly longer endoscopic resection time in the MTCC group. This suggests that the time invested in creating a viable mucosal flap for traction is more than compensated for by the accelerated closure phase, leading to a net gain in overall operative efficiency. The reduction in clip usage with MTCC is another key finding. Using fewer clips to achieve secure closure has direct implications for procedural cost-effectiveness and may reduce the risk of clip-related complications, such as tissue ischemia or interference with subsequent imaging. The ability to achieve secure apposition with fewer mechanical devices points to the superior tissue alignment afforded by the traction, minimizing the need for compensatory bridging or overlapping clip placement often required when edges are under tension or poorly visualized.

Perhaps one of the most compelling clinical outcomes observed in this study is the significantly shorter postoperative fasting time in the MTCC group (2.0 days vs. 3.0 days). This metric is a direct surrogate for clinicians’ confidence in the integrity of the gastric wall closure and the patient’s subsequent recovery trajectory. A shorter fasting period enhances patient comfort, reduces the risk of catheter-related and metabolic complications associated with prolonged nil-by-mouth status, and may contribute to shorter hospital stays, although the latter was comparable between groups in our cohort. The mechanism behind this accelerated recovery likely stems from the more robust and tension-free closure facilitated by the combination of a well-preserved mucosal flap and effective traction. By reducing mechanical strain on the suture line and promoting better tissue apposition, this dual approach may create a more favorable environment for primary healing. However, it should be acknowledged that the decision to resume oral intake is inherently subjective and dependent on the treating physician’s clinical judgment. If clinicians perceived greater confidence in the MTCC closure quality, this may have biased toward earlier resumption of feeding. Future studies incorporating objective measures of closure integrity, such as early postoperative endoscopic or radiological assessment, would help validate this finding. This is conceptually supported by research on wound management after large gastric ESDs, where the application of human-derived fibrin sealant was found to reduce delayed bleeding and accelerate the healing process of artificial ulcers²⁶. While MTCC is a mechanical rather than a biomaterial solution, it shares the same goal of optimizing the wound environment for healing. The fact that complication rates (bleeding, perforation, fever) remained comparable between the MTCC and CCC groups, despite the faster return to oral intake, strongly reinforces the safety and reliability of the MTCC closure.

The subgroup analyses provide robust evidence for the generalizability of the MTCC technique. Its superiority in reducing closure time and clip usage was consistent for both small (≤ 2.5 cm) and large (2.5 cm) defects. Managing larger defects is particularly challenging, as they are associated with increased technical difficulty and longer operative times^1,9. The ability of MTCC to maintain its efficiency advantage in this subgroup is therefore clinically significant. For large defects, the preserved mucosal flap provides a proportionally larger tissue anchor, and traction can be applied with greater vector control to reduce tension across a wider span. Furthermore, the technique proved especially beneficial for tumors located in the gastric fundus, which constituted the majority of cases in this study. The fundus is a notoriously difficult location for endoscopic procedures due to its anatomical orientation, acute angles, and the tendency for instruments to lose triangulation. However, we recognize that closure can also be challenging in other anatomical sites, such as the gastric antrum, where a thicker muscular wall generates higher tissue tension, potentially impeding adequate tissue approximation during clip placement. Prior studies have highlighted the technical challenges of resecting SMTs in the cardia and fundus^10,27. In such locations, preserving a mucosal flap is technically more demanding, and applying effective traction requires careful vector selection. However, when both elements are successfully executed, the benefits are magnified: the flap provides a secure anchor in an otherwise unstable field, and traction pulls the defect into a more accessible position. The significant reductions in total procedure time, closure time, and clip usage observed in the fundus subgroup suggest that MTCC effectively mitigates these anatomical hurdles, and we hypothesize that similar benefits may extend to other high-tension locations, including the antrum. The combination of a preserved flap and strategic traction can help pull the defect into a more favorable position within the endoscopic field, overcoming spatial and tension-related constraints. This addresses a common limitation noted in other endoscopic techniques where location-specific challenges can impact outcomes²⁸.

When evaluating the observed adverse event profile, it is reassuring that MTCC was not associated with an increased rate of postoperative bleeding, perforation, or fever in this cohort. Nevertheless, the absence of a statistically significant difference does not establish comparative safety. The primary goal of any novel closure technique must be to maintain or improve upon the safety standards of existing methods, and this question will require larger prospective studies with adequate power to detect differences in rare adverse events. The incidence of adverse events in our CCC group is consistent with rates reported in other large series of EFTR and endoscopic resection for gastric SMTs^1,9,29. For example, a study on EFTR for extra-luminal gastric SMTs reported adverse events including delayed bleeding and perforation at low rates⁹, similar to our findings. Another multi-center study on device-assisted EFTR for gastroduodenal lesions also reported a low rate of severe adverse events³⁰. The comparable safety outcomes between MTCC and CCC, despite the more efficient closure, suggest that neither mucosal flap preservation nor traction application compromises tissue viability or introduces new risks. It is worth noting that the management of post-resection defects remains an area of active investigation, with some studies questioning the necessity of routine closure for all ESD defects regarding bleeding prevention³¹. However, for full-thickness defects created during EFTR, secure closure is unequivocally mandatory to prevent life-threatening peritonitis. Our study confirms that MTCC fulfills this requirement effectively and safely.

Several limitations of this study should be acknowledged. First, the retrospective, non-randomized design inherently carries the risk of selection bias. The substantial imbalance in group sizes (MTCC, n = 32; CCC, n = 186) reflects the phased introduction of the novel MTCC technique across participating centers, which may have introduced temporal bias and selection bias, as operators may have preferentially applied the new technique in select cases. To mitigate these concerns, we performed a 1:2 PSM analysis, which successfully balanced all measured baseline characteristics between the two groups. Importantly, the superior efficacy of MTCC—including significantly shorter defect closure time, reduced clip usage, and faster postoperative recovery—persisted in the matched cohort, strengthening the validity of our conclusions. However, PSM can only adjust for measured confounders; the possibility of residual confounding due to unmeasured or unknown variables (e.g., operator experience, intraoperative decision-making, subtle differences in defect morphology, or the learning curve associated with flap creation and traction application) cannot be entirely excluded. Second, multi-center heterogeneity also introduces variability. Differences in endoscopic equipment, electrosurgical settings, perioperative protocols (e.g., fasting and discharge criteria), and operator experience across the seven hospitals may have influenced outcomes. We did not perform center-stratified or mixed-effects analyses, nor did we formally assess inter-center comparability of key outcome measurements (e.g., closure time, defect diameter measurement), which represents a further analytical limitation. Future prospective studies should incorporate standardized protocols and center-stratified analyses to address this issue. Third, despite PSM, the sample size of the MTCC group remained relatively small (n = 32), which limits the statistical power to detect differences in rare adverse events such as postoperative bleeding or perforation. The comparable complication rates observed between groups should therefore be interpreted with caution, and larger prospective studies are needed to definitively establish the safety profile of MTCC. This is particularly true for the subgroup analyses, which were based on very small sample sizes and should therefore be viewed as exploratory. Finally, the follow-up data in this study primarily focused on short-term perioperative outcomes; long-term follow-up regarding recurrence is essential. Encouragingly, other studies on EFTR for gastric SMTs and GISTs have demonstrated excellent long-term oncological outcomes with minimal recurrence^1,9,32, and it is reasonable to hypothesize that MTCC, by enabling a secure R0-capable resection and closure, would not adversely affect these long-term results.

In conclusion, this multi-center study provides preliminary evidence that the MTCC technique is feasible and may improve the efficiency of managing gastric wall defects after EFTR. Its efficacy derives from the synergistic combination of two interdependent principles: maximal preservation of a robust mucosal flap during resection, and strategic application of traction using that flap to linearize the defect and facilitate sequential clip placement. This dual mechanism significantly improves procedural efficiency, reduces clip consumption, and accelerates postoperative recovery, while maintaining a safety profile comparable to conventional closure. While prospective randomized trials are warranted to confirm these findings and evaluate long-term outcomes, MTCC emerges as a valuable technical refinement in the field of EFTR. It exemplifies how the intelligent application of basic surgical principles, such as tissue preservation and traction-assisted manipulation, can yield substantial improvements in minimally invasive therapeutic endoscopy, ultimately aiming to enhance patient care through safer, faster, and more efficient procedures. Future research should focus on prospectively comparing MTCC against other advanced closure methods, including endoscopic suturing, and on developing standardized training modules to integrate this technique into the practice of advanced endoscopists.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Author contributions

Qian Zhang and Fengcheng Zang contributed equally to this work. Qian Zhang, Fengcheng Zang, Jie Xu, Yunfu Feng, Zhibing Wang, Chao Ma, Luojie Liu, and Qiushi Tian contributed to data collection and writing. Qian Zhang and Fengcheng Zang were responsible for statistical analysis. Luojie Liu and Qiushi Tian contributed to revising this dissertation. Xianjing Fan and Qiushi Tian managed this project. All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Data availability

Data availability can be requested from the corresponding author with a reasonable justification.

Declarations

Conflict of interest

Qian Zhang, Fengcheng Zang, Jie Xu, Yunfu Feng, Zhibing Wang, Chao Ma, Luojie Liu, Xianjing Fan, and Qiushi Tian disclosed no conflict of interest for this article.

Financial support

This study was supported by the Changshu Science and Technology Program (CY202339), the Suzhou Youth Science and Technology Project for the Advancement of Science, Education, and Health (KJXW2023067), Specialized Project for Clinical Key Disease Diagnosis and Treatment Technology in Suzhou (LCZX202334), and the Suzhou 23rd Batch of Science and Technology Development Plan (Clinical Trial Institution Capability Enhancement) Project (SLT 2023006).