Comparison between interrupted closure technique and traditional closure technique in endoscopic full-thickness resection for treating gastric subepithelial lesions

Abstract

BACKGROUND

Gastric subepithelial lesions (SELs) are elevated lesions originating from the muscularis mucosa, submucosa, or muscularis propria, and may also include extraluminal lesions. For small SELs (less than 5 cm), complete endoscopic excision is the preferred treatment. Endoscopic full-thickness resection (EFTR) has proven to be an effective approach.

AIM

To evaluate the efficacy of the interrupted closure technique compared to the traditional closure technique in EFTR for gastric SELs.

METHODS

This single-center, prospective, randomized controlled trial was conducted at a tertiary hospital from September 2023 to September 2024. A total of 90 patients who underwent EFTR for gastric SELs were randomly allocated to either the interrupted closure group (n = 44) or the traditional closure group (n = 46).

RESULTS

All patients had complete resection and wound closure without any severe postoperative complications. The incidence of intraoperative gas-related complications was significantly lower in the interrupted closure group than in the traditional closure group (2.27% vs 26.09%, P = 0.001), demonstrating interrupted closure technique can reduce the incidence of gas-related issues. Statistical analysis revealed that the incidence of postoperative infection was significantly lower in the experimental group than in the control group (15.91% vs 41.30%, P = 0.008). Additionally, the median duration of antibiotic use was lower in the experimental group (3.5 days vs 5 days, P = 0.013). Abdominal pain levels on postoperative days 1 and 4 were also lower in the experimental group compared to the control group (P < 0.001).

CONCLUSION

The interrupted closure technique in EFTR for treating gastric SELs is safe and effective, reducing the incidence of intraoperative gas complications and postoperative infections.

Key Words: Endoscopic full-thickness resection; Interrupted closure; Gastric subepithelial lesion; Gas complication; Postoperative infection

Core Tip: The interrupted closure technique involves performing a two-thirds circumferential full-thickness incision around the diseased gastric wall, followed by the immediate closure of either the proximal or distal end of the defect using metallic clips. With minimal necessary exposure, lesion dissection and defect closure are performed alternately until complete tumor resection and wound closure are achieved. Our study findings demonstrate that this technique is an effective approach for the treatment of gastric subepithelial lesions, significantly reducing the incidence of intraoperative gas-related complications and postoperative infections compared to traditional endoscopic full-thickness resection.

INTRODUCTION

Gastric subepithelial lesions (SELs) are elevated lesions originating from the muscularis mucosa, submucosa, or muscularis propria, and may also include extraluminal lesions[1,2]. The prevalence of SELs in patients undergoing gastroscopy is approximately 0.3% to 0.76%[3,4]. Common types of SELs including gastrointestinal stromal tumors, leiomyomas, ectopic pancreas, and lipomas. Although most SELs are benign, up to 13% may have malignant potential[5]. For SELs with malignant potential, such as gastrointestinal stromal tumors, guidelines recommend resection for lesions with a diameter of 2 cm or greater[6-8]. Benign lesions, should also be treated if they cause symptoms, or if the patient expresses a strong desire for treatment.

Complete endoscopic excision is the preferred treatment for small SELs (less than 5 cm)[9,10]. Endoscopic submucosal dissection (ESD), a common resection method for gastrointestinal lesions, enables complete resection of mucosal and submucosal lesions. However, SELs often originate from the muscularis propria layer, the perforation rate associated with ESD is significantly elevated (approximately 4%-13%)[11-13], which somewhat limits its applicability for SELs. Endoscopic full-thickness resection (EFTR) extends ESD by resecting the lesion along with the involved gastrointestinal wall, which has gradually emerged as a treatment option for refractory mucosal and submucosal lesions[14-16]. However, EFTR may lead to iatrogenic perforation, with the passage of gas and gastric contents into the peritoneal cavity potentially causing complications[17]. Consequently, strategies to reduce gas and fluid exudation, successfully close the wound, and minimize complications are critical considerations in EFTR.

Traditional EFTR techniques facilitate the complete excision of lesions; however, they can result in substantial gastric wall defects, allowing significant intragastric gas and fluid to enter the intraperitoneal space, making it difficult to ensure a good surgical field and increasing the risk of postoperative infection. To optimize the traditional EFTR, a novel technique, EFTR interrupted closure technique was developed, in this technique, after performing a full-thickness incision around the lesion's edge, proximal or distal of the wound is immediately closed with metallic clips, followed by alternating dissection of the lesion and wound closure. Retrospective studies[18,19] have demonstrated the safety and clinical utility of the EFTR interrupted closure technique, based on this, we conducted a prospective study to compare the efficacy of the EFTR interrupted closure technique with the traditional closure technique in the treatment of gastric SELs.

MATERIALS AND METHODS

Study design

This single-center, prospective, randomized controlled trial was designed to assess the efficacy of interrupted closure vs traditional closure in EFTR for gastric SELs. The study population consisted of patients who underwent EFTR for gastric SELs in the Department of Endoscopy at Northern Theater General Hospital from September 2023 to September 2024. The study was approved by the ethics committee and registered on ClinicalTrials.gov (ChiCTR2400086500). Informed consent was obtained from all patients.

Inclusion criteria: (1) Age between 18 and 80 years; (2) SEL confirmed by endoscopic ultrasound; (3) No definitive metastasis detected on imaging; and (4) Maximum transverse diameter of the lesion ≤ 5.0 cm.

Exclusion criteria: (1) Submucosal lesion surface ulceration; (2) Multiple EFTR treatments were performed this time; (3) Allergy to lidocaine mucilage, pronase, and simethicone; (4) Pregnancy or lactation; (5) Known bleeding and coagulation disorders or use of anticoagulants or antiplatelet drugs cannot be discontinued; (6) Patients with other systemic malignancies; (7) Patients with psychoneurotic diseases, severe cardiopulmonary diseases, liver and kidney diseases, immune and blood system diseases, and electrolyte imbalance; and (8) American Society of Anesthesiologists classification ≥ IV before surgery.

Endoscopic equipment and accessories

Standard single-accessory-channel endoscopy (GIF-Q260J, GIF-H290T, Olympus) equipped with a transparent cap (D-201-11804; Olympus) and a high-frequency generator (VIO 200D; ERBE) and CO₂ insufflator were used in these procedures. Other equipment included MFK knife [AMH-EK-O-2.4X1800(4)-N, Anrei], injection needle (NM-4 L-1, Olympus), hot biopsy forceps (FD-410 LR; Olympus), metal hemostatic clip (ROCC-D-26-195, Micro-Tech; AG-51044-1950-090-16, AGS MEDTECH), and nylon rope coil (Loop-30,loop-40; LeCamp™Endoloops).

Surgical procedure

Experienced endoscopists, each having performed at least 100 EFTR procedures, conducted all operations. Selective submucosal injection of succinylated gelatin containing methylene blue was administered, and an MFK knife was employed to incise the diseased mucosal layer for adequate tumor exposure, the serosal layer of the diseased gastric wall was incised for "artificial perforation", traction devices were selectively utilized during surgery.

Experimental group (interrupted closure group): After performing a two-thirds circumferential full-thickness incision of the diseased gastric wall, the metal clips were performed at the proximal or distal to the wound. With minimal necessary exposure, the dissection and closure procedure were repeated until the tumor completely resected, then the defect was closed (Figure 1).

Figure 1 Resection of gastric submucosal lesions by interrupted closure in endoscopic full-thickness resection. A: The gastric subepithelial lesion in the gastric body was initially detected by endoscopy; B: A full-thickness incision around two-thirds circumferential of the lesion; C: The metallic clips closure were performed at the proximal of the wound surface; D: The wound was completely sutured with metal clips.

Control group (traditional closure group): Following the complete resection of the lesion, the wound was closed. Smaller wounds were closed with metal clips, larger wounds were managed using purse-string closures (Figure 2).

Figure 2 Resection of gastric submucosal lesions by traditional closure in endoscopic full-thickness resection. A: The gastric subepithelial lesion in the gastric body was initially detected by endoscopy; B: The full-thickness incision around the lesion was performed; C: The gastric wall defect after the lesion resection; D: The wound was completely sutured with metallic clips.

In cases of severe pneumoperitoneum during surgery, a syringe was used for puncture at McBurney's point. A nasogastric tube was routinely placed in the stomach.

Postoperative treatment and follow-up

Patients were closely monitored postoperatively, and abdominal computed tomography and endoscopic examination were performed promptly for patients suspected of having postoperative perforation or bleeding. Proton pump inhibitors were routinely administered postoperatively, while antibiotics were given only if infection appeared. Patients fasted for 3 to 5 days based on their condition and were started on a liquid diet after nasogastric tube removal. Discharge was allowed once they tolerated oral intake without abdominal pain, hematemesis, or other symptoms.

Three follow-up visits were scheduled for this study: on postoperative days 1, 4, and 28, to assess the occurrence of postoperative complications. On postoperative days 1 and 4, the physician evaluated EFTR-related complications through physical examination. On postoperative day 28, delayed complications were identified either through telephone follow-up or by consulting medical records.

Outcome measures

Primary Outcome Measures: Incidence of postoperative complications and success rate of wound closure.

Secondary Outcome Measures: Incidence of intraoperative gas-related complications, incidence of intraoperative bleeding, complete resection rate, operation time, hospital stay, and associated costs.

Definitions

Gas complications: The patient developed subcutaneous emphysema (manifested as emphysema on the face, neck, chest wall, and scrotum), pneumomediastinum (gastroscopy showed epiglottic swelling), severe pneumothorax and pneumoperitoneum. In this study, the occurrence of severe pneumoperitoneum requiring puncture and deflation via McBurney's point during surgery was included in the statistical analysis[1].

Intraoperative bleeding: Defined as significant active bleeding during surgery that is difficult to control endoscopically, requiring interruption of the procedure and/or blood transfusion. Minor bleeding controlled by endoscopic was not included in the assessment of intraoperative bleeding.

Severe postoperative complications: Delayed postoperative bleeding and delayed perforation.

Minor postoperative complications: Postoperative abdominal pain, fever, and infection.

Postoperative infection: Defined as the occurrence of fever (temperature > 37.3 °C) several hours to 3 days after EFTR, accompanied by localized abdominal pain [numerical rating scale (NRS) ≥ 3] and elevated inflammatory markers, with imaging excluding the possibility of obvious perforation.

Zero to ten NRS: Also known as a numerical analog scale, it allows patients to describe pain intensity using numbers from 0 to 10, with 0 representing no pain and 10 representing severe pain. NRS ≥ 3 is defined as abdominal pain.

Randomization and statistical calculations

After screening patients based on the inclusion and exclusion criteria, subjects were numbered using Microsoft Excel according to their order of enrollment, and randomized numbers were assigned. They were then allocated randomly to either the experimental or control group.

Previous literature suggests that the incidence rate of minor complications after conventional EFTR was 56.1%[20], 26.3% with interrupted closure technique[21], and 1:1 in the experimental group and the control group. A one-sided test assuming 5% significance level (α = 0.05) and 80% power (β = 0.2) was used. The required sample size was calculated as n₁ =n₂ = 39 per group using PASS 15.0; we ultimately determined a sample size of 43 per group, accounting for a 10% dropout rate since the study outcome was a short-term endpoint.

Statistical analysis

Statistical analysis was performed using SPSS 27.0, continuous variables were expressed as mean ± SD for normally distributed data and median (Q25, Q75) for non-normally distributed data, with comparisons performed using the independent sample t-test and Wilcoxon rank-sum test, respectively. Categorical data were analyzed using frequency or percentage, and enumeration data were compared using the χ² test or Fisher exact test for analysis. P < 0.05 was considered statistically significant. Logistic regression analysis was employed to evaluate the risk factors for complications in patients undergoing EFTR. Only variables with P < 0.05 in the univariable logistic regression model were included in multivariable analysis.

RESULTS

Baseline characteristics of patients and resected lesions

A total of 92 patients participated in this study between September 2023 and September 2024. Of these, 2 patients were excluded due to lesion diameter exceeding 5 cm, and 90 patients were finally analyzed. Patients were randomly divided into an experimental group (interrupted closure group, n = 44) and a control group (traditional closure group, n = 46).

There were no differences in the baseline characteristics of the patients between the two groups. Statistical analysis of lesion-related data revealed that the maximum lesion diameters were comparable between the two groups, with median lesion diameters of 29.5 mm. The majority of lesions in the experimental group were found in the body of the stomach (52.27%), whereas in the control group were located in the fundus of the stomach (56.52%). In both groups, lesions predominantly originated from the muscularis propria of the stomach, exhibiting both intraluminal and extraluminal growth patterns. Postoperative pathological analysis showed that the most common pathological types in both groups were stromal tumors. Detailed information is provided in Table 1.

Table 1 Baseline characteristics of patients and resected lesions, n (%).

	Experimental group (n = 44)	Control group (n = 46)	P value
Age [M, (Q1, Q3), years]	56.5 (49.0, 64.5)	58.5 (49.0, 62.0)	0.994
Sex			0.411
Male	21 (47.73)	18 (39.13)
Female	23 (52.27)	28 (60.87)
Maximum lesion diameter [M, (Q1, Q3), mm]	29.5 (20.0, 39.0)	29.5 (20.0, 48.0)	0.474
Lesion location			0.296
Fundus of stomach	18 (40.90)	26 (56.52)
Gastric body	23 (52.27)	18 (39.13)
Antrum of stomach	3 (6.82)	2 (4.35)
Level of origin of lesion			0.072
Not reaching muscularis propria	3 (6.82)	0 (0)
Muscularis propria	41 (93.18)	46 (100)
Growth pattern			0.875
Intraluminal growth	3 (6.82)	3 (6.52)
Extraluminal growth	2 (4.54)	1 (2.17)
Intraluminal and extraluminal growth	39 (88.64)	42 (91.30)
Pathological type			0.634
Stromal tumor	25 (56.82)	33 (71.74)
Leiomyoma	10 (22.73)	6 (13.04)
Ectopia pancreas	2 (4.54)	1 (2.17)
Neurofibroma	3 (6.82)	2 (4.35)
Other	3 (6.82)	4 (8.70)

M: Median.

Intraoperative measures analysis

In this study, all patients achieved complete excision of the lesion and successful wound closure. About gas complications, one patient in the experimental group required puncture at McBurney's point due to severe pneumoperitoneum, the gas-related complication rate was 2.27%, in contrast, in control group was 26.09%, with a total of 12 patients affected (P = 0.001). In terms of operative time, the median operative time was same in both groups at 29.5 minutes. There was no requirement for traction techniques in this study, 8 patients in the experimental group and 10 patients in the control group employed traction technique. Regarding the complete lesion removal rate, one patient from each group required gastric segmentation due to large lesions. Detailed information is provided in Table 2.

Table 2 Comparison of intraoperative characteristics between the two groups, n (%).

	Experimental group (n = 44)	Control group (n = 46)	P value
Complete wound closure			> 0.999
Yes	44 (100)	46 (100)
No	0 (0)	0 (0)
En bloc resection			> 0.999
Yes	44 (100)	46 (100)
No	0 (0)	0 (0)
Intraoperative bleeding			> 0.999
Yes	0 (0)	0 (0)
No	44 (100)	46 (100)
Gas complication
Subcutaneous emphysema	0 (0)	0 (0)	> 0.999
Pneumomediastinum	0 (0)	0 (0)	> 0.999
Severe pneumothorax	0 (0)	0 (0)	> 0.999
Severe pneumoperitoneum	1 (2.27)	12 (26.09)	0.001
Operative time [M, (Q1, Q3), minute]	29.5 (20.0, 39.0)	29.5 (20.0, 48.0)	0.474
Intraoperative traction			0.673
Yes	8 (18.18)	10 (21.74)
No	36 (81.82)	36 (78.26)
Complete lesion removal			0.975
Yes	43 (97.73)	45 (97.83)
No	1 (2.27)	1 (2.17)

M: Median.

Analysis of postoperative indicators

The postoperative laboratory-related examination results showed that the postoperative C-reactive protein, white blood cell count, and neutrophil percentage in the experimental group were slightly lower than those in the control group, however, the differences had no statistical significance (P > 0.05). The median postoperative maximum body temperature was 36.8 °C in both groups. Regarding the degree of postoperative abdominal pain, the NRS scores at 1 and 4 days after the operation were lower in the experimental group compared to the control group. No cases of delayed complications were reported in either group. The incidence of postoperative infection was 15.91% in the experimental group and 41.30% in the control group (P = 0.008). In terms of antibiotic use, the duration was 3.5 days (1.3, 5) in the experimental group compared to 5 days (3.8, 6) in the control group (P = 0.013). The median number of fasting days was similar in both groups, at 5 days, with no statistically significant difference. Additionally, there was no statistical difference in postoperative hospital stays [6 (5, 7) days vs 6 (5, 7) days, P = 0.314] and total hospital costs [19843.3 (18361.6, 22260.5) China yuan (CNY) vs 20862.5 (19475.0, 24398.3) CNY, P = 0.075] between the experimental group and the control group. Detailed information is provided in Table 3.

Table 3 Comparison of postoperative characteristics between the two groups, n (%).

	Experimental group (n = 44)	Control group (n = 46)	P value
Postoperative CRP [M, (Q1, Q3), mg/L]	17.98 (7.3, 34.2)	30.8 (16.1, 47.1)	0.066
Postoperative white blood cell count [M, (Q1, Q3), 109/L]	8.9 (7.9, 11.2)	10.6 (9.0, 13.0)	0.069
Postoperative neutrophil percentage [M, (Q1, Q3), %]	81.7 (75.7, 84.9)	82.4 (76.7, 84.7)	0.577
Postoperative maximum body temperature [M, (Q1, Q3), °C]	36.8 (36.7, 36.9)	36.8 (36.8, 37.5)	0.149
NRS 1 day postoperative [M, (Q1, Q3), points]	3 (2, 3.6)	4 (3, 5)	< 0.001
NRS 4 days postoperative [M, (Q1, Q3), points]	1 (1, 2.0)	2 (1, 2.3)	< 0.001
Delayed postoperative bleeding			> 0.999
Yes	0 (0)	0 (0)
No	44 (100)	46 (100)
Delayed postoperative perforation			> 0.999
Yes	0 (0)	0 (0)
No	44 (100)	46 (100)
Postoperative infection			0.008
Yes	7 (15.91)	19 (41.30)
No	37 (84.09)	27 (58.70)
The days of use antibiotic [M, (Q1, Q3), day]	3.5 (1.3, 5)	5 (3.8, 6)	0.013
Postoperative hospital stays [M, (Q1, Q3), day]	6 (5, 7)	6 (5, 7)	0.314
Total hospitalization costs [M, (Q1, Q3), CNY]	19843.3 (18361.6, 22260.5)	20862.5 (19475.0, 24398.3)	0.075

M: Median; CNY: China yuan; CRP: C-reactive protein; NRS: Numerical rating scale.

DISCUSSION

EFTR is appropriate for lesions that originate from deeper layers and exhibit significant fibrosis. Multiple clinical studies have demonstrated the safety and efficacy of EFTR for treating gastrointestinal submucosal lesions[14-16]. In this study, all patients who underwent EFTR achieved complete resection of the lesions.

A critical factor of EFTR is the complete closure of the defect. To achieve this objective, researchers have endeavored to develop suitable suturing techniques and devices for EFTR wound closure, including Over-the-scope clip[22,23], OverStitch[24], Full-thickness resection device devices[25,26], and X-Tack suturing systems[27]. However, the excessive costs and complex procedures associated limit their clinical application. Suturing with metal clips is a straightforward and widely used in clinical practice. In this study, after incising two-thirds of the gastric wall surrounding the lesion, metal clips were immediately applied to proximal or distal end of the wound, dissection and closure procedure were repeated until the tumor completely resected. The interrupted closure technique alleviates wound tension and prevents mucosal edge edema associated with prolonged surgery, thereby facilitating easier and more efficient suturing. It also prevents the tumor from falling into the abdominal. In this study, all patients in the experimental group achieved complete wound closure without delayed complications. Proper management of gastric wall vessels is crucial prior to using the interrupted closure technique, as residual metallic clips could hinder subsequent hemostasis in case of bleeding.

Traditional EFTR involves resecting the entire lesion and subsequently closing the defect, however, due to prolonged perforation, this approach may lead to insufficient gastric cavity distension, compromised intraoperative visibility, and an increased incidence of gas-related complications[28,29]. In contrast, the interrupted closure technique performs suturing during lesion resection, reducing the defect area, effectively decreasing intraoperative gas leakage into the abdominal, maintaining better visibility, and decreasing the incidence of gas complications[19]. In this study, the gas complications observed were limited to pneumoperitoneum. And only one patient in the experimental group experienced pneumoperitoneum, compared to 12 in the control group (P = 0.001), multivariate logistic analysis indicated that the interrupted closure technique served as a protective factor against gas complications (Supplementary materials). Notably, the lesions studied were larger in diameter compared to previous literature. In our clinical practice, smaller lesions derive limited benefit from the new technique. When lesions are small, even if they lead to active perforation, the perforation duration is short, the metallic clips placed in advance may hinder subsequent dissection. Therefore, this technique may be more suitable for larger lesions. Although EFTR has been reported for lesions > 5 cm, we limit its use to those < 5 cm based on guidelines[1] and clinical experience.

Previous studies have indicated a high possibility of postoperative infection following EFTR, attributed to several factors: (1) EFTR targets lesions originating from deeper layers, where thermal damage from intraoperative electrocautery directly affecting the muscularis mucosa, muscularis propria, and serosa[30]; and (2) Iatrogenic perforation results in communication between the abdominal and gastric cavities, increasing the risk of abdominal infection. Consequently, exploring methods to reduce the incidence of postoperative infection after EFTR is significant. In this study, the postoperative infection rate in control group was 41.30%, while the experimental group was only 15.91%. These results indicate that the interrupted closure technique can effectively reduce the incidence of postoperative infection (P = 0.008). Additionally, in multivariable analysis, interrupted closure serves as a protective factor against postoperative infection (Supplementary materials). Although the interrupted closure technique in EFTR can reduce postoperative infection rates, all patients remained hospitalized for 3-5 days postoperatively for close monitoring and to ensure their safety.

The use of traction techniques during lesion resection remains a prominent topic in endoscopic research. The application of traction during EFTR offers several advantages: (1) It increases cutting tension, enhances cutting efficiency, and shortens operative time[31]; (2) It provides a clear visual field allowing for lesion dissection under direct vision and reducing vascular injury[32]; (3) It pulls the lesion into the gastric lumen, reducing the risk of mucosal incision damage to abdominal organs[33]; and (4) It prevents the lesion from falling into the abdominal cavity, reducing the possibility of intraperitoneal spread. In this study, a total of 18 patients underwent intraoperative traction. Among these, one patient utilized snare traction technique, while the others used dental floss technique. Notably, the median operative time was significantly shorter in the traction group compared to the non-traction group (28.0 minutes vs 36.5 minutes, P = 0.019), indicating the effectiveness of selective traction in optimizing surgical efficiency.

This study has several limitations: (1) It is a single-center study; (2) NRS scoring for abdominal pain is subjective, and individual variability among patients may exist; (3) The study outcomes were short-term, with no long-term follow-up was conducted; and (4) The classification of lesion locations was imprecise, and no efficacy assessment was performed for each gastric wall segment.

Not all locations may benefit from the interrupted closure technique, particularly the lesser curvature of the upper gastric body. First, lesions in this area can be challenging to dissect from the anal side; initiating dissection from the oral side with partial wound closure may complicate subsequent operations. Second, the omentum covering the gastric wall somewhat reduces leakage of gastric contents[34], thereby decreasing the risk of severe pneumoperitoneum and postoperative infection. Further research is needed to evaluate the appropriateness of the interrupted closure technique for this specific location.

CONCLUSION

The interrupted closure technique is a safe and effective approach in EFTR for gastric SELs, reducing the incidence of intraoperative gas complications and postoperative infections.