Indexed In : Science Citation Index Expanded(SCIE), MEDLINE,
Pubmed/Pubmed Central, Elsevier Bibliographic, Google Scholar,
Databases(Scopus & Embase), KCI, KoreaMed, DOAJ
Gut and Liver is an international journal of gastroenterology, focusing on the gastrointestinal tract, liver, biliary tree, pancreas, motility, and neurogastroenterology. Gut atnd Liver delivers up-to-date, authoritative papers on both clinical and research-based topics in gastroenterology. The Journal publishes original articles, case reports, brief communications, letters to the editor and invited review articles in the field of gastroenterology. The Journal is operated by internationally renowned editorial boards and designed to provide a global opportunity to promote academic developments in the field of gastroenterology and hepatology. +MORE
Yong Chan Lee |
Professor of Medicine Director, Gastrointestinal Research Laboratory Veterans Affairs Medical Center, Univ. California San Francisco San Francisco, USA |
Jong Pil Im | Seoul National University College of Medicine, Seoul, Korea |
Robert S. Bresalier | University of Texas M. D. Anderson Cancer Center, Houston, USA |
Steven H. Itzkowitz | Mount Sinai Medical Center, NY, USA |
All papers submitted to Gut and Liver are reviewed by the editorial team before being sent out for an external peer review to rule out papers that have low priority, insufficient originality, scientific flaws, or the absence of a message of importance to the readers of the Journal. A decision about these papers will usually be made within two or three weeks.
The remaining articles are usually sent to two reviewers. It would be very helpful if you could suggest a selection of reviewers and include their contact details. We may not always use the reviewers you recommend, but suggesting reviewers will make our reviewer database much richer; in the end, everyone will benefit. We reserve the right to return manuscripts in which no reviewers are suggested.
The final responsibility for the decision to accept or reject lies with the editors. In many cases, papers may be rejected despite favorable reviews because of editorial policy or a lack of space. The editor retains the right to determine publication priorities, the style of the paper, and to request, if necessary, that the material submitted be shortened for publication.
Tae-Se Kim , Byung-Hoon Min , Yang Won Min , Hyuk Lee , Poong-Lyul Rhee , Jae J. Kim , Jun Haeng Lee
Correspondence to: Jun Haeng Lee
ORCID https://orcid.org/0000-0002-5272-1841
E-mail stomachlee@gmail.com
Tae-Se Kim and Byung-Hoon Min contributed equally to this work as first authors.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Gut Liver 2024;18(1):40-49. https://doi.org/10.5009/gnl220508
Published online May 10, 2023, Published date January 15, 2024
Copyright © Gut and Liver.
Background/Aims: Delayed perforation is a rare but serious adverse event of gastric endoscopic submucosal dissection (ESD). The aim of this study was to clarify the clinical features and appropriate management strategy of patients with delayed perforation.
Methods: Among 11,531 patients who underwent gastric ESD, the clinical features and outcomes of patients who experienced delayed perforation were retrospectively reviewed and compared with those of the control group.
Results: Delayed perforation occurred in 15 of 11,531 patients (0.13%). The patients with delayed perforation were significantly older than those without delayed perforation (p=0.027). The median time to diagnosis of delayed perforation was 28.8 hours (range, 14 to 71 hours). All 15 patients with delayed perforation complained of severe abdominal pain after gastric ESD and underwent subsequent chest X-rays (CXRs) for evaluation. In subsequent CXR, free air was found in 12 patients (80%). For three (20%) patients without free air in CXR, delayed perforation was finally diagnosed by computed tomography. Leukocytosis was significantly less frequent in the patients without free air in CXR (p=0.022). A perforation hole smaller than 1 cm in size was more frequently observed in the six patients who underwent successful non-surgical treatments than in the nine patients who underwent surgery (p<0.001). There was no mortality related to delayed perforation.
Conclusions: One-fifth of the patients with delayed perforation did not show free air in CXR and exhibited less leukocytosis than those with free air. Non-surgical treatments including endoscopic closure might be considered as an initial treatment modality for delayed perforation smaller than 1 cm.
Keywords: Endoscopic mucosal resection, Intestinal perforation, Postoperative complications, Stomach neoplasms
Endoscopic submucosal dissection (ESD) has become a standard treatment modality for early gastric cancer with a negligible risk of lymph node metastases because of its minimal invasiveness and comparable long-term outcomes to a surgery.1 Major adverse events of gastric ESD are bleeding and perforation.2 ESD-related perforations can be subdivided into intraoperative perforations occurring during ESD and delayed perforations occurring after completion of procedure. The incidence of intraoperative perforation ranges from 1.2% to 5.2%.2 In most cases, intraoperative perforations can be successfully treated conservatively by endoscopic closure alone.3 The incidence of delayed perforation is much lower ranging from 0.1% to 0.45%.4-7 In a recent Japanese multicenter prospective study, the incidence of delayed perforation was 0.4%.8 In contrast to intraoperative perforation, delayed perforation is a serious adverse event. It is often associated with life-threatening peritonitis at the time of diagnosis. Despite this severity, few studies have reported on delayed perforation after gastric ESD and the optimal management strategy for delayed perforation remains unclear.
Thus, the objective of the present study was to clarify the clinical features and outcomes of patients undergoing delayed perforation after gastric ESD. In addition, we attempted to identify the appropriate management strategy for this serious adverse event of ESD.
A total of 11,813 patients with 13,108 lesions underwent gastric ESD at Samsung Medical Center between January 2006 and May 2022. Among them, 282 patients with 372 lesions were excluded from the study population because of unavailable ESD pathology report (109 lesions) or ESD pathology other than adenoma or adenocarcinoma (263 lesions). Finally, 11,531 patients with 12,736 lesions were analyzed. Delayed perforation was defined when all of the following criteria were met: (1) no visible gastric wall defect during ESD; (2) no free air in the chest X-ray (CXR) taken immediately after ESD; and (3) free air detected in subsequent CXRs or intraperitoneal free air with evident wall defect found in the abdominal computed tomography (CT) along with symptoms of peritoneal irritation after ESD. According to this definition, delayed perforation occurred in 15 (0.13%) of 11,531 patients who underwent ESDs for gastric adenoma or adenocarcinoma (Fig. 1). Clinicopathological data and outcomes after ESD were obtained through retrospective review of medical records from intranet resources of Samsung Medical Center. The study protocol was approved by the Institutional Review Board of Samsung Medical Center (IRB number: 2022-11-005). Written informed patient consent was waived by the Institutional Review Board due to the retrospective nature of the study. This study was conducted in accordance with guidelines of the Declaration of Helsinki.
ESD was indicated according to Japanese Gastric Cancer Treatment Guidelines9 and ESD procedures were performed as previously reported.10-12 After tumor margins were delineated by chromoendoscopy with indigocarmine spray,13 marking dots were placed around the lesion. After circumferential marking, normal saline mixed with epinephrine, indigocarmine and glycerol was injected to the submucosal layer to separate the lesion from the muscle layer. After making a submucosal cushion, circumferential precutting and submucosal dissection were then performed with a dual knife (KD-650L; Olympus, Tokyo, Japan) or insulated-tip-2 knife (KD-611L; Olympus). VIO 300D was used as the electrosurgical unit (ERBE, Tübingen, Germany). Minor bleeding during submucosal dissection was controlled using electrosurgical knives in a swift coagulation mode. Bleeding from large vessel during precutting or submucosal dissection and visible vessels in ESD-induced artificial ulcer after dissection were coagulated using hemostatic forceps (FD-410LR; Olympus) in a soft coagulation mode at a current of 80W. Perforation detected during ESD was closed with endoclips (HX-600-090L; Olympus).
All patients took CXR immediately after the procedure. In our institution, scheduled second-look endoscopy was not routinely performed.14 When patients showed symptoms or signs of peritoneal irritation, additional CXRs were initially taken and abdominal CT or upper endoscopy was selectively performed at the discretion of the attending physician.
When delayed perforation was diagnosed by subsequent CXR or abdominal CT, patients were managed conservatively or by surgical repair. Surgery was recommended by the attending physician in following cases: (1) rapidly deteriorating clinical condition due to severe peritonitis; (2) size of perforation was deemed too large for endoscopic therapy; (3) failed endoscopic clipping; or (4) non-curative resection requiring gastrectomy.
In conservative treatment, patients were managed with fasting, fluid therapy, intravenous antibiotics, and proton pump inhibitors. L-tube drainage was applied at the discretion of the attending physician.
Endoscopic tumor shape was described according to the Paris classification.15 For analyses, type I and type IIa were categorized into the elevated group and others (IIb, IIc, and IIa+IIc) were assigned to the flat/depressed/mixed group. Axial tumor location was categorized as lower third (pylorus and antrum), middle third (angle, low body, and mid-body) and upper third (high body, fundus, and cardia).
Complete resection was defined as en bloc resection with negative lateral and vertical resection margins. For differentiated-type early gastric cancer, curative resection was defined when a tumor was completely resected, did not have lymphovascular invasion, and fulfilled one of the following criteria:9 (1) tumor size ≤2 cm, mucosal cancer, no ulcer; (2) tumor size >2 cm, mucosal cancer, no ulcer; (3) tumor size ≤3 cm, mucosal cancer, ulcer present; or (4) tumor size ≤3 cm, SM1 cancer (submucosal invasion depth <500 µm from the muscularis mucosa layer). For undifferentiated-type early gastric cancer, curative resection was defined as complete resection with tumor size ≤2 cm confined to mucosa without ulcer or lymphovascular invasion.9
The size of perforation hole was determined based on endoscopy report (mainly based on opening width of endoclip) or abdominal CT findings (based on the size of gastric wall defect) which was obtained before the management of delayed perforation.
Clinicopathological characteristics were compared between those with and without delayed perforation. For patients with delayed perforation, further detailed clinicopathological characteristics were presented. Clinicopathological characteristics were compared between those who underwent non-surgical treatments only (non-surgery group: conservative treatments or successful endoscopic closure with clipping) and those who received surgery (surgery group) and also between those with and without free air in subsequent CXR. The differences between groups were evaluated using the Student t-test or Mann-Whitney test for continuous variables and the chi-square test or Fisher exact test for categorical variables. Statistical significance was set at p<0.05. All analyses were performed using SPSS version 25.0 (IBM SPSS Statistics for Windows, Version 25.0; IBM Corp., Armonk, NY, USA).
Delayed perforation occurred in 15 of 11,531 patients (0.13%) who underwent gastric ESD (Fig. 1). The clinicopathological characteristics of patients with and without delayed perforation are compared in Table 1. Patients with delayed perforation were significantly older than those without delayed perforation (70.0±11.7 vs 64.4±9.9, p=0.027).
Comparison of the Clinicopathological Characteristics of Patients with and without Delayed Perforation in Whole Study Population
Characteristic | Delayed perforation (n=15) | Without delayed perforation (n=11,516) | p-value |
---|---|---|---|
Age, yr | 0.027 | ||
Mean±SD | 70.0±11.7 | 64.4±9.9 | |
Median (range) | 71 (39–86) | 65 (20-98) | |
Sex | 0.562 | ||
Male | 10 (66.7) | 8,464 (73.5) | |
Female | 5 (33.3) | 3,052 (26.5) | |
Diabetes mellitus | 0.383 | ||
No | 13 (86.7) | 8,605 (74.7) | |
Yes | 2 (13.3) | 2,911 (25.3) | |
Hypertension | 0.127 | ||
No | 11 (73.3) | 6,087 (52.9) | |
Yes | 4 (26.7) | 5,429 (47.1) | |
Location* | 0.188 | ||
Lower third | 7 (46.6) | 7,348 (57.7) | |
Middle third | 4 (26.7) | 4,143 (32.6) | |
Upper third | 4 (26.7) | 1,082 (8.5) | |
Remnant stomach | 0 | 101 (0.8) | |
Gastric tube | 0 | 47 (0.4) | |
ESD pathology* | 0.203 | ||
Adenoma | 2 (13.3) | 3,788 (29.8) | |
Differentiated-type EGC | 12 (80.0) | 8,551 (67.2) | |
Undifferentiated-type EGC | 1 (6.7) | 382 (3.0) | |
Tumor size (pathology), mm* | 0.968 | ||
Mean±SD | 15.7±9.6 | 15.4±9.8 | |
Median (range) | 12.0 (8.0–36.0) | 13.0 (0.1–110.0) | |
Tumor depth* | 1.000 | ||
Mucosa | 13 (86.7) | 11,053 (86.9) | |
Submucosa | 2 (13.3) | 1,659 (13.0) | |
Muscularis propria | 0 | 9 (0.1) |
Data are presented as number (%) unless otherwise indicated.
ESD, endoscopic submucosal dissection; EGC, early gastric cancer.
*12,736 Lesions were analyzed (15 with delayed perforation and 12,721 without delayed perforation).
A representative image of delayed perforation is shown in Fig. 2. Clinicopathological features of patients with delayed perforation are summarized in Table 2. Details of each case are described in Table 3. The median age was 71 years (range, 39 to 86 years). Male patients accounted for 66.7%. Regarding axial location, tumors were most frequently (46.7%) located in the lower third of the stomach. Regarding circumferential location, tumors were most frequently found in the lesser curvature and greater curvature (33.3% and 33.3%, respectively). The median ESD specimen size was 4.2 cm (range, 3.0 to 6.6 cm). The median procedure time was 1.0 hours (range, 0.3 to 2.3 hours). The median time to diagnosis of delayed perforation was 28.8 hours (range, 14 to 71 hours). All 15 patients with delayed perforation complained of severe abdominal pain after gastric ESD and underwent subsequent CXRs for evaluation. Fever and leukocytosis were present in 53.3% and 66.7% of patients, respectively. No cases of delayed perforation occurred in the gastric tube following esophagectomy or remnant stomach after gastrectomy.
Clinicopathological Features of Patients with Delayed Perforation after Endoscopic Submucosal Dissection
Variable | Total (n=15) |
---|---|
Age, yr | |
Mean±SD | 70.0±11.7 |
Median (range) | 71 (39–86) |
Sex | |
Male | 10 (66.7) |
Female | 5 (33.3) |
Comorbidities | |
None | 9 (60.0) |
Diabetes mellitus | 2 (13.3) |
Hypertension | 4 (26.7) |
Axial location | |
Lower third | 7 (46.7) |
Middle third | 4 (26.7) |
Upper third | 4 (26.7) |
Remnant stomach or gastric tube | 0 |
Circumferential location | |
Anterior wall | 2 (13.4) |
Posterior wall | 3 (20.0) |
Lesser curvature | 5 (33.3) |
Greater curvature | 5 (33.3) |
Tumor shape (endoscopy) | |
Elevated | 5 (33.3) |
I | 2 (13.3) |
IIa | 3 (20.0) |
Flat/depressed/mixed | 10 (66.7) |
IIb | 1 (6.7) |
IIc | 3 (20.0) |
IIa+IIc | 6 (40.0) |
Histology | |
Adenoma | 2 (13.3) |
Differentiated-type EGC | 12 (80.0) |
Undifferentiated-type EGC | 1 (6.7) |
Tumor size (pathology), cm | |
Mean±SD | 1.6±1.0 |
Median (range) | 1.2 (0.8–3.6) |
Tumor depth | |
Mucosa | 13 (86.7) |
Submucosa | 2 (13.3) |
Specimen size (pathology), cm | |
Mean±SD | 4.5±1.3 |
Median (range) | 4.2 (3.0–6.6) |
Procedure time, hr | |
Mean±SD | 1.0±0.6 |
Median (range) | 1.0 (0.3–2.3) |
Curative resection | |
No | 1 (6.7) |
Yes | 14 (93.3) |
Post-procedural abdominal pain (yes) | 15 (100) |
Post-procedural fever (yes)* | 8 (53.3) |
Post-procedural leukocytosis (yes)† | 10 (66.7) |
Time to diagnosis, hr | |
Mean±SD | 30.0±14.7 |
Median (range) | 28.8 (14–71) |
Free air in subsequent CXRs | |
No | 3 (20.0) |
Yes | 12 (80.0) |
Size of perforation‡ | |
<1 cm | 5 (35.7) |
≥1 cm | 9 (64.3) |
Data are presented as number (%) unless otherwise indicated.
EGC, early gastric cancer; CXR, chest X-ray.
*Post-procedural fever was defined as a body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as a white blood count of 10,000/μL or higher17; ‡Perforation size could be identified in 14 patients.
Summary of the Clinical Characteristics and Treatment Outcomes of Patients with Delayed Perforation
No. | Age, yr | Sex | Histologic type | Tumor shape | Location | Tumor size, cm | Specimen size, cm | Tumor depth | Time to diagnosis, hr | Free air in CXR | Perforation size, cm | Determination of perforation size | Treatment | Reason for surgery | Hospital days | Death |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | 80 | M | MD | IIa+IIc | Angle AW | 2.6 | 6.6 | MM | 31.8 | Yes | Non-visible in EGD | EGD | Supportive only | NA | 18 | No |
2 | 39 | F | SRC | IIc | Fundus GC | 0.8 | 3.0 | LP | 24.1 | Yes | 0.5 | CT | Supportive only | NA | 7 | No |
3 | 79 | M | MD | IIb | Antrum LC | 0.8 | 3.0 | LP | 47.0 | Yes | N/A | NA | Supportive only | NA | 11 | No |
4 | 77 | F | WD | IIa+IIc | Antrum GC | 0.8 | 4.5 | LP | 28.9 | Yes | 0.5 | EGD | Endoscopic closure with clipping | NA | 10 | No |
5 | 62 | F | WD | IIa | Antrum LC | 0.9 | 3.4 | LP | 13.7 | Yes | 0.6 | EGD | Endoscopic closure with clipping | NA | 9 | No |
6 | 60 | M | WD | IIa+IIc | Antrum GC | 0.8 | 3.3 | LP | 40.0 | Yes | 0.8 | EGD | Endoscopic closure with clipping and Neoveil packing | NA | 12 | No |
7 | 86 | M | MD | IIc | Antrum AW | 0.8 | 4.2 | LP | 24.0 | Yes | 1.0 | EGD | Clipping→primary surgical repair | Unsuccessful clipping | 22 | No |
8 | 67 | M | WD | IIc | Antrum LC | 1.8 | 4.5 | MM | 14.7 | Yes | 1.2 | EGD | Clipping→laparoscopic segmental resection | Unsuccessful clipping | 11 | No |
9 | 77 | M | HGD | IIa | HB LC | 3.2 | 6.5 | Mucosa | 34.7 | Yes | 1.2 | EGD | Clipping→open wedge resection | Unsuccessful clipping | 14 | No |
10 | 68 | M | WD | IIa+IIc | Antrum GC | 1.2 | 5.1 | MM | 27.4 | Yes | 1.4 | CT | Primary surgical repair | Rapid clinical deterioration (intractable pain) | 13 | No |
11 | 62 | M | WD | I | LB GC | 1.3 | 4.0 | MM | 20.7 | Yes | 2.0 | EGD | Laparoscopic wedge resection | Large perforation size | 11 | No |
12 | 77 | F | LGD | I | HB PW | 1.2 | 3.1 | Mucosa | 28.8 | Yes | 1.7 | CT | Open wedge resection | Large perforation size | 14 | No |
13 | 80 | M | WD | IIa+IIc | LB PW | 1.2 | 4.0 | MM | 14.4 | No | 1.8 | EGD | Primary surgical repair | Large perforation size | 12 | No |
14 | 71 | M | MD | IIa+IIc | MB LC | 3.6 | 6.4 | SM 1400 | 29.0 | No | 2.0 | CT | Total gastrectomy | Non-curative resection | 12 | No |
15 | 65 | F | MD | IIa | HB PW | 2.6 | 5.7 | SM 50 | 71.1 | No | 2.2 | CT | Primary surgical repair & left colectomy due to descending colon ischemia | Rapid clinical deterioration (septic shock) | 192 | No |
CXR, chest X-ray; M, male; F, female; MD, moderately differentiated type; SRC, signet ring cell carcinoma; WD, well-differentiated type; HGD, high-grade dysplasia; LGD, low-grade dysplasia; AW, anterior wall; GC, greater curvature; LC, lesser curvature; HB, high body; LB, low body; PW, posterior wall; MB, mid-body; MM, muscularis mucosae; LP, lamina propria; SM, submucosa; EGD, esophagogastroduodenoscopy; CT, computed tomography; NA, not applicable.
In subsequent CXRs, free air was found and diagnosis of delayed perforation could be made in 12 (80%) patients. For three patients (20%; patients #13, #14, and #15 in Table 3) without free air detected in subsequent CXRs, diagnoses of delayed perforation were finally made by abdominal CT at 14.4, 29.0, and 71.1 hours after procedure (Fig. 2). For patient #15, CXRs were taken daily because the patient complained of severe abdominal pain after ESD. As serial CXRs did not show free air, the patient was conservatively managed with fasting and antibiotics. On the third day after ESD, the patient underwent septic shock and emergency abdominal CT scan was taken, which showed free air and panperitonitis. This patient required left colectomy with end colostomy due to colonic ischemia and prolonged intensive care unit treatment.
Among the 15 patients with delayed perforation (Table 3), three patients (20%) received conservative treatment only and recovered without undergoing endoscopic or surgical procedures. Three patients (20%) were successfully managed with endoscopic closure with clipping (Fig. 3). Nine patients (60%) received surgery for repairing perforation site including three patients (patients #7, #8, and #9 in Table 3) whose initial attempt of endoscopic closure had failed. Reasons for surgery were summarized in Table 3. There was no delayed perforation-related death.
Table 4 summarizes the comparison of clinical characteristics of patients with and without free air in subsequent CXR. Among patients undergoing delayed perforation, patients without free air in subsequent CXR showed significantly lower rate of post-procedural leukocytosis compared to those with free air (0% and 83.3%, respectively).
Comparison of the Clinical Characteristics between Patients with and without Free Air in Subsequent Chest X-Ray
Characteristic | No free air (n=3) | Free air (n=12) | p-value |
---|---|---|---|
Age, yr | 0.734 | ||
Mean±SD | 72.0±7.6 | 69.5±12.7 | |
Median (range) | 71 (65–80) | 73 (39–86) | |
Male sex | 2 (66.7) | 8 (66.7) | 1.000 |
Comorbidities (yes) | 1 (33.3) | 5 (41.7) | 1.000 |
Axial location | 0.123 | ||
Lower third | 0 | 7 (58.3) | |
Middle third | 2 (66.7) | 2 (16.7) | |
Upper third | 1 (33.3) | 3 (25.0) | |
Circumferential location | 0.154 | ||
Anterior wall | 0 | 2 (16.7) | |
Posterior wall | 2 (66.7) | 1 (8.3) | |
Lesser curvature | 1 (33.3) | 4 (33.3) | |
Greater curvature | 0 | 5 (41.7) | |
Tumor shape (endoscopy) | 1.000 | ||
Elevated | 1 (33.3) | 4 (33.3) | |
Flat/depressed/mixed | 2 (66.7) | 8 (66.7) | |
Specimen size (pathology), cm | 0.308 | ||
Mean±SD | 5.4±1.2 | 4.3±1.3 | |
Median (range) | 5.7 (4.0–6.4) | 4.1 (3.0–6.6) | |
Procedure time, hr | 0.295 | ||
Mean±SD | 1.4±0.8 | 0.9±0.6 | |
Median (range) | 1.0 (1.0–2.3) | 0.7 (0.3–2.1) | |
Post-procedural abdominal pain (yes) | 3 (100) | 12 (100) | 1.000 |
Post-procedural fever (yes)* | 2 (66.7) | 6 (50.0) | 1.000 |
Post-procedural leukocytosis (yes)† | 0 | 10 (83.3) | 0.022 |
Time to diagnosis, hr | 0.734 | ||
Mean±SD | 38.2±29.5 | 28.0±9.7 | |
Median (range) | 29 (14–71) | 28 (14–47) | |
Size of perforation‡ | 0.258 | ||
<1 cm | 0 | 5 (45.5) | |
≥1 cm | 3 (100) | 6 (54.5) | |
Operator | 1.000 | ||
A | 0 | 3 (25.0) | |
B | 2 (66.7) | 5 (41.6) | |
C | 0 | 2 (16.7) | |
D | 1 (33.3) | 2 (16.7) | |
Treatment method | 0.229 | ||
Non-surgery | 0 | 6 (50.0) | |
Surgery | 3 (100) | 6 (50.0) |
Data are presented as number (%) unless otherwise indicated.
*Post-procedural fever was defined as a body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as white a blood count of 10,000/μL or higher17; ‡The perforation size could be identified in 14 patients.
Among patients undergoing delayed perforation, clinicopathological characteristics of patients undergoing non-surgical and surgical treatments are compared in Table 5. The mean size of tumor in pathology was smaller in the non-surgery group than in the surgery group (1.1±0.7 cm and 1.9±1.0 cm, respectively). In the non-surgery group (patients #1, #2, #3, #4, #5, and #6 in Table 3), the size of the perforation hole was smaller than 1 cm in all patients. In contrast, all patients in the surgery group had perforation holes no smaller than 1 cm. Size of a perforation hole smaller than 1 cm was significantly associated with successful non-surgical treatment for delayed perforation.
Comparison of the Clinicopathological Characteristics between Patients Improved with Non-Surgical Treatments Only or with Surgery
Characteristic | Non-surgery (n=6) | Surgery (n=9) | p-value |
---|---|---|---|
Age, yr | 0.317 | ||
Mean±SD | 66.2±15.9 | 72.6±7.9 | |
Median (range) | 69.5 (39–80) | 71.0 (62–86) | |
Male sex | 3 (50.0) | 7 (77.8) | 0.329 |
Comorbidities (yes) | 2 (33.3) | 4 (44.4) | 1.000 |
Axial location | 0.513 | ||
Lower third | 4 (66.6) | 3 (33.3) | |
Middle third | 1 (16.7) | 3 (33.3) | |
Upper third | 1 (16.7) | 3 (33.3) | |
Circumferential location | 0.461 | ||
Anterior wall | 1 (16.7) | 1 (11.1) | |
Posterior wall | 0 | 3 (33.3) | |
Lesser curvature | 2 (33.3) | 3 (33.3) | |
Greater curvature | 3 (50.0) | 2 (22.3) | |
Tumor shape (endoscopy) | 0.580 | ||
Elevated | 1 (16.7) | 4 (44.4) | |
Flat/depressed/mixed | 5 (83.3) | 5 (55.6) | |
Histology | 0.301 | ||
Adenoma | 0 | 2 (22.2) | |
Differentiated-type EGC | 5 (83.3) | 7 (77.8) | |
Undifferentiated-type EGC | 1 (16.7) | 0 | |
Tumor size (pathology), cm | 0.033 | ||
Mean±SD | 1.1±0.7 | 1.9±1.0 | |
Median (range) | 0.8 (0.8–2.6) | 1.3 (0.8–3.6) | |
Tumor depth | 0.486 | ||
Mucosa | 6 (100) | 7 (77.8) | |
Submucosa | 0 | 2 (22.2) | |
Specimen size (pathology), cm | 0.191 | ||
Mean±SD | 4.0±1.4 | 4.8±1.2 | |
Median (range) | 3.4 (3.0–6.6) | 4.5 (3.1–6.5) | |
Procedure time, hr | 0.188 | ||
Mean±SD | 0.8±0.4 | 1.2±0.7 | |
Median (range) | 0.6 (0.3–1.4) | 1.0 (0.4–2.3) | |
Post-procedural abdominal pain (yes) | 6 (100) | 9 (100) | - |
Post-procedural fever (yes)* | 2 (33.3) | 6 (66.7) | 0.315 |
Post-procedural leukocytosis (yes)† | 4 (66.7) | 6 (66.7) | 1.000 |
Time to diagnosis, hr | 0.529 | ||
Mean±SD | 30.9±11.7 | 29.4±17.0 | |
Median (range) | 30.4 (14–47) | 27.4 (14–71) | |
Free air in subsequent CXRs (yes) | 6 (100) | 6 (66.7) | 0.229 |
Size of perforation‡ | <0.001 | ||
<1 cm | 5 (100) | 0 | |
≥1 cm | 0 | 9 (100) |
Data are presented as number (%) unless otherwise indicated.
EGC, early gastric cancer; CXR, chest X-ray.
*Post-procedural fever was defined as body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as white blood count of 10,000/μL or higher17; ‡The perforation size could be identified in 14 patients.
Delayed perforation is a rare but potentially life-threatening adverse event of gastric ESD with a reported incidence ranging from 0.1% to 0.45%.4-8 Due to its rarity, clinical features and outcomes of patients with delayed perforation have been scarcely reported and its optimal treatment strategy remains uncertain. The main limitation of previous studies on delayed perforation was that they were small case series without a control group (Kang et al.18: two cases, Homma et al.19: one case, and Yano et al.6: four cases). In the present study, we analyzed 15 patients who experienced delayed perforation after gastric ESD and compared their results with control group without delayed perforation. To the best of our knowledge, this is the largest study on delayed perforation after gastric ESD. In this study, we found that free air in the subsequent CXR was persistently absent in 20% of delayed perforation cases and that leukocytosis was significantly less frequent among patients without free air than those with free air. We also found that 40% of patients improved with non-surgical treatments only and small size of perforation (<1 cm) was significantly associated with successful non-surgical treatment.
The risk factors for delayed perforation after gastric ESD are not well known because of its rarity. In the present study, patients with delayed perforation were significantly older than those without delayed perforation. In contrast, Yamamoto et al.20 and Suzuki et al.7 reported that there were no differences in the proportion of patients older than 70 years between patients with and without delayed perforation. Meanwhile, Suzuki et al.7 reported that patients with gastric tube following esophagectomy were significantly associated with delayed perforation. Yabuuchi et al.21 reported that delayed perforation occurred in 1.3% of patients with remnant stomach, which was higher than those reported in previous studies (0.1% to 0.45%). In this study, however, no cases of delayed perforation occurred in the gastric tube following esophagectomy or remnant stomach after gastrectomy. Further large studies are required to evaluate whether old age or postoperative stomach is associated with the occurrence of delayed perforation after gastric ESD.
It is important to note that free air may not always be present in follow-up CXRs after ESD in patients with delayed perforation. In a recent study, the sensitivity of erect CXR for detecting gastric perforation was only 75%.22 In a study by Yamamoto et al.20 including five patients with delayed perforation, CXR performed in two patients did not show free air in either case. In the present study, free air in CXR was detected in only 80% (12/15) of patients and three patients were persistently negative for free air in CXR (20%, 3/15). Interestingly, patients #13 and #15 in Table 3 showed strikingly different clinical course although their CXRs were both negative for free air. In patient #15, the abdominal pain started approximately 24 hours after ESD. However, diagnosis of delayed perforation was confirmed with CT at 71.1 hours after ESD when septic shock had already occurred. This patient required prolonged intensive care unit treatment and left colectomy due to colonic ischemia. Meanwhile, diagnosis of delayed perforation was made at 14.4 hours with abdominal CT in patient #13 whose abdominal pain started six hours after ESD (Fig. 2). This patient underwent primary repair without further adverse events. As such, absence of free air in CXR may delay the diagnosis of delayed perforation and complicate the clinical course. Therefore, it is imperative not to depend on the CXR alone for the diagnosis of delayed perforation to avoid delay in diagnosis and to improve clinical outcomes. Keeping a high level of clinical suspicion and performing CT scan are required if a patient develops symptoms and signs of peritonitis after ESD even when follow-up CXRs are persistently negative for free air. In addition, approximately 50% of patients with delayed perforation did not develop fever or leukocytosis in the present study. Furthermore, leukocytosis was significantly less frequent among patients without free air than those with free air, which again emphasized the importance of a clinical suspicion in diagnosis of delayed perforation. Similarly, Yamamoto et al.20 reported that more than half of their patients with delayed perforation were asymptomatic. Clinicians should be careful not to rule out the possibility of delayed perforation based on the absence of fever or leukocytosis.
The optimal treatment strategy for delayed perforation remains controversial. In the case series of Hanaoka et al.,4 83.3% of patients (5/6) required surgery due to severe peritonitis or worsening clinical course. However, Yamamoto et al.20 reported that all five patients with delayed perforation avoided surgery because the perforation was small (5 mm or less in all cases) and the diagnosis was made early (all cases within 24 hours after ESD). In the present study, size of a perforation hole smaller than 1 cm was significantly associated with successful non-surgical treatment for delayed perforation. In all three patients whose initial attempt of endoscopic closure failed (patients #7, #8, and #9 in Table 3), the size of perforation hole was equal to or larger than 1 cm. As delayed perforation may be caused by blunt damage due to ischemic change rather than sharp laceration due to electrosurgical knife, the size of the necrotic area around the perforation hole can be large. This may make endoscopic closure with clipping difficult and can result in late dehiscence even after successful clipping, especially when the size of perforation hole is large. Based on these results, we suggest that endoscopic closure might be considered as the initial treatment modality for delayed perforation if the size of perforation hole is smaller than 1 cm. Also, in patients who do not improve after endoscopic treatment and in those with severe ongoing peritonitis, surgery should not be delayed because duration of symptoms is known to be a factor influencing the prognosis after surgery for peptic ulcer perforation.23 Another potential treatment option for delayed perforation was shielding with polyglycolic acid sheets (Neoveil; Gunze Co., Osaka, Japan). Polyglycolic acid sheets are composed of bioabsorbable soft-tissue reinforcement material and can be used to cover the perforation site. Ono et al.24 reported a case of delayed perforation following gastric ESD which was successfully treated with shielding with polyglycolic acid sheets after failure of endoclip closure. This shielding method with polyglycolic acid sheets can be considered when the size of perforation hole is large or the ESD-induced ulcer floor is fragile, which makes endoscopic closure with clipping difficult.
This study has several limitations. First, this was a retrospective study performed at a single tertiary referral center. Second, the number of study population was small. Further large-scale studies are required.
In conclusion, delayed perforation is a rare but potentially life-threatening adverse event of gastric ESD which requires active evaluation for diagnosis even when free air in the CXR is persistently absent. Endoscopic closure might be considered as an initial treatment modality for delayed perforation if the size of perforation hole is smaller than 1 cm.
No potential conflict of interest relevant to this article was reported.
Study concept and design: T.S.K., B.H.M. Data acquisition: T.S.K., B.H.M., Y.W.M., H.L., P.L.R., J.J.K., J.H.L. Data analysis and interpretation: T.S.K., B.H.M. Drafting of the manuscript: T.S.K., B.H.M. Critical revision of the manuscript for important intellectual content: T.S.K., B.H.M., Y.W.M., H.L., P.L.R., J.J.K., J.H.L. Statistical analysis: T.S.K., B.H.M. Administrative, technical, or material support; study supervision: J.H.L. Approval of final manuscript: all authors.
Gut and Liver 2024; 18(1): 40-49
Published online January 15, 2024 https://doi.org/10.5009/gnl220508
Copyright © Gut and Liver.
Tae-Se Kim , Byung-Hoon Min , Yang Won Min , Hyuk Lee , Poong-Lyul Rhee , Jae J. Kim , Jun Haeng Lee
Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
Correspondence to:Jun Haeng Lee
ORCID https://orcid.org/0000-0002-5272-1841
E-mail stomachlee@gmail.com
Tae-Se Kim and Byung-Hoon Min contributed equally to this work as first authors.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background/Aims: Delayed perforation is a rare but serious adverse event of gastric endoscopic submucosal dissection (ESD). The aim of this study was to clarify the clinical features and appropriate management strategy of patients with delayed perforation.
Methods: Among 11,531 patients who underwent gastric ESD, the clinical features and outcomes of patients who experienced delayed perforation were retrospectively reviewed and compared with those of the control group.
Results: Delayed perforation occurred in 15 of 11,531 patients (0.13%). The patients with delayed perforation were significantly older than those without delayed perforation (p=0.027). The median time to diagnosis of delayed perforation was 28.8 hours (range, 14 to 71 hours). All 15 patients with delayed perforation complained of severe abdominal pain after gastric ESD and underwent subsequent chest X-rays (CXRs) for evaluation. In subsequent CXR, free air was found in 12 patients (80%). For three (20%) patients without free air in CXR, delayed perforation was finally diagnosed by computed tomography. Leukocytosis was significantly less frequent in the patients without free air in CXR (p=0.022). A perforation hole smaller than 1 cm in size was more frequently observed in the six patients who underwent successful non-surgical treatments than in the nine patients who underwent surgery (p<0.001). There was no mortality related to delayed perforation.
Conclusions: One-fifth of the patients with delayed perforation did not show free air in CXR and exhibited less leukocytosis than those with free air. Non-surgical treatments including endoscopic closure might be considered as an initial treatment modality for delayed perforation smaller than 1 cm.
Keywords: Endoscopic mucosal resection, Intestinal perforation, Postoperative complications, Stomach neoplasms
Endoscopic submucosal dissection (ESD) has become a standard treatment modality for early gastric cancer with a negligible risk of lymph node metastases because of its minimal invasiveness and comparable long-term outcomes to a surgery.1 Major adverse events of gastric ESD are bleeding and perforation.2 ESD-related perforations can be subdivided into intraoperative perforations occurring during ESD and delayed perforations occurring after completion of procedure. The incidence of intraoperative perforation ranges from 1.2% to 5.2%.2 In most cases, intraoperative perforations can be successfully treated conservatively by endoscopic closure alone.3 The incidence of delayed perforation is much lower ranging from 0.1% to 0.45%.4-7 In a recent Japanese multicenter prospective study, the incidence of delayed perforation was 0.4%.8 In contrast to intraoperative perforation, delayed perforation is a serious adverse event. It is often associated with life-threatening peritonitis at the time of diagnosis. Despite this severity, few studies have reported on delayed perforation after gastric ESD and the optimal management strategy for delayed perforation remains unclear.
Thus, the objective of the present study was to clarify the clinical features and outcomes of patients undergoing delayed perforation after gastric ESD. In addition, we attempted to identify the appropriate management strategy for this serious adverse event of ESD.
A total of 11,813 patients with 13,108 lesions underwent gastric ESD at Samsung Medical Center between January 2006 and May 2022. Among them, 282 patients with 372 lesions were excluded from the study population because of unavailable ESD pathology report (109 lesions) or ESD pathology other than adenoma or adenocarcinoma (263 lesions). Finally, 11,531 patients with 12,736 lesions were analyzed. Delayed perforation was defined when all of the following criteria were met: (1) no visible gastric wall defect during ESD; (2) no free air in the chest X-ray (CXR) taken immediately after ESD; and (3) free air detected in subsequent CXRs or intraperitoneal free air with evident wall defect found in the abdominal computed tomography (CT) along with symptoms of peritoneal irritation after ESD. According to this definition, delayed perforation occurred in 15 (0.13%) of 11,531 patients who underwent ESDs for gastric adenoma or adenocarcinoma (Fig. 1). Clinicopathological data and outcomes after ESD were obtained through retrospective review of medical records from intranet resources of Samsung Medical Center. The study protocol was approved by the Institutional Review Board of Samsung Medical Center (IRB number: 2022-11-005). Written informed patient consent was waived by the Institutional Review Board due to the retrospective nature of the study. This study was conducted in accordance with guidelines of the Declaration of Helsinki.
ESD was indicated according to Japanese Gastric Cancer Treatment Guidelines9 and ESD procedures were performed as previously reported.10-12 After tumor margins were delineated by chromoendoscopy with indigocarmine spray,13 marking dots were placed around the lesion. After circumferential marking, normal saline mixed with epinephrine, indigocarmine and glycerol was injected to the submucosal layer to separate the lesion from the muscle layer. After making a submucosal cushion, circumferential precutting and submucosal dissection were then performed with a dual knife (KD-650L; Olympus, Tokyo, Japan) or insulated-tip-2 knife (KD-611L; Olympus). VIO 300D was used as the electrosurgical unit (ERBE, Tübingen, Germany). Minor bleeding during submucosal dissection was controlled using electrosurgical knives in a swift coagulation mode. Bleeding from large vessel during precutting or submucosal dissection and visible vessels in ESD-induced artificial ulcer after dissection were coagulated using hemostatic forceps (FD-410LR; Olympus) in a soft coagulation mode at a current of 80W. Perforation detected during ESD was closed with endoclips (HX-600-090L; Olympus).
All patients took CXR immediately after the procedure. In our institution, scheduled second-look endoscopy was not routinely performed.14 When patients showed symptoms or signs of peritoneal irritation, additional CXRs were initially taken and abdominal CT or upper endoscopy was selectively performed at the discretion of the attending physician.
When delayed perforation was diagnosed by subsequent CXR or abdominal CT, patients were managed conservatively or by surgical repair. Surgery was recommended by the attending physician in following cases: (1) rapidly deteriorating clinical condition due to severe peritonitis; (2) size of perforation was deemed too large for endoscopic therapy; (3) failed endoscopic clipping; or (4) non-curative resection requiring gastrectomy.
In conservative treatment, patients were managed with fasting, fluid therapy, intravenous antibiotics, and proton pump inhibitors. L-tube drainage was applied at the discretion of the attending physician.
Endoscopic tumor shape was described according to the Paris classification.15 For analyses, type I and type IIa were categorized into the elevated group and others (IIb, IIc, and IIa+IIc) were assigned to the flat/depressed/mixed group. Axial tumor location was categorized as lower third (pylorus and antrum), middle third (angle, low body, and mid-body) and upper third (high body, fundus, and cardia).
Complete resection was defined as en bloc resection with negative lateral and vertical resection margins. For differentiated-type early gastric cancer, curative resection was defined when a tumor was completely resected, did not have lymphovascular invasion, and fulfilled one of the following criteria:9 (1) tumor size ≤2 cm, mucosal cancer, no ulcer; (2) tumor size >2 cm, mucosal cancer, no ulcer; (3) tumor size ≤3 cm, mucosal cancer, ulcer present; or (4) tumor size ≤3 cm, SM1 cancer (submucosal invasion depth <500 µm from the muscularis mucosa layer). For undifferentiated-type early gastric cancer, curative resection was defined as complete resection with tumor size ≤2 cm confined to mucosa without ulcer or lymphovascular invasion.9
The size of perforation hole was determined based on endoscopy report (mainly based on opening width of endoclip) or abdominal CT findings (based on the size of gastric wall defect) which was obtained before the management of delayed perforation.
Clinicopathological characteristics were compared between those with and without delayed perforation. For patients with delayed perforation, further detailed clinicopathological characteristics were presented. Clinicopathological characteristics were compared between those who underwent non-surgical treatments only (non-surgery group: conservative treatments or successful endoscopic closure with clipping) and those who received surgery (surgery group) and also between those with and without free air in subsequent CXR. The differences between groups were evaluated using the Student t-test or Mann-Whitney test for continuous variables and the chi-square test or Fisher exact test for categorical variables. Statistical significance was set at p<0.05. All analyses were performed using SPSS version 25.0 (IBM SPSS Statistics for Windows, Version 25.0; IBM Corp., Armonk, NY, USA).
Delayed perforation occurred in 15 of 11,531 patients (0.13%) who underwent gastric ESD (Fig. 1). The clinicopathological characteristics of patients with and without delayed perforation are compared in Table 1. Patients with delayed perforation were significantly older than those without delayed perforation (70.0±11.7 vs 64.4±9.9, p=0.027).
Comparison of the Clinicopathological Characteristics of Patients with and without Delayed Perforation in Whole Study Population.
Characteristic | Delayed perforation (n=15) | Without delayed perforation (n=11,516) | p-value |
---|---|---|---|
Age, yr | 0.027 | ||
Mean±SD | 70.0±11.7 | 64.4±9.9 | |
Median (range) | 71 (39–86) | 65 (20-98) | |
Sex | 0.562 | ||
Male | 10 (66.7) | 8,464 (73.5) | |
Female | 5 (33.3) | 3,052 (26.5) | |
Diabetes mellitus | 0.383 | ||
No | 13 (86.7) | 8,605 (74.7) | |
Yes | 2 (13.3) | 2,911 (25.3) | |
Hypertension | 0.127 | ||
No | 11 (73.3) | 6,087 (52.9) | |
Yes | 4 (26.7) | 5,429 (47.1) | |
Location* | 0.188 | ||
Lower third | 7 (46.6) | 7,348 (57.7) | |
Middle third | 4 (26.7) | 4,143 (32.6) | |
Upper third | 4 (26.7) | 1,082 (8.5) | |
Remnant stomach | 0 | 101 (0.8) | |
Gastric tube | 0 | 47 (0.4) | |
ESD pathology* | 0.203 | ||
Adenoma | 2 (13.3) | 3,788 (29.8) | |
Differentiated-type EGC | 12 (80.0) | 8,551 (67.2) | |
Undifferentiated-type EGC | 1 (6.7) | 382 (3.0) | |
Tumor size (pathology), mm* | 0.968 | ||
Mean±SD | 15.7±9.6 | 15.4±9.8 | |
Median (range) | 12.0 (8.0–36.0) | 13.0 (0.1–110.0) | |
Tumor depth* | 1.000 | ||
Mucosa | 13 (86.7) | 11,053 (86.9) | |
Submucosa | 2 (13.3) | 1,659 (13.0) | |
Muscularis propria | 0 | 9 (0.1) |
Data are presented as number (%) unless otherwise indicated..
ESD, endoscopic submucosal dissection; EGC, early gastric cancer..
*12,736 Lesions were analyzed (15 with delayed perforation and 12,721 without delayed perforation)..
A representative image of delayed perforation is shown in Fig. 2. Clinicopathological features of patients with delayed perforation are summarized in Table 2. Details of each case are described in Table 3. The median age was 71 years (range, 39 to 86 years). Male patients accounted for 66.7%. Regarding axial location, tumors were most frequently (46.7%) located in the lower third of the stomach. Regarding circumferential location, tumors were most frequently found in the lesser curvature and greater curvature (33.3% and 33.3%, respectively). The median ESD specimen size was 4.2 cm (range, 3.0 to 6.6 cm). The median procedure time was 1.0 hours (range, 0.3 to 2.3 hours). The median time to diagnosis of delayed perforation was 28.8 hours (range, 14 to 71 hours). All 15 patients with delayed perforation complained of severe abdominal pain after gastric ESD and underwent subsequent CXRs for evaluation. Fever and leukocytosis were present in 53.3% and 66.7% of patients, respectively. No cases of delayed perforation occurred in the gastric tube following esophagectomy or remnant stomach after gastrectomy.
Clinicopathological Features of Patients with Delayed Perforation after Endoscopic Submucosal Dissection.
Variable | Total (n=15) |
---|---|
Age, yr | |
Mean±SD | 70.0±11.7 |
Median (range) | 71 (39–86) |
Sex | |
Male | 10 (66.7) |
Female | 5 (33.3) |
Comorbidities | |
None | 9 (60.0) |
Diabetes mellitus | 2 (13.3) |
Hypertension | 4 (26.7) |
Axial location | |
Lower third | 7 (46.7) |
Middle third | 4 (26.7) |
Upper third | 4 (26.7) |
Remnant stomach or gastric tube | 0 |
Circumferential location | |
Anterior wall | 2 (13.4) |
Posterior wall | 3 (20.0) |
Lesser curvature | 5 (33.3) |
Greater curvature | 5 (33.3) |
Tumor shape (endoscopy) | |
Elevated | 5 (33.3) |
I | 2 (13.3) |
IIa | 3 (20.0) |
Flat/depressed/mixed | 10 (66.7) |
IIb | 1 (6.7) |
IIc | 3 (20.0) |
IIa+IIc | 6 (40.0) |
Histology | |
Adenoma | 2 (13.3) |
Differentiated-type EGC | 12 (80.0) |
Undifferentiated-type EGC | 1 (6.7) |
Tumor size (pathology), cm | |
Mean±SD | 1.6±1.0 |
Median (range) | 1.2 (0.8–3.6) |
Tumor depth | |
Mucosa | 13 (86.7) |
Submucosa | 2 (13.3) |
Specimen size (pathology), cm | |
Mean±SD | 4.5±1.3 |
Median (range) | 4.2 (3.0–6.6) |
Procedure time, hr | |
Mean±SD | 1.0±0.6 |
Median (range) | 1.0 (0.3–2.3) |
Curative resection | |
No | 1 (6.7) |
Yes | 14 (93.3) |
Post-procedural abdominal pain (yes) | 15 (100) |
Post-procedural fever (yes)* | 8 (53.3) |
Post-procedural leukocytosis (yes)† | 10 (66.7) |
Time to diagnosis, hr | |
Mean±SD | 30.0±14.7 |
Median (range) | 28.8 (14–71) |
Free air in subsequent CXRs | |
No | 3 (20.0) |
Yes | 12 (80.0) |
Size of perforation‡ | |
<1 cm | 5 (35.7) |
≥1 cm | 9 (64.3) |
Data are presented as number (%) unless otherwise indicated..
EGC, early gastric cancer; CXR, chest X-ray..
*Post-procedural fever was defined as a body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as a white blood count of 10,000/μL or higher17; ‡Perforation size could be identified in 14 patients..
Summary of the Clinical Characteristics and Treatment Outcomes of Patients with Delayed Perforation.
No. | Age, yr | Sex | Histologic type | Tumor shape | Location | Tumor size, cm | Specimen size, cm | Tumor depth | Time to diagnosis, hr | Free air in CXR | Perforation size, cm | Determination of perforation size | Treatment | Reason for surgery | Hospital days | Death |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | 80 | M | MD | IIa+IIc | Angle AW | 2.6 | 6.6 | MM | 31.8 | Yes | Non-visible in EGD | EGD | Supportive only | NA | 18 | No |
2 | 39 | F | SRC | IIc | Fundus GC | 0.8 | 3.0 | LP | 24.1 | Yes | 0.5 | CT | Supportive only | NA | 7 | No |
3 | 79 | M | MD | IIb | Antrum LC | 0.8 | 3.0 | LP | 47.0 | Yes | N/A | NA | Supportive only | NA | 11 | No |
4 | 77 | F | WD | IIa+IIc | Antrum GC | 0.8 | 4.5 | LP | 28.9 | Yes | 0.5 | EGD | Endoscopic closure with clipping | NA | 10 | No |
5 | 62 | F | WD | IIa | Antrum LC | 0.9 | 3.4 | LP | 13.7 | Yes | 0.6 | EGD | Endoscopic closure with clipping | NA | 9 | No |
6 | 60 | M | WD | IIa+IIc | Antrum GC | 0.8 | 3.3 | LP | 40.0 | Yes | 0.8 | EGD | Endoscopic closure with clipping and Neoveil packing | NA | 12 | No |
7 | 86 | M | MD | IIc | Antrum AW | 0.8 | 4.2 | LP | 24.0 | Yes | 1.0 | EGD | Clipping→primary surgical repair | Unsuccessful clipping | 22 | No |
8 | 67 | M | WD | IIc | Antrum LC | 1.8 | 4.5 | MM | 14.7 | Yes | 1.2 | EGD | Clipping→laparoscopic segmental resection | Unsuccessful clipping | 11 | No |
9 | 77 | M | HGD | IIa | HB LC | 3.2 | 6.5 | Mucosa | 34.7 | Yes | 1.2 | EGD | Clipping→open wedge resection | Unsuccessful clipping | 14 | No |
10 | 68 | M | WD | IIa+IIc | Antrum GC | 1.2 | 5.1 | MM | 27.4 | Yes | 1.4 | CT | Primary surgical repair | Rapid clinical deterioration (intractable pain) | 13 | No |
11 | 62 | M | WD | I | LB GC | 1.3 | 4.0 | MM | 20.7 | Yes | 2.0 | EGD | Laparoscopic wedge resection | Large perforation size | 11 | No |
12 | 77 | F | LGD | I | HB PW | 1.2 | 3.1 | Mucosa | 28.8 | Yes | 1.7 | CT | Open wedge resection | Large perforation size | 14 | No |
13 | 80 | M | WD | IIa+IIc | LB PW | 1.2 | 4.0 | MM | 14.4 | No | 1.8 | EGD | Primary surgical repair | Large perforation size | 12 | No |
14 | 71 | M | MD | IIa+IIc | MB LC | 3.6 | 6.4 | SM 1400 | 29.0 | No | 2.0 | CT | Total gastrectomy | Non-curative resection | 12 | No |
15 | 65 | F | MD | IIa | HB PW | 2.6 | 5.7 | SM 50 | 71.1 | No | 2.2 | CT | Primary surgical repair & left colectomy due to descending colon ischemia | Rapid clinical deterioration (septic shock) | 192 | No |
CXR, chest X-ray; M, male; F, female; MD, moderately differentiated type; SRC, signet ring cell carcinoma; WD, well-differentiated type; HGD, high-grade dysplasia; LGD, low-grade dysplasia; AW, anterior wall; GC, greater curvature; LC, lesser curvature; HB, high body; LB, low body; PW, posterior wall; MB, mid-body; MM, muscularis mucosae; LP, lamina propria; SM, submucosa; EGD, esophagogastroduodenoscopy; CT, computed tomography; NA, not applicable..
In subsequent CXRs, free air was found and diagnosis of delayed perforation could be made in 12 (80%) patients. For three patients (20%; patients #13, #14, and #15 in Table 3) without free air detected in subsequent CXRs, diagnoses of delayed perforation were finally made by abdominal CT at 14.4, 29.0, and 71.1 hours after procedure (Fig. 2). For patient #15, CXRs were taken daily because the patient complained of severe abdominal pain after ESD. As serial CXRs did not show free air, the patient was conservatively managed with fasting and antibiotics. On the third day after ESD, the patient underwent septic shock and emergency abdominal CT scan was taken, which showed free air and panperitonitis. This patient required left colectomy with end colostomy due to colonic ischemia and prolonged intensive care unit treatment.
Among the 15 patients with delayed perforation (Table 3), three patients (20%) received conservative treatment only and recovered without undergoing endoscopic or surgical procedures. Three patients (20%) were successfully managed with endoscopic closure with clipping (Fig. 3). Nine patients (60%) received surgery for repairing perforation site including three patients (patients #7, #8, and #9 in Table 3) whose initial attempt of endoscopic closure had failed. Reasons for surgery were summarized in Table 3. There was no delayed perforation-related death.
Table 4 summarizes the comparison of clinical characteristics of patients with and without free air in subsequent CXR. Among patients undergoing delayed perforation, patients without free air in subsequent CXR showed significantly lower rate of post-procedural leukocytosis compared to those with free air (0% and 83.3%, respectively).
Comparison of the Clinical Characteristics between Patients with and without Free Air in Subsequent Chest X-Ray.
Characteristic | No free air (n=3) | Free air (n=12) | p-value |
---|---|---|---|
Age, yr | 0.734 | ||
Mean±SD | 72.0±7.6 | 69.5±12.7 | |
Median (range) | 71 (65–80) | 73 (39–86) | |
Male sex | 2 (66.7) | 8 (66.7) | 1.000 |
Comorbidities (yes) | 1 (33.3) | 5 (41.7) | 1.000 |
Axial location | 0.123 | ||
Lower third | 0 | 7 (58.3) | |
Middle third | 2 (66.7) | 2 (16.7) | |
Upper third | 1 (33.3) | 3 (25.0) | |
Circumferential location | 0.154 | ||
Anterior wall | 0 | 2 (16.7) | |
Posterior wall | 2 (66.7) | 1 (8.3) | |
Lesser curvature | 1 (33.3) | 4 (33.3) | |
Greater curvature | 0 | 5 (41.7) | |
Tumor shape (endoscopy) | 1.000 | ||
Elevated | 1 (33.3) | 4 (33.3) | |
Flat/depressed/mixed | 2 (66.7) | 8 (66.7) | |
Specimen size (pathology), cm | 0.308 | ||
Mean±SD | 5.4±1.2 | 4.3±1.3 | |
Median (range) | 5.7 (4.0–6.4) | 4.1 (3.0–6.6) | |
Procedure time, hr | 0.295 | ||
Mean±SD | 1.4±0.8 | 0.9±0.6 | |
Median (range) | 1.0 (1.0–2.3) | 0.7 (0.3–2.1) | |
Post-procedural abdominal pain (yes) | 3 (100) | 12 (100) | 1.000 |
Post-procedural fever (yes)* | 2 (66.7) | 6 (50.0) | 1.000 |
Post-procedural leukocytosis (yes)† | 0 | 10 (83.3) | 0.022 |
Time to diagnosis, hr | 0.734 | ||
Mean±SD | 38.2±29.5 | 28.0±9.7 | |
Median (range) | 29 (14–71) | 28 (14–47) | |
Size of perforation‡ | 0.258 | ||
<1 cm | 0 | 5 (45.5) | |
≥1 cm | 3 (100) | 6 (54.5) | |
Operator | 1.000 | ||
A | 0 | 3 (25.0) | |
B | 2 (66.7) | 5 (41.6) | |
C | 0 | 2 (16.7) | |
D | 1 (33.3) | 2 (16.7) | |
Treatment method | 0.229 | ||
Non-surgery | 0 | 6 (50.0) | |
Surgery | 3 (100) | 6 (50.0) |
Data are presented as number (%) unless otherwise indicated..
*Post-procedural fever was defined as a body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as white a blood count of 10,000/μL or higher17; ‡The perforation size could be identified in 14 patients..
Among patients undergoing delayed perforation, clinicopathological characteristics of patients undergoing non-surgical and surgical treatments are compared in Table 5. The mean size of tumor in pathology was smaller in the non-surgery group than in the surgery group (1.1±0.7 cm and 1.9±1.0 cm, respectively). In the non-surgery group (patients #1, #2, #3, #4, #5, and #6 in Table 3), the size of the perforation hole was smaller than 1 cm in all patients. In contrast, all patients in the surgery group had perforation holes no smaller than 1 cm. Size of a perforation hole smaller than 1 cm was significantly associated with successful non-surgical treatment for delayed perforation.
Comparison of the Clinicopathological Characteristics between Patients Improved with Non-Surgical Treatments Only or with Surgery.
Characteristic | Non-surgery (n=6) | Surgery (n=9) | p-value |
---|---|---|---|
Age, yr | 0.317 | ||
Mean±SD | 66.2±15.9 | 72.6±7.9 | |
Median (range) | 69.5 (39–80) | 71.0 (62–86) | |
Male sex | 3 (50.0) | 7 (77.8) | 0.329 |
Comorbidities (yes) | 2 (33.3) | 4 (44.4) | 1.000 |
Axial location | 0.513 | ||
Lower third | 4 (66.6) | 3 (33.3) | |
Middle third | 1 (16.7) | 3 (33.3) | |
Upper third | 1 (16.7) | 3 (33.3) | |
Circumferential location | 0.461 | ||
Anterior wall | 1 (16.7) | 1 (11.1) | |
Posterior wall | 0 | 3 (33.3) | |
Lesser curvature | 2 (33.3) | 3 (33.3) | |
Greater curvature | 3 (50.0) | 2 (22.3) | |
Tumor shape (endoscopy) | 0.580 | ||
Elevated | 1 (16.7) | 4 (44.4) | |
Flat/depressed/mixed | 5 (83.3) | 5 (55.6) | |
Histology | 0.301 | ||
Adenoma | 0 | 2 (22.2) | |
Differentiated-type EGC | 5 (83.3) | 7 (77.8) | |
Undifferentiated-type EGC | 1 (16.7) | 0 | |
Tumor size (pathology), cm | 0.033 | ||
Mean±SD | 1.1±0.7 | 1.9±1.0 | |
Median (range) | 0.8 (0.8–2.6) | 1.3 (0.8–3.6) | |
Tumor depth | 0.486 | ||
Mucosa | 6 (100) | 7 (77.8) | |
Submucosa | 0 | 2 (22.2) | |
Specimen size (pathology), cm | 0.191 | ||
Mean±SD | 4.0±1.4 | 4.8±1.2 | |
Median (range) | 3.4 (3.0–6.6) | 4.5 (3.1–6.5) | |
Procedure time, hr | 0.188 | ||
Mean±SD | 0.8±0.4 | 1.2±0.7 | |
Median (range) | 0.6 (0.3–1.4) | 1.0 (0.4–2.3) | |
Post-procedural abdominal pain (yes) | 6 (100) | 9 (100) | - |
Post-procedural fever (yes)* | 2 (33.3) | 6 (66.7) | 0.315 |
Post-procedural leukocytosis (yes)† | 4 (66.7) | 6 (66.7) | 1.000 |
Time to diagnosis, hr | 0.529 | ||
Mean±SD | 30.9±11.7 | 29.4±17.0 | |
Median (range) | 30.4 (14–47) | 27.4 (14–71) | |
Free air in subsequent CXRs (yes) | 6 (100) | 6 (66.7) | 0.229 |
Size of perforation‡ | <0.001 | ||
<1 cm | 5 (100) | 0 | |
≥1 cm | 0 | 9 (100) |
Data are presented as number (%) unless otherwise indicated..
EGC, early gastric cancer; CXR, chest X-ray..
*Post-procedural fever was defined as body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as white blood count of 10,000/μL or higher17; ‡The perforation size could be identified in 14 patients..
Delayed perforation is a rare but potentially life-threatening adverse event of gastric ESD with a reported incidence ranging from 0.1% to 0.45%.4-8 Due to its rarity, clinical features and outcomes of patients with delayed perforation have been scarcely reported and its optimal treatment strategy remains uncertain. The main limitation of previous studies on delayed perforation was that they were small case series without a control group (Kang et al.18: two cases, Homma et al.19: one case, and Yano et al.6: four cases). In the present study, we analyzed 15 patients who experienced delayed perforation after gastric ESD and compared their results with control group without delayed perforation. To the best of our knowledge, this is the largest study on delayed perforation after gastric ESD. In this study, we found that free air in the subsequent CXR was persistently absent in 20% of delayed perforation cases and that leukocytosis was significantly less frequent among patients without free air than those with free air. We also found that 40% of patients improved with non-surgical treatments only and small size of perforation (<1 cm) was significantly associated with successful non-surgical treatment.
The risk factors for delayed perforation after gastric ESD are not well known because of its rarity. In the present study, patients with delayed perforation were significantly older than those without delayed perforation. In contrast, Yamamoto et al.20 and Suzuki et al.7 reported that there were no differences in the proportion of patients older than 70 years between patients with and without delayed perforation. Meanwhile, Suzuki et al.7 reported that patients with gastric tube following esophagectomy were significantly associated with delayed perforation. Yabuuchi et al.21 reported that delayed perforation occurred in 1.3% of patients with remnant stomach, which was higher than those reported in previous studies (0.1% to 0.45%). In this study, however, no cases of delayed perforation occurred in the gastric tube following esophagectomy or remnant stomach after gastrectomy. Further large studies are required to evaluate whether old age or postoperative stomach is associated with the occurrence of delayed perforation after gastric ESD.
It is important to note that free air may not always be present in follow-up CXRs after ESD in patients with delayed perforation. In a recent study, the sensitivity of erect CXR for detecting gastric perforation was only 75%.22 In a study by Yamamoto et al.20 including five patients with delayed perforation, CXR performed in two patients did not show free air in either case. In the present study, free air in CXR was detected in only 80% (12/15) of patients and three patients were persistently negative for free air in CXR (20%, 3/15). Interestingly, patients #13 and #15 in Table 3 showed strikingly different clinical course although their CXRs were both negative for free air. In patient #15, the abdominal pain started approximately 24 hours after ESD. However, diagnosis of delayed perforation was confirmed with CT at 71.1 hours after ESD when septic shock had already occurred. This patient required prolonged intensive care unit treatment and left colectomy due to colonic ischemia. Meanwhile, diagnosis of delayed perforation was made at 14.4 hours with abdominal CT in patient #13 whose abdominal pain started six hours after ESD (Fig. 2). This patient underwent primary repair without further adverse events. As such, absence of free air in CXR may delay the diagnosis of delayed perforation and complicate the clinical course. Therefore, it is imperative not to depend on the CXR alone for the diagnosis of delayed perforation to avoid delay in diagnosis and to improve clinical outcomes. Keeping a high level of clinical suspicion and performing CT scan are required if a patient develops symptoms and signs of peritonitis after ESD even when follow-up CXRs are persistently negative for free air. In addition, approximately 50% of patients with delayed perforation did not develop fever or leukocytosis in the present study. Furthermore, leukocytosis was significantly less frequent among patients without free air than those with free air, which again emphasized the importance of a clinical suspicion in diagnosis of delayed perforation. Similarly, Yamamoto et al.20 reported that more than half of their patients with delayed perforation were asymptomatic. Clinicians should be careful not to rule out the possibility of delayed perforation based on the absence of fever or leukocytosis.
The optimal treatment strategy for delayed perforation remains controversial. In the case series of Hanaoka et al.,4 83.3% of patients (5/6) required surgery due to severe peritonitis or worsening clinical course. However, Yamamoto et al.20 reported that all five patients with delayed perforation avoided surgery because the perforation was small (5 mm or less in all cases) and the diagnosis was made early (all cases within 24 hours after ESD). In the present study, size of a perforation hole smaller than 1 cm was significantly associated with successful non-surgical treatment for delayed perforation. In all three patients whose initial attempt of endoscopic closure failed (patients #7, #8, and #9 in Table 3), the size of perforation hole was equal to or larger than 1 cm. As delayed perforation may be caused by blunt damage due to ischemic change rather than sharp laceration due to electrosurgical knife, the size of the necrotic area around the perforation hole can be large. This may make endoscopic closure with clipping difficult and can result in late dehiscence even after successful clipping, especially when the size of perforation hole is large. Based on these results, we suggest that endoscopic closure might be considered as the initial treatment modality for delayed perforation if the size of perforation hole is smaller than 1 cm. Also, in patients who do not improve after endoscopic treatment and in those with severe ongoing peritonitis, surgery should not be delayed because duration of symptoms is known to be a factor influencing the prognosis after surgery for peptic ulcer perforation.23 Another potential treatment option for delayed perforation was shielding with polyglycolic acid sheets (Neoveil; Gunze Co., Osaka, Japan). Polyglycolic acid sheets are composed of bioabsorbable soft-tissue reinforcement material and can be used to cover the perforation site. Ono et al.24 reported a case of delayed perforation following gastric ESD which was successfully treated with shielding with polyglycolic acid sheets after failure of endoclip closure. This shielding method with polyglycolic acid sheets can be considered when the size of perforation hole is large or the ESD-induced ulcer floor is fragile, which makes endoscopic closure with clipping difficult.
This study has several limitations. First, this was a retrospective study performed at a single tertiary referral center. Second, the number of study population was small. Further large-scale studies are required.
In conclusion, delayed perforation is a rare but potentially life-threatening adverse event of gastric ESD which requires active evaluation for diagnosis even when free air in the CXR is persistently absent. Endoscopic closure might be considered as an initial treatment modality for delayed perforation if the size of perforation hole is smaller than 1 cm.
No potential conflict of interest relevant to this article was reported.
Study concept and design: T.S.K., B.H.M. Data acquisition: T.S.K., B.H.M., Y.W.M., H.L., P.L.R., J.J.K., J.H.L. Data analysis and interpretation: T.S.K., B.H.M. Drafting of the manuscript: T.S.K., B.H.M. Critical revision of the manuscript for important intellectual content: T.S.K., B.H.M., Y.W.M., H.L., P.L.R., J.J.K., J.H.L. Statistical analysis: T.S.K., B.H.M. Administrative, technical, or material support; study supervision: J.H.L. Approval of final manuscript: all authors.
Comparison of the Clinicopathological Characteristics of Patients with and without Delayed Perforation in Whole Study Population
Characteristic | Delayed perforation (n=15) | Without delayed perforation (n=11,516) | p-value |
---|---|---|---|
Age, yr | 0.027 | ||
Mean±SD | 70.0±11.7 | 64.4±9.9 | |
Median (range) | 71 (39–86) | 65 (20-98) | |
Sex | 0.562 | ||
Male | 10 (66.7) | 8,464 (73.5) | |
Female | 5 (33.3) | 3,052 (26.5) | |
Diabetes mellitus | 0.383 | ||
No | 13 (86.7) | 8,605 (74.7) | |
Yes | 2 (13.3) | 2,911 (25.3) | |
Hypertension | 0.127 | ||
No | 11 (73.3) | 6,087 (52.9) | |
Yes | 4 (26.7) | 5,429 (47.1) | |
Location* | 0.188 | ||
Lower third | 7 (46.6) | 7,348 (57.7) | |
Middle third | 4 (26.7) | 4,143 (32.6) | |
Upper third | 4 (26.7) | 1,082 (8.5) | |
Remnant stomach | 0 | 101 (0.8) | |
Gastric tube | 0 | 47 (0.4) | |
ESD pathology* | 0.203 | ||
Adenoma | 2 (13.3) | 3,788 (29.8) | |
Differentiated-type EGC | 12 (80.0) | 8,551 (67.2) | |
Undifferentiated-type EGC | 1 (6.7) | 382 (3.0) | |
Tumor size (pathology), mm* | 0.968 | ||
Mean±SD | 15.7±9.6 | 15.4±9.8 | |
Median (range) | 12.0 (8.0–36.0) | 13.0 (0.1–110.0) | |
Tumor depth* | 1.000 | ||
Mucosa | 13 (86.7) | 11,053 (86.9) | |
Submucosa | 2 (13.3) | 1,659 (13.0) | |
Muscularis propria | 0 | 9 (0.1) |
Data are presented as number (%) unless otherwise indicated.
ESD, endoscopic submucosal dissection; EGC, early gastric cancer.
*12,736 Lesions were analyzed (15 with delayed perforation and 12,721 without delayed perforation).
Clinicopathological Features of Patients with Delayed Perforation after Endoscopic Submucosal Dissection
Variable | Total (n=15) |
---|---|
Age, yr | |
Mean±SD | 70.0±11.7 |
Median (range) | 71 (39–86) |
Sex | |
Male | 10 (66.7) |
Female | 5 (33.3) |
Comorbidities | |
None | 9 (60.0) |
Diabetes mellitus | 2 (13.3) |
Hypertension | 4 (26.7) |
Axial location | |
Lower third | 7 (46.7) |
Middle third | 4 (26.7) |
Upper third | 4 (26.7) |
Remnant stomach or gastric tube | 0 |
Circumferential location | |
Anterior wall | 2 (13.4) |
Posterior wall | 3 (20.0) |
Lesser curvature | 5 (33.3) |
Greater curvature | 5 (33.3) |
Tumor shape (endoscopy) | |
Elevated | 5 (33.3) |
I | 2 (13.3) |
IIa | 3 (20.0) |
Flat/depressed/mixed | 10 (66.7) |
IIb | 1 (6.7) |
IIc | 3 (20.0) |
IIa+IIc | 6 (40.0) |
Histology | |
Adenoma | 2 (13.3) |
Differentiated-type EGC | 12 (80.0) |
Undifferentiated-type EGC | 1 (6.7) |
Tumor size (pathology), cm | |
Mean±SD | 1.6±1.0 |
Median (range) | 1.2 (0.8–3.6) |
Tumor depth | |
Mucosa | 13 (86.7) |
Submucosa | 2 (13.3) |
Specimen size (pathology), cm | |
Mean±SD | 4.5±1.3 |
Median (range) | 4.2 (3.0–6.6) |
Procedure time, hr | |
Mean±SD | 1.0±0.6 |
Median (range) | 1.0 (0.3–2.3) |
Curative resection | |
No | 1 (6.7) |
Yes | 14 (93.3) |
Post-procedural abdominal pain (yes) | 15 (100) |
Post-procedural fever (yes)* | 8 (53.3) |
Post-procedural leukocytosis (yes)† | 10 (66.7) |
Time to diagnosis, hr | |
Mean±SD | 30.0±14.7 |
Median (range) | 28.8 (14–71) |
Free air in subsequent CXRs | |
No | 3 (20.0) |
Yes | 12 (80.0) |
Size of perforation‡ | |
<1 cm | 5 (35.7) |
≥1 cm | 9 (64.3) |
Data are presented as number (%) unless otherwise indicated.
EGC, early gastric cancer; CXR, chest X-ray.
*Post-procedural fever was defined as a body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as a white blood count of 10,000/μL or higher17; ‡Perforation size could be identified in 14 patients.
Summary of the Clinical Characteristics and Treatment Outcomes of Patients with Delayed Perforation
No. | Age, yr | Sex | Histologic type | Tumor shape | Location | Tumor size, cm | Specimen size, cm | Tumor depth | Time to diagnosis, hr | Free air in CXR | Perforation size, cm | Determination of perforation size | Treatment | Reason for surgery | Hospital days | Death |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | 80 | M | MD | IIa+IIc | Angle AW | 2.6 | 6.6 | MM | 31.8 | Yes | Non-visible in EGD | EGD | Supportive only | NA | 18 | No |
2 | 39 | F | SRC | IIc | Fundus GC | 0.8 | 3.0 | LP | 24.1 | Yes | 0.5 | CT | Supportive only | NA | 7 | No |
3 | 79 | M | MD | IIb | Antrum LC | 0.8 | 3.0 | LP | 47.0 | Yes | N/A | NA | Supportive only | NA | 11 | No |
4 | 77 | F | WD | IIa+IIc | Antrum GC | 0.8 | 4.5 | LP | 28.9 | Yes | 0.5 | EGD | Endoscopic closure with clipping | NA | 10 | No |
5 | 62 | F | WD | IIa | Antrum LC | 0.9 | 3.4 | LP | 13.7 | Yes | 0.6 | EGD | Endoscopic closure with clipping | NA | 9 | No |
6 | 60 | M | WD | IIa+IIc | Antrum GC | 0.8 | 3.3 | LP | 40.0 | Yes | 0.8 | EGD | Endoscopic closure with clipping and Neoveil packing | NA | 12 | No |
7 | 86 | M | MD | IIc | Antrum AW | 0.8 | 4.2 | LP | 24.0 | Yes | 1.0 | EGD | Clipping→primary surgical repair | Unsuccessful clipping | 22 | No |
8 | 67 | M | WD | IIc | Antrum LC | 1.8 | 4.5 | MM | 14.7 | Yes | 1.2 | EGD | Clipping→laparoscopic segmental resection | Unsuccessful clipping | 11 | No |
9 | 77 | M | HGD | IIa | HB LC | 3.2 | 6.5 | Mucosa | 34.7 | Yes | 1.2 | EGD | Clipping→open wedge resection | Unsuccessful clipping | 14 | No |
10 | 68 | M | WD | IIa+IIc | Antrum GC | 1.2 | 5.1 | MM | 27.4 | Yes | 1.4 | CT | Primary surgical repair | Rapid clinical deterioration (intractable pain) | 13 | No |
11 | 62 | M | WD | I | LB GC | 1.3 | 4.0 | MM | 20.7 | Yes | 2.0 | EGD | Laparoscopic wedge resection | Large perforation size | 11 | No |
12 | 77 | F | LGD | I | HB PW | 1.2 | 3.1 | Mucosa | 28.8 | Yes | 1.7 | CT | Open wedge resection | Large perforation size | 14 | No |
13 | 80 | M | WD | IIa+IIc | LB PW | 1.2 | 4.0 | MM | 14.4 | No | 1.8 | EGD | Primary surgical repair | Large perforation size | 12 | No |
14 | 71 | M | MD | IIa+IIc | MB LC | 3.6 | 6.4 | SM 1400 | 29.0 | No | 2.0 | CT | Total gastrectomy | Non-curative resection | 12 | No |
15 | 65 | F | MD | IIa | HB PW | 2.6 | 5.7 | SM 50 | 71.1 | No | 2.2 | CT | Primary surgical repair & left colectomy due to descending colon ischemia | Rapid clinical deterioration (septic shock) | 192 | No |
CXR, chest X-ray; M, male; F, female; MD, moderately differentiated type; SRC, signet ring cell carcinoma; WD, well-differentiated type; HGD, high-grade dysplasia; LGD, low-grade dysplasia; AW, anterior wall; GC, greater curvature; LC, lesser curvature; HB, high body; LB, low body; PW, posterior wall; MB, mid-body; MM, muscularis mucosae; LP, lamina propria; SM, submucosa; EGD, esophagogastroduodenoscopy; CT, computed tomography; NA, not applicable.
Comparison of the Clinical Characteristics between Patients with and without Free Air in Subsequent Chest X-Ray
Characteristic | No free air (n=3) | Free air (n=12) | p-value |
---|---|---|---|
Age, yr | 0.734 | ||
Mean±SD | 72.0±7.6 | 69.5±12.7 | |
Median (range) | 71 (65–80) | 73 (39–86) | |
Male sex | 2 (66.7) | 8 (66.7) | 1.000 |
Comorbidities (yes) | 1 (33.3) | 5 (41.7) | 1.000 |
Axial location | 0.123 | ||
Lower third | 0 | 7 (58.3) | |
Middle third | 2 (66.7) | 2 (16.7) | |
Upper third | 1 (33.3) | 3 (25.0) | |
Circumferential location | 0.154 | ||
Anterior wall | 0 | 2 (16.7) | |
Posterior wall | 2 (66.7) | 1 (8.3) | |
Lesser curvature | 1 (33.3) | 4 (33.3) | |
Greater curvature | 0 | 5 (41.7) | |
Tumor shape (endoscopy) | 1.000 | ||
Elevated | 1 (33.3) | 4 (33.3) | |
Flat/depressed/mixed | 2 (66.7) | 8 (66.7) | |
Specimen size (pathology), cm | 0.308 | ||
Mean±SD | 5.4±1.2 | 4.3±1.3 | |
Median (range) | 5.7 (4.0–6.4) | 4.1 (3.0–6.6) | |
Procedure time, hr | 0.295 | ||
Mean±SD | 1.4±0.8 | 0.9±0.6 | |
Median (range) | 1.0 (1.0–2.3) | 0.7 (0.3–2.1) | |
Post-procedural abdominal pain (yes) | 3 (100) | 12 (100) | 1.000 |
Post-procedural fever (yes)* | 2 (66.7) | 6 (50.0) | 1.000 |
Post-procedural leukocytosis (yes)† | 0 | 10 (83.3) | 0.022 |
Time to diagnosis, hr | 0.734 | ||
Mean±SD | 38.2±29.5 | 28.0±9.7 | |
Median (range) | 29 (14–71) | 28 (14–47) | |
Size of perforation‡ | 0.258 | ||
<1 cm | 0 | 5 (45.5) | |
≥1 cm | 3 (100) | 6 (54.5) | |
Operator | 1.000 | ||
A | 0 | 3 (25.0) | |
B | 2 (66.7) | 5 (41.6) | |
C | 0 | 2 (16.7) | |
D | 1 (33.3) | 2 (16.7) | |
Treatment method | 0.229 | ||
Non-surgery | 0 | 6 (50.0) | |
Surgery | 3 (100) | 6 (50.0) |
Data are presented as number (%) unless otherwise indicated.
*Post-procedural fever was defined as a body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as white a blood count of 10,000/μL or higher17; ‡The perforation size could be identified in 14 patients.
Comparison of the Clinicopathological Characteristics between Patients Improved with Non-Surgical Treatments Only or with Surgery
Characteristic | Non-surgery (n=6) | Surgery (n=9) | p-value |
---|---|---|---|
Age, yr | 0.317 | ||
Mean±SD | 66.2±15.9 | 72.6±7.9 | |
Median (range) | 69.5 (39–80) | 71.0 (62–86) | |
Male sex | 3 (50.0) | 7 (77.8) | 0.329 |
Comorbidities (yes) | 2 (33.3) | 4 (44.4) | 1.000 |
Axial location | 0.513 | ||
Lower third | 4 (66.6) | 3 (33.3) | |
Middle third | 1 (16.7) | 3 (33.3) | |
Upper third | 1 (16.7) | 3 (33.3) | |
Circumferential location | 0.461 | ||
Anterior wall | 1 (16.7) | 1 (11.1) | |
Posterior wall | 0 | 3 (33.3) | |
Lesser curvature | 2 (33.3) | 3 (33.3) | |
Greater curvature | 3 (50.0) | 2 (22.3) | |
Tumor shape (endoscopy) | 0.580 | ||
Elevated | 1 (16.7) | 4 (44.4) | |
Flat/depressed/mixed | 5 (83.3) | 5 (55.6) | |
Histology | 0.301 | ||
Adenoma | 0 | 2 (22.2) | |
Differentiated-type EGC | 5 (83.3) | 7 (77.8) | |
Undifferentiated-type EGC | 1 (16.7) | 0 | |
Tumor size (pathology), cm | 0.033 | ||
Mean±SD | 1.1±0.7 | 1.9±1.0 | |
Median (range) | 0.8 (0.8–2.6) | 1.3 (0.8–3.6) | |
Tumor depth | 0.486 | ||
Mucosa | 6 (100) | 7 (77.8) | |
Submucosa | 0 | 2 (22.2) | |
Specimen size (pathology), cm | 0.191 | ||
Mean±SD | 4.0±1.4 | 4.8±1.2 | |
Median (range) | 3.4 (3.0–6.6) | 4.5 (3.1–6.5) | |
Procedure time, hr | 0.188 | ||
Mean±SD | 0.8±0.4 | 1.2±0.7 | |
Median (range) | 0.6 (0.3–1.4) | 1.0 (0.4–2.3) | |
Post-procedural abdominal pain (yes) | 6 (100) | 9 (100) | - |
Post-procedural fever (yes)* | 2 (33.3) | 6 (66.7) | 0.315 |
Post-procedural leukocytosis (yes)† | 4 (66.7) | 6 (66.7) | 1.000 |
Time to diagnosis, hr | 0.529 | ||
Mean±SD | 30.9±11.7 | 29.4±17.0 | |
Median (range) | 30.4 (14–47) | 27.4 (14–71) | |
Free air in subsequent CXRs (yes) | 6 (100) | 6 (66.7) | 0.229 |
Size of perforation‡ | <0.001 | ||
<1 cm | 5 (100) | 0 | |
≥1 cm | 0 | 9 (100) |
Data are presented as number (%) unless otherwise indicated.
EGC, early gastric cancer; CXR, chest X-ray.
*Post-procedural fever was defined as body temperature of 37.8°C or higher16; †Post-procedural leukocytosis was defined as white blood count of 10,000/μL or higher17; ‡The perforation size could be identified in 14 patients.