We carried out a single-center retrospective study involving 36 consecutive patients diagnosed with CFTs who underwent endoscopic resection - specifically, ESD, endoscopic full-thickness resection (EFTR), or submucosal tunneling endoscopic resection (STER) - at Zhongshan Hospital Fudan University, Shanghai, China, from December 2016 to September 2021. Only patients with available follow-up data and complete demographic and clinical information were included in the study. Preoperative assessments were based on endoscopy, computed tomography (CT), or endoscopic ultrasonography (EUS) results Tumor characteristics, endoscopic methods, complications, en bloc resection rate, complete resection rate, local recurrence, and distant metastases were evaluated for all patients. This study was approved by the Ethics Committee of Zhongshan Hospital, in accordance with the Declaration of Helsinki (B-2018-222). Informed consent was obtained from all participants.

To excise the tumor, a standard endoscope (GIF-Q260J; Olympus, Tokyo, Japan) equipped with a transparent cap (D-201-11804; Olympus, Tokyo, Japan), an I-type hybrid knife (ERBE; Erbe Elektromedizin GmbH, Tübingen, Germany), a hook knife (KD-620LR; Olympus, Tokyo, Japan), and an IT knife (KD-611L; Olympus, Tokyo, Japan). Additional accessories comprised an argon plasma coagulation unit (APC300; Erbe), hot biopsy forceps (FD-410LR; Olympus, Tokyo, Japan), and a polypectomy snare.

The patients underwent endotracheal intubation under general anesthesia. Prophylactic intravenous antibiotics were administered routinely 30 min before the procedure, and CO₂ was utilized to minimize gas-related complications during the procedure. Additionally, when necessary, a 20-gauge needle filled with normal saline was inserted into the right lower quadrant to mitigate intraoperative pneumoperitoneum.

Endoscopic mucosal resection (EMR) was indicated for GI lesions less than 20 mm in size that can be easily lifted and have a low risk of submucosal invasion[6]. The main steps of EMR were as follows: (1) After locating the lesion, a mixture of normal saline and indigo carmine was injected into the submucosa; (2) The lesion was removed using a snare; (3) Argon plasma coagulation or hot biopsy forceps were used to prevent delayed bleeding; and (4) The defect was closed with metal clips.

ESD has been used to resect lesions originating from the superficial muscularis propria (MP) with an intraluminal growth pattern[7]. The main steps were as follows: (1) A margin of 5 mm away from the edge of the lesion was marked for orientation using an endoscopic knife; (2) A mixture of normal saline and indigo carmine was injected into the submucosa; (3) The mucosal and submucosal layers were precut along the tumor base with an I-type hybrid knife; (4) The tumor was gradually dissected using an IT knife, extending as deep as the bottom around the lesion; (5) The tumor was removed with a snare; (6) Argon plasma coagulation or hot biopsy forceps were used to prevent delayed bleeding; and (7) The defect was closed with metal clips.

EFTR has been used to resect lesions originating from the deep MP or serosa with an extraluminal growth pattern[8]. The main steps of EFTR were similar to those of ESD, except that intentional perforations were performed to achieve complete tumor removal. However, the serosa was preserved whenever possible to prevent gas extravasation and intra-abdominal contamination. Closure of the defect following EFTR required advanced endoscopic skills. When the defect was smaller or slightly larger than the width of the open clip, metal clips were used for defect management. For defects larger than 15-20 mm, a combination of metal clips and nylon rope or the omental-patch method was necessary.

STER has a lower risk of leakage; however, its application is limited by the anatomical location of the lesion. The most suitable sites for STER are the esophagus and cardia. The main steps are as follows: (1) A mixture of normal saline and indigo carmine is injected into the submucosa 2-4 cm proximal to the lesion, and a mucosal incision is made to serve as the tunnel entry; (2) Injection and dissection are carried out between the submucosal and muscular layers to create a tunnel toward the lesion; (3) The tumor is dissected in a manner similar to that of ESD or EFTR; (4) The tumor is removed using a snare; (5) Argon plasma coagulation or hot biopsy forceps are used to prevent delayed bleeding; and (6) The mucosal entry is closed with metal clips.

After endoscopic resection, a nasogastric or rectal tube was placed to decompress and monitor the wound for delayed bleeding. Additionally, we monitored for postoperative symptoms, including fever, hematemesis, melena, dyspnea, chest pain, and signs of pneumothorax. If a patient had persistent fever, hematemesis, melena, or pain, emergent endoscopy and CT would be performed. Furthermore, proton pump inhibitors, antibiotics, and hemocoagulase were administered.

Assessments of the primary outcomes included: (1) The procedure duration, which began with the mucosal incision for STER or the marking of several dots around the lesion for ESD and EFTR, and concluded with the closure of the wound; (2) En bloc resection, defined as the excision of the tumor in one piece without fragmentation; and complete resection, defined on the basis of en bloc resection, excision without apparent residual tumors assessed macroscopically by the endoscopist at the resection site and with negative margins (both lateral and basal resection margins) on pathological examination; (3) Complications and the duration of the hospital stay; and (4) Local recurrence and distant metastases.

Patients underwent regular follow-up to evaluate wound healing, local recurrence, and distant metastases through endoscopy or telephone interviews at 3 and 6 months after resection. Detailed telephone interviews were conducted by trained physicians for patients who were unwilling to return for follow-up visits. These interviews included questions about any examinations and treatments received at other hospitals. The final follow-up was conducted in March 2022. For patients who experienced relapses, endoscopy and EUS were performed to check for recurrent tumors, while abdominal EUS and CT were conducted to assess for distant metastases.

Continuous variables were presented as means ± SD, N ~ (µ, σ²), while categorical variables were presented as numbers and percentages of the total. Statistical analysis was performed using SPSS software version 26.0.

A total of 36 consecutive patients (19 women and 17 men) aged 46.4 ± 12.6 years, were diagnosed with CFTs and underwent endoscopic resection at Zhongshan Hospital, Fudan University, Shanghai, China. The majority of patients were asymptomatic (n = 21, 58.3%), while 12 experienced abdominal discomfort. Symptoms such as dysphagia, vomiting, and diarrhea were also observed (Table 1).

The most commonly affected sites were the stomach (n = 26, 72.2%), which included 20 cases in the gastric body, four in the gastric fundus, and two in the gastric antrum. This was followed by the large intestine (n = 6, 16.7%), the esophagus (n = 3, 8.3%), and the small intestine (terminal ileum, n = 1, 2.8%). Within the large intestine, the rectum was the most frequent location (n = 4). During preoperative endoscopy, all CFTs showed a tenacious texture and broad pedicles with smooth mucosa. Most CFTs presented with an intraluminal growth pattern (n = 32, 88.9%). The gross morphology revealed that the well-defined CFTs were either round (n = 34, 94.4%) or dumbbell-shaped (n = 2, 5.6%), with a gray surface. The maximum diameter of the CFTs ranged from 10.6 ± 4.3 cm. The MP was the most commonly involved layer (n = 29, 80.6%), followed by the submucosa (n = 7, 19.4%). None of the patients had mucosal involvement or accompanying CFTs in other parts of the body (Supplementary Table 1).

Endoscopic resection is the primary treatment method for CFTs. In this study, all 36 patients underwent endoscopic resections, including STER (n = 5, 13.9%) (Figure 1), EMR (n = 1, 2.8%), ESD (n = 14, 38.9%), and EFTR (n = 16, 44.4%) (Figures 2 and 3). Sixteen (44.4%) CFTs were resected using EFTR. The suture method included metal clips (n = 25, 69.4%) (Figure 1), a combination of metal clips and nylon rope (n = 8, 22.2%) (Figure 2), and fully covered removable nitinol self-expandable metal stents (SEMS) (n = 1, 2.8%) (Figure 3). Postoperative GI decompression was performed for 16 patients (44.4%). The mean duration of the procedures ranged from 39.8 ± 29.8 min. Minor bleeding occurred in all cases during resection, but there were no intraoperative complications. Overall, en bloc resection was performed for 34 CFTs (94.4%), and complete resection was achieved in 34 cases (94.4%).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Submucosal tunneling endoscopic resection for a calcifying fibrous tumor of the rectum originating from the muscularis propria. A: Calcifying fibrous tumor of the rectum; B: The mass as observed under the endoscopic ultrasonography; C: The mass within the submucosal tunnel; D: Entrance to the submucosal tunnel; E: Metal clip used to suture the wound; F: Gross appearance of the resected mass.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Endoscopic full-thickness resection for a calcifying fibrous tumor of the stomach originating in the muscularis propria. A: Calcifying fibrous tumor of the stomach body; B: The mass as observed under the endoscopic ultrasonography; C: The mass originating from the muscularis propria; D: Omentum visible through the wound; E: Combination of nylon rope and metal clips to suture the wound; F: Gross appearance of the resected mass.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Endoscopic full-thickness resection for a calcifying fibrous tumor of the esophagus originating from the muscularis propria. A: Calcifying fibrous tumor in the esophagus; B: The mass as observed under endoscopic ultrasonography; C: The mass originating from the muscularis propria; D: The large wound; E: Metal stent used to seal the wound; F: Gross appearance of the resected mass.

For the 16 CFTs that underwent full-thickness resection, they were located in the MP of the esophagus (n = 1, 6.2%), stomach (n = 13, 81.2%), colon (n = 1, 6.2%), and rectum (n = 1, 6.2%). Furthermore, the wounds were sutured using metal clips (n = 7, 43.7%), purse-string sutures combined with metal clips and nylon rope (n = 8, 50.0%), and SEMS (n = 1, 6.2%). The majority of patients (n = 12, 75%) underwent GI decompression following resection. A CFT located in the deep layer of the esophageal MP was excised using EFTR, with a stent employed intraoperatively to repair the large wound defect in the esophagus (Figure 3).

The average duration of hospital stay post-surgery was 2.9 ± 1.2 d. None of the patients experienced delayed wound bleeding, GI tract leakage, secondary peritoneal infection, or abdominal infection. Additionally, two patients developed new growths of CFTs near the sites of the primary tumors that had been completely resected. No patient developed distant metastases during the follow-up period (Supplementary Table 1). The two patients with new growths underwent re-treatment with ESD, and no recurrences were observed during follow-up.

Calcifying fibrous tumors (CFTs) are rare mesenchymal lesions found in various body sites, including the tubular gastrointestinal (GI) tract.

To the best of our knowledge, limited research has been conducted on GI CFTs, which is the main focus of this study.

Analyze clinical findings in 36 patients with GI tract CFTs to guide diagnosis and treatment.

Retrospective study of 36 GI tract CFT patients, collecting demographic and clinical data, with regular follow-ups for local recurrence assessment.

Stomach involvement was most common (72.2% of 36 CFTs). Various endoscopic resection techniques were employed, with 94.4% achieving complete resection in 39.8 ± 29.8 min. Average tumor diameter was 10.6 ± 4.3 cm. No complications occurred during a 2.9 ± 1.2 d hospital stay. Two patients had new CFT growth near primary sites, and no distant metastases occurred in follow-up.

GI tract CFTs, rare and mostly benign, can be effectively managed through endoscopic procedures.

In the future, we need to carry out studies with prospective large samples.