We reviewed all cases of endoscopy and ESD to treat gastrointestinal tumors performed at the Shirakawa Clinic from April 2006 to March 2011. ESD was performed not only for treatment purposes, but also to make accurate histological diagnoses. Indications for ESD primarily included mucosal types of gastrointestinal cancers, excepting poorly differentiated cancers. A detailed description of the indications has been previously reported[9-12]. Depending on the procedure, we followed the guidelines of either the Japan Esophageal Society[14], the Japanese Gastric Cancer Association[15] or the Japanese Society for Cancer of the Colon and Rectum[16].

The Shirakawa Clinic includes 19 inpatient beds, 5 endoscopists and 5 endoscopical technicians certified by the Japan Gastroenterological Endoscopy Society. The endoscopic treatment room and treatment procedures are shown in Figure 1. The treatment teams included two doctors (one operator and one doctor who administered propofol sedation), two nurses and two endoscopical technicians. The years of experience of the endoscopists who performed the endoscopic treatment, including EMR, were as follows: 24 (Onozato Y), 20 (Sohara N), 16 (Hagiwara S), 14 (Iizuka H), and 11 (Arai R) years. The number of years of experience with ESD treatment were 10 (Onozato Y), 5 (Sohara N), 4 (Hagiwara S), 8 (Iizuka H), and 2 (Arai R) years. The numbers of ESD treatments performed were about 1000 (Onozato Y), 250 (Sohara N), 150 (Hagiwara S), 300 (Iizuka H), and 50 (Arai R) cases, respectively. All ESD procedures were supervised by Onozato Y, who had most experience performing ESD among the five endoscopists. As a result, all ESD cases were similarly treated because of the careful supervision.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Endoscopic treatment room (A) and treatment procedures (B) of the Shirakawa Clinic. The treatment team included two doctors [one operator (1) and one doctor to administer propofol sedation (2)], two nurses who care for the patients (3) and two endoscopical technicians (not in photo). 4: Vital sign monitor.

The intravenous administration of pethidine hydrochloride and propofol was used for sedation, except in colorectal ESDs. The tumors were treated using standard ESD procedures as previously described[9-12]. For each patient, a data collection sheet was used to obtain relevant clinical information about the patient, tumor, procedure and complications. The data sheets were reviewed retrospectively. Findings from abdominal X-P, physical examinations and blood tests were assessed using a clinical pathway as a basis. Patients without complications were permitted to eat soft foods two days after ESD and then were discharged after endoscopic follow-up on the seventh day. The efficacy and complications of ESD were assessed.

One-piece resection was defined as en bloc resection. The resections were considered to be curative when tumor-free vertical and horizontal margins were achieved, en bloc resection was performed and resection criteria were met[14-16]. The operation time was calculated from the data of digital versatile disc recording systems for endoscopic procedures. Time from the start of the mucosal incision to the completion of the dissection was defined as the operation time. The time for marking or hemostasis after dissection was excluded from the operation time.

Procedure-related bleeding was subdivided into intra- and postoperative groups. Intraoperative bleeding was defined as large amounts of bleeding during the procedure and difficult cases of hemostasis. Postoperative bleeding was defined as clinical evidence of bleeding as evidenced by hematemesis or melena at 0-30 d after ESD and requiring endoscopic treatment. Furthermore, postoperative bleeding was subdivided into early postoperative (within three days of ESD) and delayed postoperative (more than three days after ESD) groups. We also recorded the incidence of perforation during endoscopy and after the procedures based on clinical evidence.

Endoscopic follow-up examinations were performed routinely one week, three months, nine months and every six months after ESD. To exclude distant metastases, abdominal ultrasonography (and/or computed tomography) was performed before ESD and three months and every six months after ESD on the same day as endoscopy.

The continuous variables are presented as the mean ± SD. The data were analyzed using the Fisher’s exact probability test, and Mann-Whitney’s U-test. A P value < 0.05 was considered to be significant.

From April 2006 to March 2011, a total of 35 540 endoscopies were performed at the Shirakawa Clinic. A total of 1047 lesions in 873 patients were resected with ESD (Table 1). Sixty-four patients with esophageal cancer (E), 850 patients with gastric tumors (G: 764 patients with cancer, 82 patients with adenomas and four others), four patients with duodenal cancer (D) and 129 patients with colorectal tumors (C: 94 patients with cancer, 21 patients with adenomas and 14 others) were treated. The en bloc resection rate was 94.3% (E: 96.9%, G: 95.8%, D: 100%, C: 79.8%). The median operation time was 46 min (range: 4-360 min) and the mean size of the resected specimens was 18 mm (range: 2-150 mm). No mortal complications were observed in association with the ESD procedures. Perforation occurred in 12 cases (1.1%, E: 1 case, G: 9 cases, D: 1 case, C: 1 case) and postoperative bleeding occurred in 53 cases (5.1%, G: 51 cases, D: 1 case, C: 1 case); however, no cases required emergency surgery or blood transfusion. All of the perforations were resolved using endoscopic clipping or hemostasis. The other problematic complication observed was pneumonia, which was treated with conservative therapy.

The follow-up results of the 64 patients with esophageal cancer are shown in Figure 2A. The median follow-up period was 24 mo (range: 12-62 mo). Forty-six patients met the absolute indication criteria of the guidelines issued by the Japan Esophageal Society in 2007[14]. Twelve patients with lesions did not meet the conventional absolute criteria and instead met expanded indication criteria. Four patients did not meet the indication criteria. The en bloc resection rate was 96.9%. One of the 12 patients who met the expanded indication criteria and two of the four patients who did not meet the indication criteria received additional treatment. No local recurrences were observed in the patients who met the indication criteria.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 The clinical courses after the endoscopic submucosal dissection of esophageal cancers (A), gastric lesions (B), duodenal cancers (C) and colorectal lesions (D). ESD: Endoscopic submucosal dissection; EP: Epithelium; MM: Muscularis mucosae; LPM: Lamina propria mucosae; Well diff.: Well differentiated adenocarcinoma.

The follow-up results of the 850 patients with gastric tumors (764 patients with cancer, 82 patients with adenomas and 4 others) are shown in Figure 2B. The median follow-up period was 26 mo (range: 12-62 mo). Four hundred and thirty-three of the 764 patients with gastric cancer met the absolute indication criteria of the guidelines issued by the Japanese Gastric Cancer Association in 2010[15]. Two hundred and thirty-nine patients with lesions did not meet the conventional absolute indication criteria and instead met expanded indication criteria. Ninety-two patients with lesions exceeded the indication criteria. The en bloc resection rate was 95.8%. Curative resection was achieved in 653 of the 764 patients with gastric cancer (85.5%). Nineteen patients with lesions underwent non-curative resection (non en bloc resection was performed, etc.) but were diagnosed curative clinically. No local recurrences were observed in the patients who met the indication criteria or in those who underwent clinically curative resection.

The follow-up results of the four patients with duodenal cancer are shown in Figure 2C. The median follow-up period was 16 mo (range: 14-36 mo). All four patients underwent curative resection, and the en bloc resection rate was 100%.

The follow-up results of the 129 patients with colorectal tumors (C: 94 patients with cancer, 21 patients with adenomas and 14 others) are shown in Figure 2D. The median follow-up period was 32 mo (range: 12-62 mo). Twenty-eight of the 94 patients with colorectal cancer met the absolute indication criteria of the guidelines issued by the Japanese Society for Cancer of the Colon and Rectum in 2010[16]. Sixty-one patients with lesions did not meet the conventional absolute indication criteria and instead met expanded indication criteria. Five patients with lesions exceeded the indication criteria. The en bloc resection rate was 79.8%. Curative resection was achieved in 74 of the 94 patients with colorectal cancer (78.7%). Six patients with lesions underwent non-curative resection (because non en bloc resection was performed, etc.) but were diagnosed curative clinically. No local recurrences were observed in the patients who met the indication criteria or in those who underwent clinically curative resection.

No mortal complications were observed in association with the ESD procedures. Perforations occurred in 12 cases (1.1%, E: 1 case, G: 9 cases, D: 1 case, C: 1 case) and postoperative bleeding occurred in 53 cases (5.1%, G: 51 cases, D: 1 case, C: 1 case) (Table 1). No problematic complications related to sedation were observed. One patient who experienced postoperative pneumonia was treated with an infusion of antibiotics and recovered in one week.

Operation times and tumor size for the cases with and without complications are shown in Table 2. Because the number of esophageal, duodenal and colorectal cases complicated by perforation or bleeding was small, it was not possible to make a statistical comparison between a no complication group and a complication group. It was possible to analyze the cases of gastric lesions statistically. Tumor size and operation times in the gastric cases with perforations were significantly larger than those observed in the cases without complications (P < 0.05).

All perforations were managed endoscopically using endoclips, a patch of omentum or conservative therapy. None of the patients with perforations required further surgery, and each of these patients were hospitalized for an additional zero to two days. A representative case of perforation is shown in Figure 3. Briefly, after detecting a perforation hole, the air was removed using an endoscopic fiber. Next, endoscopic clipping was performed. When the omentum became visible through the perforation hole, we attempted to aspirate the omentum through the perforation hole and clip the omentum to the hole.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Representative cases of perforation and bleeding treated endoscopically. A: A perforation was detected (arrows); B: When the omentum became visible through the perforation hole, we attempted to aspirate the omentum through the perforation hole, then clipped the gastric wall and omentum together; C: Several clips were used to seal the hole completely; D: The appearance three months after clipping; E: Bleeding from a post-endoscopic submucosal dissection ulcer was detected; F: After removing the coagulation, the bleeding point was detected; G: Clips were used to achieve hemostasis endoscopically; H: The appearance 1 wk after clipping.

All cases of intraoperative bleeding were controlled with endoscopic procedures such as soft coagulation with hemostatic forceps (Coagrasper; FD-410LR or FD-410QR, Olympus, Tokyo). No case of intraoperative bleeding required surgical intervention or blood transfusion. Postoperative bleeding occurred in 53 cases. Early postoperative bleeding (within three days of ESD) occurred in 25/53 cases and delayed postoperative bleeding (more than three days after ESD) occurred in 28/53 lesions. All cases of postoperative bleeding were controlled with endoscopic treatment (clipping and/or electrocoagulation). No case of postoperative bleeding required surgical intervention or blood transfusion. A representative case of bleeding treated endoscopically is shown in Figure 3. Briefly, bleeding from a post-ESD ulcer was detected. After removing the coagulation, the bleeding point was detected. Clips were used to achieve hemostasis endoscopically.

Endoscopic submucosal dissection (ESD) is a minimally invasive treatment for gastrointestinal cancer. It is usually performed in general hospitals because of the high frequency of complications and the need for a high level of technical skill. The aim of this study was to determine whether ESD can be safely performed at small clinics. The efficacy, technical feasibility and associated complications of the procedures were assessed.

ESD is usually performed in general hospitals. Treatment data and outcome of ESD at general hospitals were reported. However, such reports from clinics were rare. The efficacy, technical feasibility and associated complications of the ESD procedures at a local clinic were assessed.

This study showed that ESD can be safely performed in clinics with established therapeutic methods and medical services to address potential complications. ESD is associated with a few problematic complications. However, in this study, all complications were resolved using non-surgical therapies such as endoscopic clipping.

ESD can be safely performed in clinics with established therapeutic methods and medical services to address potential complications.

ESD is a minimally invasive treatment for gastrointestinal cancer. One-piece resection is considered the gold standard for endoscopic treatment because it provides an accurate histological assessment and reduces the risk of recurrence. ESD enable us to resect large and ulcerative lesions en bloc.

This study investigated whether ESD can be safely performed at small clinics. The authors indicated ESD can be safely performed in their small 19-bed clinic with similar ESD results as previously published from general hospitals. This is an interesting topic.