Complete endoscopic closure of gastric perforation induced by endoscopic resection of early gastric cancer using endoclips can prevent surgery (with video)
Article Outline
- Abstract
- Patients and methods
- Results
- Discussion
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- References
- Copyright
Background
When gastric perforation occurs during endoscopic resection for early gastric cancer, a surgical treatment generally is performed. Considering the increasing number of EMRs and the possibility of perforation, our research sought to investigate whether endoscopic treatment for gastric perforation is possible.
Methods
From 1987 to 2004, 121 of 2460 patients who underwent gastric EMR at the National Cancer Center Hospital had gastric perforation during EMR (4.9%). The initial 4 patients were treated with emergent surgery. The subsequent 117 patients who were treated with endoclips formed our study population.
Results
Endoscopic closure with endoclips in 115 patients (98.3%) was successful. Two patients with unsuccessful endoscopic closure underwent emergent surgery. In the past 6 years, patients with perforation during gastric EMR treated with endoscopic closure had a recovery rate similar to that of the nonperforation cases.
Conclusions
Gastric perforation during endoscopic resection can be conservatively treated by complete endoscopic closure with endoclips.
What is already known on this topic
What this study adds to our knowledge
Endoscopic resection should be safe, effective, and applicable to a variety of clinical situations. However, as the number of patients who undergo EMR increases, the number of complications, such as perforation and bleeding, might increase. Bleeding is the most common complication, occurring in up to 8% of patients who undergo standard EMR.3, 4 The recently developed endoscopic resection procedure, endoscopic submucosal dissection (ESD),5, 6 has a higher risk of complications. In particular, perforation is seen more frequently during ESD compared with standard EMR procedures. The risk of perforation has been reported as about 4% in ESD7 and about 0.5% in standard EMR procedures.8 Despite requiring significant additional technical skills and a longer procedure time, the ESD technique is rapidly gaining popularity in Japan, primarily because of the ability to resect large EGCs en bloc.9, 10, 11, 12
Consequently, physicians must be able to treat patients with such complications. When gastric perforation occurs during EMR or ESD, emergency surgery generally has been performed, and the merits of the endoscopic resection are lost. We investigated whether a nonsurgical treatment modality such as endoscopic closure with endoclips would prove useful as a less invasive method of the treatment of gastric perforations after endoscopic resection.
Patients and methods
A total of 2460 patients with EGC underwent endoscopic resection at the National Cancer Center Hospital from 1987 to 2004. Gastric perforation occurred in 121 patients (4.9%) during the endoscopic procedures (101 men and 20 women; median age 67 years, range 36-84 years). Informed consent, which described the merits and the complications of gastric EMR or ESD, was routinely obtained from each patient before these procedures.
Two methods of endoscopic closure were carried out: a “single-closure method” and an “omental-patch method” with endoclips (HX-600-090; Olympus Optical Co, Ltd, Tokyo, Japan). A single-closure method is done to treat small defects (Fig. 1A to C; see video clip no. 1 online at www.giejournal.org). An omental-patch method is performed for comparatively larger defects by using either the greater omentum or the lesser omentum as a patch (Fig. 2A to D; see video clip no. 2 online at www.giejournal.org).
Figure 1.
A, Gastric perforation occurs during ESD. B, A single-closure method is done to treat small defect by using endoclips only. C, The healed perforation 3 months after endoscopic closure.
Figure 2.
A, Gastric perforation occurs during EMR. B and C, An omental-patch method is performed for comparatively larger defect by using the greater omentum as a patch. D, The healed perforation 6 months after endoscopic closure.
During the early period of our study, the management of gastric perforation consisted of 3 days of nasogastric suction and 9 days of fasting with intravenous hyperalimentation (IVH), with oral intake allowed only after a complete defect closure was confirmed radiologically when using Gastrografin (Nihon Schering K.K., Osaka, Japan), and resulted in hospitalization for 14 days. Recently, these treatments have been simplified into 1 day of nasogastric suction and 1 to 2 days of fasting without IVH or defect closure confirmation, reducing the period of hospitalization to 4 to 7 days (Table 1).
Table 1. Management after endoscopic treatment for gastric perforation
| Method | Days |
|---|---|
| Drip infusion | 2-3 |
| Nasogastric tube | 1 |
| Antibiotics∗ | 2 |
| Fasting | 1-2 |
| Admission period | 4-7 |
∗Second-generation cephalosporin. |
We reviewed the medical records of each patient with gastric perforation for tumor location and their clinical course, such as the hospitalization period and the patient's condition. The statistical analysis was carried out by using the SAS program (SAS Institute, Inc, SAS Campus Drive, Cary, NC).
Results
The rate of gastric perforation for each study period is shown in Table 2. Among all cases with perforation, 30 cases had been treated by standard EMR procedures (30/566 [5.3%]) and 91 cases by the ESD procedure (91/1894 [4.8%]). There were no statistically significant differences in the perforation rate between standard EMR procedures and the ESD procedure.
Table 2. The rate of gastric perforation according to the period
| Period | Gastric perforation rate (%) |
|---|---|
| 1987 to 1993 | 4/155 (2.6) |
| 1994 to 1996 | 12/219 (5.5) |
| 1997 to 1999 | 33/397 (8.3) |
| 2000 to 2002 | 39/856 (4.6) |
| 2003 to 2004 | 33/833 (4.0) |
| Total | 121/2460 (4.9) |
There were no statistically significant differences in the perforation rates of tumor locations (upper third of the stomach 36/431 [8.4%]; middle, 62/1209 [5.1%]; lower, 23/1000 [2.3%]). The rates of gastric perforation according to the aspect of gastric wall are shown in Figure 3.
Figure 3.
The rates of gastric perforation according to the aspect of gastric wall.
The results of 121 patients with gastric perforation during endoscopic resections for early gastric cancer are shown in Figure 4. The initial 4 patients, who underwent EMR from 1987 to 1993, were treated by emergent surgery. Two additional patients among the remaining 117 patients initially were treated with endoclips, but closure of the perforation was not possible because of severe bleeding or deterioration of general condition, and the patients were finally treated by open surgery.
Figure 4.
The results of 121 patients with gastric perforation during endoscopic resections for early gastric cancer.
In 115 patients (98.3%), endoscopic closure was possible and the clinical courses of the patients were favorable under antibiotic therapy with a second-generation cephalosporin. The most frequent clinical symptoms after closure of the gastric-wall defect were abdominal discomfort (51/115 [44%]) and low-grade fever, less than 38°C (31/115 [27%]).
The investigation of the clinical course after endoscopic treatment for gastric perforation is shown in Table 3. In the initial period (1994-1998), the average period of fasting after EMR was 6.3 days, and the median hospitalization period was 10.1 days. With the present treatment over the past 6 years (1999-2004), patients had complete endoscopic closure and a faster recovery, as evidenced by an average period of fasting after EMR of 2.1 days and a median hospitalization period of 6.7 days.
Table 3. Investigations of clinical courses after endoscopic treatment for gastric perforation
| Cases with perforation | ||
|---|---|---|
| ∼1998 | 1999∼ | |
| Fasting, d | 6.3 | 2.1 |
| Admission period, d | 10.1 | 6.7 |
There was no mortality caused by gastric perforation during endoscopic resections.
Discussion
Because of technical limitation by standard EMR procedures in the past, an accepted indication for EMR of EGC included the resection of small intramucosal cancer less than 2 cm in size of intestinal histology type.13 These standard techniques cannot be used to resect lesions larger than 15 mm in one piece.14 Piecemeal resections can cause the pathologist to render pathologic staging with inadequate certainty, and there is a high risk of a recurrence after piecemeal resections.15, 16
However, by using a large database that involves more than 5000 patients who underwent gastrectomy with meticulous R2-level lymph-node dissection, expanded groups of patients with larger EGCs with no risks of LNM were identified. Intramucosally differentiated adenocarcinoma without lymphatic venous invasion, less than 30 mm in size, irrespective of ulcer findings, had a risk of LNM of 95% confidence interval (CI) [0%, 0.3%]. Intramucosally differentiated adenocarcinoma without lymphatic venous invasion and ulcer findings, irrespective of tumor size, had a risk of LNM of 95% CI [0%, 0.4%]. Submucosal differentiated adenocarcinoma without lymphatic venous invasion less than 30 mm in size had a risk of LNM of 95% CI [0%, 2.5%].17 These groups of patients were shown to have no risks of LNM rather than the risks of surgical mortality.
In regard to patient prognosis, a decision needs to be made whether additional radical surgery is necessary after a precise histologic assessment. Such a precise staging, unfortunately, cannot be substituted accurately with any imaging technique currently available.18 Final staging is available only through an accurate histologic assessment.19 Because piecemeal resections can lead to uncertainty of pathologic staging and because there is a high risk of recurrence after piecemeal resections, a new endoscopic technique has been introduced. This technique, which uses a modified needle knife,20 uses direct dissection of the submucosa and is classified as endoscopic submucosal dissection (ESD), instead of conventional EMR. This promising procedure has the advantage of achieving large en bloc resection and allows not only precise histologic staging but may also prevent disease recurrence.
The complications of endoscopic resection for EGCs include abdominal pain, localized peritonitis, bleeding, and perforation. Although bleeding is the most common complication occurring in patients who undergo endoscopic resections, perforation is seen relatively more commonly during ESD compared with standard EMR.
Even though perforation must be avoided, understanding its conservative management when using noninvasive methods is important and is used whenever possible. This retrospective study intended to clarify whether endoscopic closure with endoclips for gastric perforation during endoscopic resections is possible and useful.
The overall rate of gastric perforation during endoscopic resection in this study was 4.9%. The perforation rate decreased to 2.6% in 1987 to 1993 compared with 8.3% in 1997 to 1999. Initially, and until the middle of 1996, the conventional method for EMR at our hospital was strip biopsy. The ESD procedure with an insulation-tipped diathermic knife was developed to resect larger lesions en bloc in 1996 and has been used in the majority of cases since 1997. Our perforation rate also has improved in the past 5 years.
Rates of gastric perforation in the upper and the middle third of the stomach, especially greater curvature of the gastric body, were higher than in the lower third of the stomach, probably because the gastric wall in these locations is comparatively thin. The endoscopic procedure for lesions on the upper and middle third of the stomach had to be achieved with a retroflexed view, compared as the availability of a straight view for the lesions on the lower third of the stomach. In addition, sufficient insufflation, which makes endoscopic resection easier, is difficult to achieve during endoscopic procedures for lesions located on the greater curvature. Especially with such lesions, an important step is to perform gastric endoscopic resection while carefully maintaining a satisfactory position of gastroscope.
Endoscopic closure with endoclips for gastric perforation was attempted, because the stomach of these patients was thought to be comparatively clean during gastric EMR or ESD because of the antibacterial effect of gastric acid. Two methods of endoscopic closure were performed: a single-closure method and an omental-patch method. When the size of perforations was smaller than 1 cm, which was the opening width of the endoclip, and the shape of perforation was linear, it was possible to seal these by the single-closure method, using one or two endoclips between muscle layer and muscle layer. An omental-patch method was carried out for comparatively larger defects (more than 1 cm in size) by using either the greater omentum or the lesser omentum as a patch. The omentum was endoscopically suctioned from the perforation, and it was brought into the inner aspect of the stomach as much as possible. Endoscopic closure by using several endoclips was attempted around a defect. In all of the 115 cases where endoclip treatment was successfully performed, emergency surgery was not necessary.
Vital signs such as blood pressure, oxygen saturation, and electrocardiograms must be continuously checked during these endoscopic procedures. If abdominal fullness because of air leakage from the perforated lesion is severe, breathing deterioration or neurogenic shock can occur. To prevent these complications when gastric perforation occurs, frequent abdominal palpation is recommended to check the degree of abdominal fullness with air. If severe abdominal fullness is noted, decompression of the pneumoperitoneum must be performed with a 14-gauge puncture needle with side slits. Before puncture, we tested with a 23-gauge needle syringe filled with saline solution after we confirmed by transabdominal US (see video clip no. 3 online at www.giejournal.org). Among 117 perforations, 55 abdominal punctures were achieved without any further complications. Also, no severe peritonitis or prolonged hospitalization related to abdominal puncture was noted.
The most frequent symptoms after endoscopic closure of a gastric perforation were abdominal discomfort and low-grade fever. Abdominal fullness because of the remaining air also was observed in some cases. There were no patients with severe complications, such as peritonitis or abscess formation, and all patients were followed, without surgical treatment. However, close clinical monitoring is crucial. Based on previous reports of perforated peptic ulcers, patients may enter an asymptomatic phase. The acute pain of peritonitis may subside, and the abdominal rigidity may lessen, while the leakage of enteric contents continues, leading to subsequent clinical deterioration.21
For a gastric perforation after endoscopic resection for EGC, strict treatment in the initial period was gradually simplified, because the clinical course of patients with gastric perforations had been extremely favorable. Especially in the last 6 years (1999-2004), the average period of fasting after ESD was 2.1 days and the median admission period was 6.7 days. We allow hospital discharge after we confirm the resolution of the patient's symptoms and abnormal laboratory data. On the other hand, standard cases with no complications have 1 to 2 days of fasting and 4 to 7 days of admission after ESD. If defect closure is complete and patients receive the present treatment at our hospital, they have a recovery rate similar to that of standard cases with no complications.
In conclusion, endoscopic closure with endoclips was sufficient for patients with gastric perforation during endoscopic resections, and these patients had a fast recovery.
Appendix. Supplementary data
Video clip 1. A single-closure method is done to treat small defect by using endoclips only.
Appendix. Supplementary data
Video clip 2. An omental-patch method is performed for comparatively larger defect by using the greater omentum as a patch.
Appendix. Supplementary data
Video clip 3. Decompression of the pneumoperitoneum is performed by 14-gauge puncture needle with side slits after testing with a 23-gauge needle syringe filled with saline solution.
References
- . Japanese classification of gastric carcinoma. [2nd English edition] Gastric Cancer. 1998;1:10–24
- Endoscopic mucosal resection. Gastrointest Endosc. 2003;52:567–579
- The survey of gastrointestinal endoscopic complications in Japan. [in Japanese] Gastroenterol Endosc. 2000;42:308–313
- . One piece resection and piecemeal resection of early gastric cancer by strip biopsy. [in Japanese with English abstract] Tokyo: Igaku-Shoin; 1998;p. 68-87
- Endoscopic resection of early gastric cancer and other tumors with local injection of hypertonic saline-epinephrine. Gastrointest Endosc. 1988;34:264–269
- Endoscopic mucosal resection for treatment of early gastric cancer. Gut. 2001;48:225–229
- Endoscopic submucosal dissection for early gastric cancer: technical feasibility, operation time and complications from a large consecutive cases. Dig Endosc. 2005;17:54–58
- Outcome of endoscopic mucosal resection for early gastric cancer: review of the Japanese literature. Gastrointest Endosc. 1998;48:550–555
- . Aggressive endoscopic mucosal resection in the upper GI tract: Hook knife EMR method. Minim Invasive Ther Allied Technol. 2002;11:291–295
- A novel method of endoscopic mucosal resection using sodium hyaluronate. Gastrointest Endosc. 1999;50:251–256
- Success rate of curative endoscopic mucosal resection with circumferential mucosal incision assisted by submucosal injection of sodium hyaluronate. Gastrointest Endosc. 2002;56:507–512
- Endoscopic submucosal dissection for early gastric cancer using the tip of an electrosurgical snare (thin type). Dig Endosc. 2004;16:34–38
- Risk factors for lymph node metastasis from intramucosal gastric carcinoma. Cancer. 1996;77:602–606
- Clinicopathological features of mucosal carcinoma of the stomach with lymph node metastasis in eleven patients. Br J Surg. 1986;73:431–433
- Clinical outcome of endoscopic aspiration mucosectomy for early stage gastric cancer. Gastrointest Endosc. 2002;56:708–713
- Is endoscopic one-piece mucosal resection essential for early gastric cancer?. Dig Endosc. 2003;15:113–116
- Incidence of lymph node metastasis from early gastric cancer: estimation with a large number of cases at two large centers. Gastric Cancer. 2000;3:219–222
- Clinical impact of strip biopsy for early gastric cancer. Gastrointest Endosc. 2004;60:771–777
- Efficacy, safety, and clinical outcomes of endoscopic mucosal resection: a study of 101 cases. Gastrointest Endosc. 2002;55:390–396
- A new endoscopic mucosal resection (EMR) procedure using an insulation-tipped diathermic (IT) knife for rectal flat lesions. Gastrointest Endosc. 1999;50:560–563
- A randomized trial of nonoperative treatment for perforated peptic ulcer. N Engl J Med. 1989;320:970–973
See CME section; p. 677.
PII: S0016-5107(05)02452-1
doi:10.1016/j.gie.2005.07.029
© 2006 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.