Submucosal tumor (SMT) is a disease that is commonly discovered during endoscopic examination. With advances in endoscopic ultrasonography (EUS) technology, this technique has become the primary screening method for the diagnosis of upper gastrointestinal SMTs. The present study summarized the clinical data of patients who were examined and diagnosed with upper gastrointestinal SMTs by EUS, underwent endoscopic therapy or surgical treatment, and received final pathological results in our hospital between January 2011 and September 2014. Our results show that endoscopic therapy has become the main approach for the treatment of upper gastrointestinal SMTs with the development and maturation of endoscopic technology in recent years. Our conclusion suggests that the selection of endoscopic methods, such as endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), and peroral submucosal tunneling endoscopic resection (STER), under the guidance of EUS is safe and effective for the treatment of upper gastrointestinal SMTs.
Submucosal tumors (SMTs) are a class of protruding lesions with normal mucosa-covered surfaces. SMT is a disease that is commonly discovered during endoscopic examination. With advances in endoscopic ultrasonography (EUS) technology, this technique has become the primary screening method for the diagnosis of upper gastrointestinal SMTs. In the past, as the majority of upper gastrointestinal SMTs are benign, SMTs with diameters < 2 cm were primarily subjected to follow-up observation. If a tumor has a larger diameter or clinical symptoms have developed, the tumor is resected. With the development and maturation of endoscopic technology in recent years, endoscopic therapy has become the main approach for the treatment of upper gastrointestinal SMTs. In particular, the development and maturation of endoscopic submucosal dissection (ESD) and peroral submucosal tunneling endoscopic resection (STER) allow for the endoscopic resection of SMTs with large diameters (>2 cm) and SMTs originating from the deeper layer of the gastrointestinal tract (muscularis propria). The present study summarized the clinical data of patients who were examined and diagnosed with upper gastrointestinal SMTs by EUS, underwent endoscopic therapy or surgical treatment, and received final pathological results in our hospital between January 2011 and September 2014. In addition, the present study analyzed different treatment choices and treatment-induced complications and provides the basis for the choice of treatment of upper gastrointestinal SMTs.
A retrospective analysis was conducted on 154 patients who (i) were examined and diagnosed with upper gastrointestinal SMTs by EUS in our hospital between January 2011 and September 2014, (ii) underwent endoscopic therapy or surgical treatment in our hospital, and (iii) received a final pathological diagnosis. The patients carried a total of 165 lesions. Among the 154 patients, 76 were male and 78 were female (male : female = 0.97 : 1). The age of the patients ranged from 17 to 78 years with a mean age of 52.82 years. All of the patients who underwent endoscopic examination and treatment signed informed consent document. The present study was approved by the Ethics Committee of Peking University Third Hospital.
All of the patients were subjected to EUS examination. All of the EUS procedures were performed by experienced physician endoscopists in the Department of Gastroenterology. The following data were collected: basic information of the patients (i.e., name, gender, age, hospital admission number, pathology number, and examination number), the diagnostic results of the EUS examination, the utilized treatment methods, any postoperative complications, and the results of postoperative pathologic diagnosis.
The EUS system used in the present study consisted of a FUJINON EG530UR scope, a Fujinon SU-8000 ultrasound processor, and a SP-702 sonoprobe system (miniprobe EUS system). The utilized gastroscopes included the FUJINON EG590, FUJINON EG410, FUJINON EG450, FUJINON EC450WM5, and the FUJINON EC410. Other equipment included FD-410LR hemostatic forceps (Olympus, Japan), a KD-611L IT knife (Olympus, Japan), a KD-620L hook knife (Olympus, Japan), a ERBE200 high-frequency electrosurgical generator (ERBE Elektromedizin GmbH, Germany), and an APC300 argon plasma coagulation (APC) unit (ERBE Elektromedizin GmbH, Germany).
The data were subjected to statistical analysis using SPSS 19.0 software. Comparisons between two groups of samples were performed using Pearson’s chi-square test.
The 154 patients harbored a total of 165 lesions. Among the 154 patients, 2 patients harbored 1 lesion in the esophagus and 1 lesion in the stomach, 4 patients carried 2 lesions in the esophagus, 1 patient developed 3 lesions in the stomach, and 1 patient had 4 lesions in the stomach. The age of the patients ranged from 17 to 78 years, with a mean age of 52.82 years. The incidence rate was highest among patients aged 50–59 years (35.1%), followed by patients aged 60–69 years (20.8%). The long diameter and location of 164 lesions (the measured value for the long diameter of one gastric lesion was not available) were analyzed, and the results are shown in Table
The long diameter and location of the lesions.
Esophagus | Stomach | Duodenum | |
---|---|---|---|
Long diameter (cm) | |||
Minimum value | 0.3 | 0.6 | 1 |
Maximum value | 6 | 6.2 | 2.8 |
Mean value | 1.28 | 2.38 | 1.64 |
Number of lesions < 2 cm | 69 | 32 | 3 |
Number of lesions ≥ 2 cm | 12 | 46 | 2 |
Among the 165 lesions, 81 were esophageal, 79 were gastric, and 5 were duodenal, accounting for 49.1%, 47.9%, and 3.0% of the total lesions, respectively. Among the 81 esophageal lesions, the highest incidence rate (23.6%) was observed for lesions in the middle part of the esophagus, accounting for 48.1% (39/81) of all esophageal lesions. Among the 79 gastric lesions, the highest incidence rate (16.4%) was observed for lesions in the gastric body, followed by the lesions in the gastric antrum (13.9%). Lesions in the gastric body and gastric antrum accounted for 34.1% (27/79) and 29.1% (23/79) of all of the gastric lesions, respectively. In terms of the 5 duodenal lesions, 2 were located in the duodenal bulb, 1 was located at the junction between the duodenal bulb and the descending duodenum, and 2 were located in the descending part of the duodenum.
The originating layers of the lesions were analyzed. It was found that, among the 165 lesions, the largest percentage (76 lesions, 46.1%) originated from the 2nd layer of the gastrointestinal tract (muscularis mucosa), followed by the 4th layer (muscularis propria) (60 lesions, 36.4%). One lesion originated from the 5th layer (serosa). Postoperative pathological analysis indicated that the lesion was a gastrointestinal stromal tumor.
The methods employed to treat SMTs can be divided into endoscopic therapy (used to treat 102 lesions) and surgical intervention (used to treat 63 lesions). Endoscopic therapy included EMR (41 lesions), ESD (44 lesions), STER (14 lesions), and ESE (3 lesions) (Figures
Esophageal leiomyoma.
Heterotopic pancreas in the gastric antrum.
STER procedure for the resection of esophageal leiomyoma. (a) Endoscopic examination of esophageal leiomyoma. Diameter of the tumor: 2.5 cm. (b) Establishment of the submucosal tunnel. (c) Resection of the tumor. (d) and (e) Closure of the entrance of the submucosal tunnel. (f) Fixation of the specimens.
The medical information of the patients who received endoscopic therapy or surgical treatment was analyzed. No significant difference existed with respect to the age range of the patients. The mean ages of the endoscopically treated patients and the surgically treated patients were 51 years and 53 years, respectively.
See Table
Pathological diagnosis after resection of lesions and treatment methods.
Pathological diagnosis | EMR | ESD | STER | ESE | Endoscopic surgery | Laparotomy | Total |
---|---|---|---|---|---|---|---|
Stromal tumor | 2 | 4 | 2 | 36 | 5 | 49 | |
Leiomyoma | 34 | 22 | 10 | 1 | 7 | 74 | |
Heterotopic pancreas | 9 | 4 | 1 | 14 | |||
Lipoma | 1 | 5 | 1 | 7 | |||
Carcinoid | 1 | 2 | 3 | ||||
Brunner adenoma | 1 | 1 | 2 | ||||
Schwannoma | 1 | 2 | 3 | ||||
Cyst | 1 | 1 | 2 | ||||
Granulosa cell tumor | 2 | 2 | |||||
Bronchogenic cyst | 1 | 1 | |||||
Fibrous polyp | 1 | 1 | |||||
Spindle cell tumor | 1 | 1 | |||||
Hemangioma | 1 | 1 | |||||
Hyperplastic polyps | 1 | 1 | |||||
Amyloidosis | 1 | 1 | |||||
High density-induced compressional deformation of spindle cells | 1 | 1 | |||||
Benign gastric lesions that display significant hyperplasia of vascular fibrous tissue and hyaline degeneration | 1 | 1 | |||||
Chronic inflammation with intestinal metaplasia; irregularly arranged smooth muscle tissue and hyperplastic fibrous tissue lying beneath the mucosa | 1 | 1 | |||||
Total | 41 | 44 | 14 | 3 | 55 | 8 | 165 |
The resulting two sets of data were subjected to the
Comparison of the originating layers of SMTs and treatment choices.
2nd layer | 3rd layer | 4th layer | 5th layer | Total | |
---|---|---|---|---|---|
Endoscopic treatment | 70 | 22 | 10 | 0 | 102 |
Surgical treatment | 6 | 6 | 50 | 1 | 63 |
Total | 76 | 28 | 60 | 1 | 165 |
|
|||||
|
86.308 | ||||
|
<0.05 |
As shown in Table
Comparison of the locations and sizes of SMTs and the treatment choices.
Age (years) | Location (number of SMTs) | Diameter (cm) |
<2 cm |
≥2 cm | |||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Min. | Max. | Mean | Esophagus | Stomach | Duodenum | Min. | Max. | Mean | |||
Endoscopic treatment | 17 | 78 | 51 | 71 | 28 | 3 | 0.31 | 4 | 1.12 | 92 | 10 |
|
|||||||||||
Surgical treatment | 18 | 78 | 53 | 10 | 51 | 2 | 0.67 | 6.2 | 2.88 | 12 | 50 |
|
|||||||||||
|
46.20 | 83.41 | |||||||||
|
<0.05 | <0.05 |
The two sets of data were subjected to the
With respect to tumor locations, EMR was the most commonly used treatment for esophageal lesions (39/41), ESD was the most common treatment for lesions in the gastric antrum (15/44), STER was the most common treatment for lesions in the middle part of the esophagus (6/14) and the gastric cardia (4/14), and surgery was the most common treatment for lesions in the gastric body (24/63). The endoscopic treatment techniques employed in the present study were compared, and the detailed data shown in Table
Detailed analysis of the locations, originating layers, and long diameters of the tumors and the corresponding treatment choices.
EMR | ESD | STER | ESE | Endoscopic surgery | Laparotomy | Total | ||
---|---|---|---|---|---|---|---|---|
Esophagus | Upper part | 10 | 7 | 0 | 0 | 0 | 0 | 17 |
Middle part | 19 | 9 | 6 | 0 | 5 | 0 | 39 | |
Lower part | 10 | 8 | 2 | 0 | 5 | 0 | 25 | |
Total | 39 | 24 | 8 | 0 | 10 | 0 | 81 | |
|
||||||||
Stomach | Gastric cardia | 0 | 1 | 4 | 0 | 5 | 0 | 10 |
Gastric fundus | 0 | 1 | 0 | 2 | 6 | 2 | 11 | |
Junction between gastric body and fundus | 0 | 0 | 1 | 0 | 3 | 0 | 4 | |
Gastric body | 0 | 1 | 1 | 1 | 21 | 3 | 27 | |
Gastric transitional zone | 0 | 1 | 0 | 0 | 3 | 0 | 4 | |
Gastric antrum | 0 | 15 | 0 | 0 | 6 | 2 | 23 | |
Total | 0 | 19 | 6 | 3 | 44 | 7 | 79 | |
|
||||||||
Duodenum | Duodenal bulb | 1 | 0 | 0 | 0 | 0 | 1 | 2 |
Junction between duodenal bulb and descending duodenum | 0 | 0 | 0 | 0 | 1 | 0 | 1 | |
Descending duodenum | 1 | 1 | 0 | 0 | 0 | 0 | 2 | |
Total | 2 | 1 | 0 | 0 | 1 | 1 | 5 | |
|
||||||||
Originating layer | 2nd layer | 41 | 23 | 6 | 0 | 6 | 0 | 76 |
3rd layer | 0 | 20 | 1 | 1 | 5 | 1 | 28 | |
4th layer | 0 | 1 | 7 | 2 | 43 | 7 | 60 | |
5th layer | 0 | 0 | 0 | 0 | 1 | 0 | 1 | |
|
||||||||
Long diameter (cm) | Minimum value | 0.31 | 0.4 | 0.78 | 1.1 | 0.67 | 1.2 | |
Maximum value | 3.5 | 4 | 2.47 | 1.8 | 6 | 6.2 | ||
Mean value | 0.69 | 1.43 | 1.34 | 1.42 | 2.89 | 2.79 | ||
|
||||||||
Number of lesions < 2 cm | 40 | 36 | 13 | 3 | 10 | 2 | 104 | |
Number of lesions ≥ 2 cm | 1 | 8 | 1 | 0 | 44 | 6 | 60 | |
|
||||||||
Total | 41 | 44 | 14 | 3 | 54 (+1) | 8 | 165 |
Among the 102 patients who received endoscopic therapy, 3 patients (2.9%) developed complications. One patient suffered post-ESD bleeding, one patient suffered post-ESD perforation, and one patient developed mediastinal emphysema after STER. The complications resolved after conservative treatment.
Among the 63 patients who underwent surgical intervention, 2 patients (3.2%) developed complications. Vomiting occurred in 1 patient at 5 d after open subtotal gastrectomy. Considering that the patient developed gastrojejunal anastomotic edema and stenosis, jejunal feeding tube placement and conservative treatment were administered. The other patient suffered upper respiratory tract infection and was given conservative treatment. The conditions of both patients improved after the treatments.
SMTs refer to a class of gastrointestinal lesions that originate below the mucosal layer (primarily from the muscularis mucosa, submucosa, and muscularis propria). SMTs mainly include leiomyoma, stromal tumors, lipomas, and neurogenic tumors. Patients with SMT usually display no specific symptoms. In most cases, SMTs are discovered incidentally during endoscopic examination. Gastrointestinal SMTs are usually benign, and only a small portion are malignant, primarily including leiomyosarcomas, liposarcomas, and malignant stromal tumors. The majority of gastrointestinal SMTs have no specific endoscopic appearance. Therefore, it is very difficult to determinate the nature of SMTs using an ordinary endoscopic examination. EUS is capable of determining, in general, the nature of a lesion based on the originating layer, size, and internal echoes of the lesion. Therefore, EUS may assist in both diagnosis and guiding treatment [
EMR is the first technology for the treatment of gastrointestinal SMTs. Previously, EMR was only employed to treat lesions confined to the mucosa, such as polyps. With advances in technology and improvements in instruments, the indications for EMR have expanded to submucosal lesions. EMR is suitable for treating SMTs that originate from the superficial layers and which have a diameter between 1 and 2 cm. In terms of the originating layer of the lesions, the absolute indications for EMR are lesions that originate from the m1 and m2 layers, while the relative indications include lesions that originate from the m3 and sm1 layers. If the mucosal layer becomes separated from the muscularis propria and a local uplift of the mucosa occurs after local injection, EMR may be chosen to treat the lesions. With the invention of the IT knife, ESD began to be used to excise early gastrointestinal cancer and precancerous lesions larger than 2 cm in diameter. ESD allows a one-time, complete en bloc resection of the lesions and provides materials for accurate final pathological diagnosis, fully reflecting the superiority of endoscopic minimally invasive resection. After ESD, there are no marginal tissue residues, and the recurrence rate is rather low. ESD is suitable for treating superficially located SMTs (in the muscularis mucosa or submucosa) larger than 2 cm in diameter and SMTs with wide bases [
The present study summarized the criteria for selecting different resection methods. The originating layer of the tumor serves as one of the criteria. In general, endoscopic therapy was the most frequently selected treatment for superficially located tumors (the 2nd and 3rd layer), while surgical intervention was often chosen for tumors originating from the deeper layers (the 4th layer). The diameters of the lesions also affect method selection. The present study divided the lesions into different groups based on diameter (>2 cm or <2 cm). The results indicated that surgical resection were often selected for tumors > 2 cm in diameter, while endoscopic procedures were the most frequently selected treatment for tumors < 2 cm in diameter. We have reviewed the literature, and Japanese and European guidelines currently recommend surgery for tumors > 5 cm in diameter and EUS-guided fine needle aspiration (EUS-FNA) for tumors with diameters between 3 and 5 cm. However, if the tumor is a gastrointestinal stromal tumor (GIST), surgery is recommended. If the tumor is leiomyoma or Schwannoma, intensive follow-up is recommended. The Japanese and European guidelines provide the following recommendations in regard to tumors < 3 cm in diameter: conducting endoscopic therapy such as EMR, performing EUS-FNA to obtain pathological data, or conducting follow-up observation (follow-up is recommended for GISTs only if the diameter is <2 cm) [
In the present study, complications only occurred after endoscopic treatment of 3 lesions (2.9%), and these complications were resolved by conservative treatment. The findings therefore indicate that endoscopic therapy is safe.
The present study retrospectively summarized SMT cases in China that were diagnosed using EUS and treated by various methods, including a variety of endoscopic procedures. However, the present study has its own limitations. First, all cases that were retrospectively analyzed in the present study were confined to 1 hospital. In addition, there were no clearly defined criteria for the selection of treatment methods. Physician endoscopists chose various treatment approaches based on their own experience. A prospective study needs to be performed to further define the criteria for the selection of various treatments for different lesions. The present study included a rather large number of SMT cases and provides guidance with respect to the selection of SMT treatments. In summary, the selection of endoscopic therapies, such as EMR, ESD, and STER, under the guidance of EUS is safe and effective for the treatment of SMTs.
The authors declare that there is no conflict of interests regarding the publication of this paper.