Preset pursestring procedure: Innovative auxiliary method of endoscopic full-thickness resection in small gastric stromal tumors

Abstract

Background and study aims

Endoscopic full-thickness resection (EFR) of gastric stromal tumors carries a high peritonitis risk. Our case report implied the preset pursestring procedure (PPP) could boost EFR speed and safety, but supporting clinical evidence is lacking, prompting this clinical evaluation of PPP in gastric EFR.

Patients and methods

This retrospective cohort study analyzed adults with small gastric stromal tumors who underwent EFR (n = 31) or PPP-assisted EFR (PPP-EFR, n = 28) between August 2022 and August 2023. Tumor characteristics, adverse events, postoperative efficacy, economic outcomes, and follow-up data were compared between groups.

Results

R0 resection rates were comparable (PPP-EFR: 92.8%; EFR: 90.0%). PPP-EFR had shorter closure time (3.5 vs. 19.5 min, P = 0.001). PPP-EFR had lower incidence of postoperative fever (10.7% vs. 32.3%, P = 0.040), computed tomography-confirmed peritonitis (7.1% vs. 29.0%, P = 0.031), and elevated inflammatory markers ( P ≤ 0.003), plus shorter fasting time ( P = 0.038). Operative time, hospital stay, and cost did not differ; no patients had recurrence or metastasis on follow-up.

Conclusions

PPP optimizes EFR closure, reduces inflammatory and abdominal complications, and improves postoperative recovery for small gastric stromal tumors.

Introduction

Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal tumors of the digestive tract, with gastric GISTs accounting for 60% of all GISTs ¹ . GISTs smaller than 2 cm are defined as small gastrointestinal stromal tumors with low risk of malignancy ² . Small GISTs are usually asymptomatic but have a high incidence and potential malignancy ³ . Currently, treatment of small gastric GISTs remains controversial, and regular gastroscopic follow-up is generally recommended. However, considering the cost of long-term monitoring and risks of tumor enlargement and metastasis, some experts suggest resection once small GISTs are detected endoscopically ⁴ . In the past, surgical resection was the mainstay of GIST treatment, but in recent years, endoscopic therapy has been increasingly applied to gastric GISTs ≤ 3 cm. Compared with surgical and laparoscopic procedures, endoscopic therapy has advantages such as minimal invasiveness, fewer adverse events (AEs), faster postoperative recovery, and lower costs ⁵ . For patients with reduced physical function and reduced immunity, minimally invasive endoscopic surgery is undoubtedly a better option.

Because most gastric GISTs originate from the muscularis propria, endoscopic full-thickness resection (EFR) is more suitable for their treatment. However, EFR requires active creation of a gastric perforation to completely resect the tumor. After perforation, poor visibility due to difficulty in insufflation and time-consuming, labor-intensive wound closure often lead to postoperative AEs such as peritonitis and even systemic inflammatory response, increasing patient suffering. This is more dangerous for patients with weak immune systems. Therefore, rapid closure of the perforation is particularly important. We previously used the preset pursestring procedure (PPP) combined with endoscopic submucosal excavation (ESE) to resect a cecal submucosal tumor, achieving rapid intraoperative wound closure and avoiding abdominal infection ⁶ . Thus, we hypothesized that PPP can reduce AEs in EFR for small gastric GISTs. This study aimed to verify this hypothesis.

Patients and methods

Study subjects

A retrospective analysis was performed on 59 patients with gastric GISTs who underwent EFR or PPP-EFR at Shenzhen Third People’s Hospital from August 2022 to August 2023. EFR was performed from August 2022 to February 2023 and PPP-EFR was performed from March 2022 to August 2023; there was no overlap between the two periods. Inclusion criteria were age ≥ 18 years; postoperative histopathological confirmation of gastric GIST; surgical procedure: EFR or PPP-EFR; maximum tumor diameter < 2.0 cm; and follow-up duration > 24 months. Exclusion criteria were presence of distant metastasis; presence of other malignant tumors or multiple organ failure; severe mental disorders or uncooperative patients; coagulation disorders or hemorrhagic diseases; and incomplete clinical data.

Preoperative preparation

All patients underwent comprehensive preoperative evaluation, including medical history review, laboratory tests, contrast-enhanced computed tomography (CT), electrocardiogram (ECG), gastroscopy, and endoscopic ultrasound (EUS). Tumor size, growth pattern, origin layer, location, internal echo, and presence of distant metastasis were evaluated preoperatively. Patients undergoing endoscopic resection were instructed to avoid aspirin or other anticoagulants for 1 week before surgery. All patients signed informed consent forms and were informed of the surgical risks. Direct deep biopsy or EUS-FNA was performed for histological confirmation of GIST for each case preoperatively. All EFR and PPP-EFR procedures were performed by a single operator with more than 3 years of experience in EFR to reduce interoperator variation.

Treatment procedures

Parameter settings of electrosurgical generator (RRBE, VIO 300D)

In marking and hemostasis phase, forced coagulation (Effect setting = 2, Power = 40W) was used. In dissecting phase, incision was performed in ENDO-CUT I mode (Effect = 2, Duration = 3, Interval = 3).

EFR procedure

All surgeries were performed under endotracheal intubation, general anesthesia, and carbon dioxide insufflator assistance, by the same experienced senior physician (with > 100 EFR cases completed). The lesion area was marked using a mucosal incision knife (EK-425D; Anrei, Hangzhou). A mixture of normal saline and methylene blue was injected into the submucosa of the lesion area. Then the mucosal incision knife was used to make a circumferential incision around the lesion, followed by continued submucosal dissection until the lesion was completely separated. The tumor, including the muscularis propria and serosa, was resected and the gastric wall defect was closed with titanium clips (ROCC-D-26–195; Micro-Tech Nanjing). Dental floss external traction is used when tumor dissection is difficult.

PPP-EFR procedure

Download video file ^{(45.2MB, mp4)}

Preset pursestring procedure.

Video 1

No mark was need when the submucosal tumor was found ( Fig. 1 a ). An endoloop (HX-400U-30; Olympus) was delivered into the stomach through the working channel of a single-channel endoscope. After the endoloop was positioned around the target lesion, five to eight titanium clips (ROCC-D-26–195; Micro-Tech Nanjing) were applied to anchor the endoloop to the gastric wall surrounding the lesion, approximately 1.0 cm from the tumor edge ( Fig. 1 b , Video 1 ). Submucosal injection of fluid cushion was done within the pursestring circle. Then tumor dissection was performed with an electrosurgical knife after mucosal incision ( Fig. 1 c ). The tumor was eventually resected and removed from the stomach ( Fig. 1 d ). The wound after PPP-EFR was check carefully ( Fig. 1 e ), and if there was no bleeding, wound closure by tightening the preset pursestring was performed using the endoloop hook ( Fig. 1 f ). If there was bleeding, thermal forceps hemostasis was performed before wound closure. In addition, dental floss external traction was used when tumor dissection was difficult. Pre-placed clipping may cause interference with resection, and to ensure smooth resection, we provide a summary of technical tips: 1) Place the clips approximately 1.0 cm from the tumor edge to avoid overlapping with the resection area; and n2) During the dissection process, press the titanium clips with a transparent cap — in a manner analogous to holding the mucosal flap in ESD — can eliminate interference caused by the titanium clips. In addition, the following tips can be employed to ensure quality of clip closure when fixing endoloop with titanium clips: 1) Relax the gastric mucosa by suction to grasp a larger area of the mucosa; and 2) After opening the titanium clips, press them firmly against the mucosa appropriately to achieve a deeper clamping effect.

Fig. 1.

Operative procedure of PPP-EFR. a Submucosal tumor of the gastric body, approximately 12 mm in diameter. b Preset pursestring placement using endoloop and titanium clips. c Tumor dissection with an electrosurgical knife after injection of fluid cushion and mucosal incision. d Completely resected tumor tissue. e The wound after PPP-EFR showed that the serosal layer was preserved. f Rapid wound closure by tightening the endoloop.

Pathological evaluation

Specimens were fixed in 40% formaldehyde, embedded in paraffin, sectioned, and stained with hematoxylin-eosin. Tumor size, surgical margins, and mitotic count were recorded. Risk stratification was performed according to the 2008 National Institutes of Health (NIH) classification, based on mitotic count per 50 high-power fields, maximum pathological diameter, and lesion location. R0 resection was defined as en bloc resection (complete tumor removal with a single specimen and no endoscopically visible residual tumor) with negative margins on all sides. Immunohistochemical staining was used to distinguish GISTs from other mesenchymal tumors.

Evaluation of adverse events and outcomes

Operative time was defined as duration from endoscope insertion to withdrawal.

Resection time was defined as duration from mucosal incision to tumor detachment from the gastric wall.

Closure time was defined as time from resection completion (perforation occurrence if possible) to closure completion.

Active bleeding was defined as bleeding that impaired intraoperative visibility and required coagulation forceps.

Delayed bleeding was defined as bleeding caused by postoperative wound ulcers. Pneumoperitoneum was defined as disappearance of hepatic dullness or obvious tympany on postoperative abdominal percussion.

Peritonitis was defined as abdominal tenderness, rebound tenderness, and peritoneal irritation on physical examination, combined with CT findings of peritonitis. Presence and severity of pneumoperitoneum were evaluated by two independent personnel: a gastroenterologist (who did not participate in the procedure) and a radiologist. In case of disagreement between the two evaluators, a third senior gastroenterologist was consulted to reach a consensus.

The preoperative blood routine was tested on the second day after admission; postoperative blood routine was tested on the second day after surgery. Maximum postoperative body temperature was measured within 48 hours after surgery.

“Cost” in this study refers to the direct medical costs associated with the EFR or PPP-EFR procedure and the immediate postoperative period (within 72 hours after the procedure). Specifically, it includes the following items: disposable endoscopic devices: endoloops, titanium clips, thermal forceps, injection needles, and biopsy forceps; medications: antibiotics, analgesics, and antiemetics; Imaging examinations: abdominal x-ray or CT scans (performed to evaluate pneumoperitoneum or other complications); laboratory tests: blood routine, biochemical tests, and inflammatory markers (such as C-reactive protein) performed before and after the procedure; and nursing care costs related to the procedure.

Follow-up

All patients were followed up by telephone 1 month after endoscopic treatment. Standard gastroscopy was performed at 6 and 12 months postoperatively to assess for residual or recurrent tumors. Thereafter, annual gastroscopy and abdominal CT were performed for evaluation. In addition, patients with intermediate or high-risk GISTs were advised to take imatinib and undergo close endoscopic monitoring.

Statistical methods

SPSS 25.0 software was used for data analysis. Measurement data with normal distribution were compared between groups using independent samples t -test; data with non-normal distribution were compared using Mann-Whitney U test. Count data were expressed as cases (%), and comparisons between the two groups were performed using chi-square test or Fisher’s exact test. P < 0.05 was considered statistically significant.

Results

Baseline data and clinicopathological characteristics

A total of 59 patients were included, including 29 males and 30 females. There were 31 cases in the EFR group (median age 65 years; range 62–68) and 28 cases in the PPP-EFR group (median age 64years; range 60–67). Among the 59 patients, 14 had hypertension, seven had diabetes, and three had both hypertension and diabetes. Tumor locations were: gastric fundus (38 cases, 64.4%), gastric body (15 cases, 25.4%), gastric antrum (5 cases, 8.5%), and cardia (1 case, 1.7%). EUS showed that 54 cases (91.5%) of tumors originated from the muscularis propria and only five cases (8.5%) originated from the muscularis mucosae; 51 cases (86.4%) showed intraluminal growth and eight cases (13.6%) showed extraluminal growth. There were no significant differences between the two groups in gender, age, comorbidities, tumor location, circumferential location, or growth pattern. Postoperative pathology showed no significant difference in maximum tumor diameter between the two groups. According to the NIH risk classification, all 59 cases were classified as very low-risk or low-risk ( Table 1 ).

Table 1 Basic data and clinicopathological features of the patients with small gastric stromal tumors among different surgical method groups.

Clinical data	EFR (n = 31)	PPP-EFR (n = 28)	P value
EFR, endoscopic full-thickness resection; PPP-EFR, endoscopic full-thickness resection with preset pursestring procedure; Q, quarter.
Gender (male/female)	16/15	15/13	0.88
Age [years, M (Q1, Q3)]	65 (62, 68)	64 (60 ,67)	0.76
Comorbidities [case (%)]	12 (38.7)	9 (32.1)	0.59
Preoperative white blood cell count (×10⁹/L, x ± s)	5.3 ± 1.2	5.1 ± 1.1	0.35
Tumor location [case]
Cardia	1	0	0.47
Gastric fundus	20	18
Gastric body	7	8
Gastric antrum	3	2
Circumferential location [case]
Lesser curvature	6	5	0.10
greater curvature	8	7
Anterior wall	7	7
posterior wall	10	9
Growth pattern [case]
Intraluminal type	26	25	0.54
Extraluminal type	5	3
Maximum pathological diameter of tumor (cm, x ± s)	1.5 ± 0.3	1.6 ± 0.4	0.08
Risk classification: very low risk or low risk [case (%)]	31 (100)	28 (100)

Intraoperative and postoperative comparisons between PPP-EFR and EFR

All lesions were resected en bloc without tumor rupture, intraoperative active bleeding, or conversion to surgery. Incidence of pneumoperitoneum was 13 cases in the EFR group and three cases in the PPP-EFR group, with a statistically significant difference ( P = 0.007). Intraoperative thermal forceps hemostasis was performed in 15 cases in the EFR group and four cases in the PPP-EFR group, with a statistically significant difference ( P = 0.005). The median number of titanium clips used was six (range 4–8) in the EFR group and seven (range 4–9) in the PPP-EFR group, with no significant difference ( P = 0.781). R0 resection rates were 90.0% (28/31) in the EFR group and 92.8% (25/28) in the PPP-EFR group, with no significant difference ( P = 0.893). The PPP-EFR group had a significantly shorter closure time than the EFR group (3.5 min vs. 19.5 min, P = 0.001). Incidence of postoperative fever was significantly higher in the EFR group (32.3% vs. 10.7%, P = 0.040). Postoperative white blood cell count and neutrophil percentage were higher in the EFR group than in the PPP-EFR group ( P = 0.002 and P = 0.003, respectively). Incidence of postoperative peritonitis was higher in the EFR group (29.0% vs. 7.1%, P = 0.031) and the postoperative fasting time was longer in the EFR group (3 days vs. 2 days, P = 0.038). The utilization rate for dental floss traction in PPP-EFR group was lower compared with the EFR group (10.7% vs.32.3%, P = 0.003). Routine postoperative antibiotics (ceftriaxone) were used for infection prevention. If abdominal pain, fever, or leukocytosis occurred after 24 hours, antibiotics were upgraded (sulbactam/cefoperazone). Antibiotic upgrading was performed in 18 cases (58.0%) in the EFR group and six cases (21.4%) in the PPP-EFR group, with a statistically significant difference ( P = 0.004). No delayed bleeding occurred in any patient after surgery. All patients fasted routinely for 2 days and resumed diet if no discomfort was reported. There was no significant differences in operation time or total length of hospital stay between the two groups ( Table 2 ).

Table 2 Comparison of intraoperative and postoperative conditions in patients with small gastric stromal tumors among different surgical method groups.

Observation indicators	EFR (n = 31)	PPP-EFR (n = 28)	P value
EFR, endoscopic full-thickness resection; PPP-EFR, endoscopic full-thickness resection with preset pursestring procedure; Q, quarter.
Operative time [min, M (Q1, Q3)]	41.0 (38.5, 53.0)	42.8 (39.0, 49.6)	0.132
Closure time [min, M (Q1, Q3)]	19.5 (12.5, 24.0)	3.5 (2.0, 4.5)	0.001
Resection time [min, M (Q1, Q3)]	27.5 (15.5, 40.5)	22.0 (13.5, 26.8)	0.147
En bloc resection [n (%)]	31 (100)	28 (100)
R0 resection rate [n (%)]	28 (90.0)	25 (92.8)	0.893
Pneumoperitoneum incidence [ n(%)]	13 (41.9)	39 (10.7)	0.007
Hemostatic forceps usage rate [n (%)]	15 (48.4)	4 (14.3)	0.005
Number of titanium clips [M (Q1, Q3)]	6 (4, 8)	7 (4,9)	0.781
Postoperative white blood cell count (×10⁹/L, x ± s)	9.3 ± 3.2	6.5 ± 2.1	0.002
Postoperative neutrophil percentage (x ± s)	78.3 ± 7.6	72.4 ± 6.2	0.003
Antibiotic escalation [n (%)]	18 (58.0)	6 (21.4)	0.004
Dental floss traction [n (%)]	10 (32.3)	3 (10.7)	0.003
Postoperative fever [n (%)]	10 (32.3)	3 (10.7)	0.040
Postoperative peritonitis [n (%)]	9 (29.0)	2 (7.1)	0.031
Postoperative fasting time [day, M (Q1, Q3)]	3 (2, 5)	2(1.5, 3)	0.038
Length of hospital stay [day, M (Q1, Q3)]	7 (5, 9)	6(4.5, 7.0)	0.069

Hospitalization costs and follow-up

There was no significant difference in average total hospitalization cost between the EFR and PPP-EFR groups ( P = 0.437). Follow-up duration was 48 months for all the patients. During follow-up, all patients in both groups survived with no tumor recurrence or metastasis.

Discussion

Endoscopic surgery for small gastric GISTs has advantages such as minimal invasiveness, fewer AEs, and lower costs ⁷ . EFR is more suitable for treatment of gastric GISTs originating from the muscularis propria or closely attached to the serosa ⁸ . We previously used PPP combined with ESE to resect a cecal submucosal tumor, achieving rapid intraoperative wound closure and avoiding abdominal infection ⁶ . However, clinical evidence about whether PPP can facilitate endoscopic resection of small gastric GISTs reduce AEs is lacking.

In this study, all patients were ≥ 18 years old and the maximum pathological diameter of all tumors was < 2 cm. Based on clinical experience, Meier et al. ⁸ found that not only tumor size but also tumor characteristics (e.g., location, intraluminal or extraluminal growth) determine the possibility of complete resection. Therefore, adequate preoperative evaluation and appropriate surgical procedure selection are essential. Our study showed that both EFR and PPP-EFR achieved high complete resection rates (90.0% and 92.8%, respectively) in treatment of small gastric GISTs. Although some patients did not achieve R0 resection, no recurrence or metastasis was observed during follow-up. This may be due to tumor capsule damage by the endoscopic tip or instruments during surgery, resulting in microscopically positive margins.

A clear surgical field of view in EFR helps avoid damage to blood vessels and the serosa. Comparison showed that incidence of pneumoperitoneum and proportion of thermal forceps hemostasis were lower in the PPP-EFR group, which was related to comprehensive traction on the mucosa around the lesion by PPP, resulting in a better surgical field of view. Comprehensive traction refers to traction applied to the mucosa surrounding the tumor. Leveraging the intrinsic internal traction effect of titanium clips placed around the lesion, this technique expands the surgical field of view, achieves more adequate exposure of the tumor mass, enables more precise dissection, prevents serosal layer injury, and minimizes risk of vascular damage. This helps reduce incidence of peritonitis and use of hemostatic forceps. When suboptimal tumor exposure occurs intraoperatively due to lesion location, we combine the above approach with external traction using dental floss. The PPP-EFR group yielded superior tumor exposure outcomes, resulting in a significantly lower utilization rate for dental floss traction compared with the EFR group.

A shorter perforation exposure time can reduce leakage of gastrointestinal contents into the abdominal cavity, thereby lowering incidence of complications such as peritonitis. Wound closure in conventional EFR is not easy. Intraoperative perforation often leads to difficulty in gastric insufflation and aggravated gastric peristalsis, resulting in poor visibility, time-consuming and labor-intensive wound closure, and increased incidence of postoperative peritonitis. The greatest advantage of PPP is rapid wound closure in EFR, and thus, incidence of postoperative peritonitis was significantly lower than that for EFR alone. Due to the low incidence of postoperative fever, abdominal pain, and inflammatory reactions, the rate of antibiotic upgrading and fasting time were significantly reduced in the PPP-EFR group. However, no significant difference in length of hospital stay was found between the two groups, which may be related to the additional hospital stay required for treatment of comorbidities such as diabetes and hypertension.

In this study, the operator had more than 3 years of independent EFR experience and had completed more than 150 conventional EFR procedures before the start of this study, indicating that his technical proficiency had reached a stable plateau, thus eliminating the impact of the initial learning curve. Selection of EFR or PPP-EFR was mainly based on the time period of the procedure and technical development of our team; there was no overlap between the two periods. We objectively acknowledge the limitation of the retrospective time-based grouping design and suggest that future prospective randomized controlled trials should be conducted to more accurately control for time-related bias.

Conclusions

In summary, our study indicates that PPP is a valuable auxiliary technique for EFR in small gastric GISTs. It optimizes the closure process, reduces inflammatory and abdominal complications, and improves postoperative recovery (shorter fasting time) without increasing medical costs or compromising resection efficacy, making it a worthy option for clinical promotion in eligible cases.