Submucosal tunnel docking endoscopic resection for a large gastroesophageal junction leiomyoma: A case report

Abstract

BACKGROUND

Submucosal tumors (SMT) at the gastroesophageal junction present challenges for endoscopic treatment, particularly when they are large, irregularly shaped, or originate from deep layers. Achieving complete resection while preserving cardia function remains a key objective in the clinic.

CASE SUMMARY

A 43 year old female patient was found to have a SMT in the gastric fundus during a routine health checkup. Endoscopy revealed a lesion involving three quarters of the cardia. Endoscopic ultrasound suggested a hypoechoic lesion originating from the muscularis propria. The patient underwent submucosal tunnel docking endoscopic resection (SDER). An esophageal submucosal tunnel was first created to dissect the tumor base, followed by retroflexed endoscopic dissection in the stomach. The tumor was completely removed without bleeding or perforation. The resected specimen measured 8.0 cm × 2.5 cm × 1.5 cm. Postoperative pathology and immunohistochemistry confirmed a gastric leiomyoma.

CONCLUSION

SDER proved to be a feasible technique for this large, irregularly shaped leiomyoma at the gastroesophageal junction, achieving complete tumor removal without short term complications. However, the follow-up period was limited to 30 days; longer follow-up is required to fully establish its safety and effectiveness.

Key Words: Gastric leiomyoma; Gastroesophageal junction; Endoscopic resection; Submucosal tunnel; Case report

Core Tip: Submucosal tunnel docking endoscopic resection (SDER) is a modified technique for treating a large submucosal tumor (SMT) at the gastroesophageal junction. We report a case of an 8-cm gastric leiomyoma involving three-quarters of the cardia. SDER was performed by first creating an esophageal submucosal tunnel to dissect the tumor base, followed by retroflexed endoscopic dissection in the stomach. The tumor was completely resected without bleeding or perforation. This technique may provide a safe and effective option for large, irregularly shaped SMTs at this challenging location.

INTRODUCTION

Submucosal lesions in the gastric fundus and cardia come in various pathological types, which makes preoperative diagnosis difficult. For cases that qualify for both endoscopic and surgical treatment, when choosing a treatment, it is important to consider both completely removing the tumor and keeping the cardia function intact. Endoscopic treatment can achieve these goals and is less invasive than surgery. However, endoscopic treatment can still be quite challenging for lesions that have unusual shapes, are very large, or start deep in the gastric fundus and cardia.

CASE PRESENTATION

Chief complaints

The patient was a 43-year-old female who was admitted on September 9, 2024, after a gastroscopy 10 days prior revealed a submucosal tumor (SMT) in the gastric fundus.

History of present illness

Since the onset of the condition, the patient did not experience abdominal pain, bloating, or difficulty swallowing. She had not received any prior treatment for this SMT.

History of past illness

The patient had no significant past medical history, including hypertension, diabetes, or other chronic diseases. She had no history of prior abdominal or thoracic surgery. There was no family history of gastrointestinal tumors or other hereditary conditions. She reported no psychosocial issues that might have affected treatment or recovery.

Personal and family history

There was no family history of gastrointestinal tumors or other hereditary conditions. She reported no psychosocial issues that might have affected treatment or recovery.

Physical examination

On admission, physical examination revealed stable vital signs: Blood pressure 118/76 mmHg, heart rate 72 beats per minute, respiratory rate 16 breaths per minute, and body temperature 36.5 °C. Abdominal examination showed no tenderness, rebound tenderness, or palpable masses. No abnormal findings were noted on chest or abdominal auscultation.

Laboratory examinations

Routine laboratory tests showed no abnormalities.

Imaging examinations

A computed tomography (CT) scan with contrast showed no enlarged surrounding lymph nodes or organ metastasis.

Further diagnostic work-up

Endoscopy revealed an SMT in the gastric fundus that surrounded the cardia for three-quarters of its circumference (Figure 1A and B). Endoscopic ultrasound (EUS) revealed a hypoechoic submucosal lesion in the gastric fundus, originating from the muscularis propria (Figure 1C). Due to its large size and extension across the lower esophagus and gastric fundus, the entire lesion could not be fully visualized on ultrasound. Therefore, a local cross-sectional scan was obtained in the gastric fundus, measuring approximately 3.05 cm × 1.49 cm.

Figure 1 Endoscopic and endoscopic ultrasonographic views of the lesion. A: Gastroscopic view of the lesion from the gastric fundus side; B: Gastroscopic view of the lesion from the esophageal side; C: Endoscopic ultrasonographic view of the lesion showing a hypoechoic mass originating from the muscularis propria layer.

MULTIDISCIPLINARY EXPERT CONSULTATION

Before the procedure, a multidisciplinary team consultation was conducted, involving specialists from gastrointestinal surgery, thoracic surgery, gastroenterology, and radiology. Imaging studies showed no evidence of exophytic growth of the lesion. As surgical resection would carry higher risks, greater costs, and a longer postoperative recovery compared with endoscopic treatment, and given the absence of definite malignant features, the multidisciplinary team decided to proceed with endoscopic tumor resection. Based on the available preoperative diagnostic findings, the prognosis was considered favorable.

FINAL DIAGNOSIS

Based on the preoperative endoscopic, endoscopic ultrasonographic, and contrast-enhanced CT findings, the preprocedural differential diagnosis primarily included leiomyoma and gastrointestinal stromal tumor (GIST). While GIST could not be ruled out before surgery, EUS showed no malignant features (such as irregular borders, ulceration, or cystic changes), and CT revealed no evidence of exophytic growth, lymphadenopathy, or distant metastasis. Given these reassuring imaging features, along with the patient's strong preference for a minimally invasive approach, endoscopic treatment was deemed a reasonable initial strategy even in the presence of a possible GIST. The final diagnosis was established postoperatively as gastric leiomyoma based on histopathological and immunohistochemical findings.

TREATMENT

The patient underwent endoscopic tunnel tumor resection under general anesthesia with tracheal intubation. After entering the esophagus, the gastroscope was withdrawn to a position 35 cm from the incisors (approximately 5 cm from the cardia) for injection into the submucosa, followed by a 1.5-cm longitudinal incision in the submucosa to establish a submucosal tunnel reaching the gastric fundus (Figure 2A). Within the tunnel, the submucosa was dissected to expose the tumor, and after separating the base and margins of the tumor (Figure 2B), with the goal of completely separating the tumor base, the endoscope was withdrawn from the esophageal tunnel into the gastric cavity. With a retrograde view, an injection into the submucosa was performed on the surface of the SMT (no injection into the submucosa was needed as the tumor had already been partially separated in the esophageal tunnel), and a linear incision was made along the tumor trajectory at the apex of the tumor (Figure 2C). After submucosal incision, the tumor was exposed and completely resected using an insulation-tipped (electrosurgical) knife (IT knife) in a retrograde position (Figure 2D). Hemostatic forceps were used to control bleeding in the tunnel and at the wound site, and hemostatic clips were placed to close the opening of the esophageal tunnel and the submucosal incision in the gastric fundus (Figure 2E and F).

Figure 2 Operative procedure of submucosal tunnel docking endoscopic resection. A: Creation of a submucosal tunnel in the esophagus; B: Separation of the lesion base through the esophageal tunnel; C: Linear incision made on the gastric fundus side under retroflexed endoscopic view to expose the tumor; D: Complete separation of the tumor under retroflexed endoscopic view in the gastric cavity; E: Closure of the wound in the gastric fundus using hemostatic clips; F: Closure of the esophageal tunnel opening; G: Post-operative tumor.

OUTCOME AND FOLLOW-UP

Outcome

The specimen submitted for examination was an irregular piece of gray-white tissue, approximately 8.0 cm × 2.5 cm × 1.5 cm in size, showing a gray-white cut surface (Figure 2G), identified as a spindle cell tumor. The immunohistochemical staining results showed the following: SMA (+), Caldesmon (+), CD34 (+), CD117 (-), S-100 (-), Dog-1 (-), Ki-67 (1%+). The final diagnosis was gastric leiomyoma.

Postoperative management and follow-up

The patient fasted for 3 days after surgery and was given intravenous proton pump inhibitors for 72 hours. After 3 days, the patient had no discomfort and was allowed to eat normally, gradually moving from a liquid diet to a regular diet. At the 30-day follow-up, the patient was doing well and had no discomfort. It should be acknowledged that the follow-up period in this case was limited to 30 days, which is insufficient to evaluate tumor recurrence or long-term outcomes. Further follow-up over a longer duration is required to confirm the durability of the treatment effect and to monitor for late complications or recurrence.

Patient perspective

The patient reported a positive perioperative experience. She experienced only mild discomfort during the first 24 hours after the procedure, which resolved spontaneously without the need for additional analgesics. She was satisfied with the rapid recovery and the absence of postoperative pain or dysphagia. At the 30-day follow-up, she expressed appreciation for the minimally invasive approach and stated that she would recommend this treatment to others with similar conditions.

DISCUSSION

Preoperative assessment

The pathological types of gastric SMTs are complex, with GIST, leiomyomas, and ectopic pancreatic tissue being relatively common[1]. Most gastric SMTs are benign lesions, with only a small portion exhibiting malignancy. Endoscopic removal may be considered for gastrointestinal SMTs that have a low risk of metastasis and can potentially be completely resected[2]. For the preoperative diagnosis and assessment of gastric SMTs, conventional endoscopy combined with EUS is recommended as part of routine evaluation. For SMTs suspected of potential malignancy or those that are large (longest diameter > 2 cm), CT and magnetic resonance imaging are recommended for further evaluation. In this case, the submucosal lesion in the gastric fundus extended across the cardia and gastric fundus. Conventional endoscopy estimated a longest diameter of approximately 4 cm, EUS measured a cross-sectional size of approximately 3.05 cm × 1.49 cm, and CT in the transverse plane showed a maximum diameter of approximately 3.3 cm. All three examinations demonstrated a marked discrepancy compared with the resected tumor’s longest diameter of 8.0 cm. The discrepancy observed using conventional endoscopy was likely related to the tumor’s submucosal extension growth. Both CT and EUS provide cross-sectional measurements, making it difficult to fully capture the entire tumor morphology. For gastric submucosal lesions, a definitive diagnosis requires histopathological confirmation; therefore, preoperative diagnosis is typically based on a combination of lesion location, morphology, layer of origin, and epidemiological probability. In this case, the lesion was irregular in shape, and EUS indicated that the lesion originated from the muscularis propria layer. The preoperative diagnosis favored a leiomyoma, with GIST as the primary differential diagnosis. The patient had no significant cardiopulmonary diseases, no liver or renal dysfunction, and no hematologic disorders or coagulation abnormalities that would constitute contraindications to surgery.

Endoscopic treatment selection

For submucosal GISTs or leiomyomas smaller than 6 cm in diameter, evidence-based studies have demonstrated the safety and efficacy of endoscopic resection[3,4]. In this case, the EUS examination showed that the lesion originated from the muscularis propria layer, with no sonographic features of malignancy. Chest and abdominal CT scans showed no evidence of abdominal lymph node or distant organ metastasis and no exophytic growth. Recent studies also have further validated the role of endoscopic resection as an organ preserving strategy in technically difficult gastrointestinal conditions. For instance, Dou et al[5] reported that endoscopic treatment could achieve favorable survival outcomes in selected patients with recurrent esophageal cancer after radiotherapy. A large retrospective study on giant (≥ 7 cm) esophageal or cardiac SMTs showed that endoscopic resection achieved an en bloc resection rate of 78.9% with acceptable safety, though irregular morphology was identified as a predictor of non en bloc resection[6]. After a multidisciplinary discussion, we decided to proceed with endoscopic treatment. Common endoscopic treatment methods for submucosal lesions at the gastroesophageal junction include submucosal tunneling endoscopic resection (STER), endoscopic submucosal excavation (ESE), and endoscopic full-thickness resection. In this case, the lesion was approximately 8 cm long and could not be directly observed during endoscopy. Most of the lesion was located at the fundus of the stomach and grew in a ring-like pattern. Performing STER directly would cause the tunnel to bend after crossing the gastroesophageal junction; it is estimated that complete resection of the lesion using STER would not be possible because of the tunnel space limitations. Since the EUS showed that the lesion came from the muscularis propria, ESE could effectively remove it when the endoscope is in a retroflexed position. However, during the knife dissection before endoscopic treatment, the lesion felt hard and could not be pushed, suggesting that if ESE is performed directly, it might result in a deeper and wider wound. Moreover, since the wound would be close to the gastroesophageal junction, suturing the wound afterward would be challenging. Additionally, because of where the lesion was located, some parts were very close to the gastroesophageal junction, and the retroflexed operation was influenced by the angle of the endoscope, making it quite challenging to dissect in some areas.

Technical analysis of submucosal tunnel docking endoscopic resection procedure

The main technical challenges in this case were fourfold: (1) Why conventional STER was unsuitable for a lesion spanning the cardia; (2) Why ESE alone was not preferred given the large, irregular morphology; (3) Why creating the esophageal tunnel first was advantageous; and (4) How submucosal tunnel docking endoscopic resection (SDER) might reduce the risk of a large, difficult to close wound. For esophageal and gastric SMTs originating from the muscularis propria layer with a diameter ≤ 3.5 cm, STER is considered a safe and effective optimal treatment[7,8]. The endoscopic treatment of the lesions in this case study faced several major difficulties. First, during the retroflexed endoscopic operation in the gastric cavity, the angles for endoscopic manipulation in some areas were restricted; closing the wound from open dissection in the retroflexed gastric cavity was challenging; and when using the tunnel technique for dissection, it is difficult to dissect the lesions in the retroflexed area of the tunnel within the gastric cavity. To address these difficulties in endoscopic treatment, we need a combined endoscopic treatment technique. In this case, an endoscopic tunneling tumor resection (esophageal STER combined with gastric fundus ESE) was performed. Currently, there are few reports in the literature about endoscopic tunneling tumor resection, and there are no standard operating guidelines or consensus for this technique. Lv et al[9] reported on the treatment of 6 cases of SMTs at the gastroesophageal junction using endoscopic tunneling tumor resection. This method uses separate tunnels created in the stomach and esophagus, bypassing the complex anatomy of the cardia and avoiding the difficulty associated with dissection across the cardia. Thus, this technique provides a new approach for treating such complex SMTs. SDER has several advantages over STER. It should be noted that SDER is not a conceptually new operation but rather a modification of the procedural route and sequence based on existing STER, ESE, and dual tunnel approaches. First, by dividing a single long tunnel into two segments-one in the esophagus and one in the stomach-the lesion spanning the cardia and gastric fundus can be dissected separately within these two tunnels. This overcomes the spatial limitation of the cardia encountered in a conventional esophageal tunnel and effectively reduces technical difficulty. Second, given the complex anatomy, rich vascularity, and natural angulation of the cardia between the esophagus and stomach, the tunnel docking technique enables dissection within a controlled field, thereby reducing the risk of cardia injury and intraoperative bleeding and improving procedural safety. Notably, a similar approach has been reported by Zhang et al[10], who described a double-opening tunneling method for endoscopic resection of SMTs. In their technique, openings were created at both ends of the lesion; however, the tunnel was created only at the proximal end, while the distal opening was used for tumor extraction. The procedure started with inserting the gastroscope into the gastric cavity. Then, a submucosal injection was performed on the anal side of the SMT under retroflexion, followed by a longitudinal incision of 1.5-2.0 cm, stopping at the gastroesophageal junction, establishing a submucosal tunnel in the gastric fundus, and exposing the distal end of the SMT. The endoscope was then retracted into the esophagus, where submucosal injection was performed on the oral side of the SMT, followed by a longitudinal incision of 1.5-2.0 cm to establish a submucosal tunnel in the esophageal segment, exposing the oral side of the SMT. At the gastroesophageal junction, the esophageal segment of the tunnel was connected with the gastric segment, and the tumor was removed from the intrinsic muscle layer in both directions under digestive endoscopy, with the tumor being completely extracted. Finally, hemostatic clips were used to close the mucosal openings at both ends. In this case, we performed endoscopic tunneling tumor resection in the reverse order compared to the previous report, first establishing the tunnel in the esophagus to reach the gastric fundus, aiming to separate the base and edges of the lesion within the tunnel. Successfully separating the base of the lesion was crucial for success of the surgery. For lesions in the dome of the gastric fundus that were challenging to separate in the tunnel, dissection could be left for when the endoscope was retroflexed. In this case, the base of the lesion and the left and right edges were successfully separated within the esophageal tunnel, and part of the lesion in the dome of the gastric fundus was also separated. With this in mind, we could make a linear submucosal incision along the central raised part of the lesion while in retroflexion; in this case, it was considered that most of the raised part of the lesion had already been separated within the esophageal tunnel, so submucosal injection was deemed unnecessary for most areas during the linear incision. After the linear submucosal incision, the tumor was exposed, with most of the gastric cavity surface of the tumor already in a separated state. For the parts of the tumor that were not separated, the IT knife could be used to easily and completely remove them. The dissection results show two advantages to starting with the esophageal tunnel dissection. First, it is easier to initially separate the base and edges of the lesion within the tunnel. If the tunnel is first established in the gastric fundus, when the esophageal tunnel and gastric fundus tunnel communicate, the tunnel cannot effectively store gas, leading to limited tunnel space and visibility, making it impossible to effectively separate the edges and base of the lesion. Second, performing the esophageal tunnel dissection first is better for fully preserving the normal mucosal surface of the tumor located in the gastric fundus. In this case, a linear incision at the apex of the gastric fundus lesion was sufficient to effectively separate the residual lesion, making it easier and more effective to stitch up the gastric fundus wound.

CONCLUSION

For SMTs at the gastroesophageal area, multiple studies have shown that super-minimally invasive endoscopic treatment techniques that keep the organs intact have advantages over surgical procedures in terms of postoperative quality of life[11]. In this case, during the SMT dissection process, the mucosal layers were well-defined, and the submucosal blood vessels could be anticipated and treated beforehand. Additionally, there was no significant bleeding observed during the procedure, and common complications related to STER, such as subcutaneous emphysema and pneumothorax, were not observed. The short-term follow-up showed a good recovery after the procedure. This case demonstrates that modified STER performed orally is a feasible approach for removing a large, irregularly shaped SMT at the gastroesophageal area in this selected patient. The 30 days follow up showed a good recovery without short term complications. However, longer follow up is needed to evaluate outcomes such as cardia function, gastroesophageal reflux, stenosis, delayed perforation, and potential recurrence. We hope to see more clinical studies in the future to validate these preliminary findings.