BACKGROUND AND AIMS: Despite advances in endoscopic submucosal dissection (ESD), perforation can still occur. The purpose of this study is to determine the clinical course and effectiveness of endoscopic closure in addition to the clinicopathologic features related to perforation. METHODS: A total of 935 lesions in 900 consecutive patients between February 1998 and February 2013 underwent ESD for colorectal tumors at our institution. We studied the clinical course and histologic features of perforation through a matched case-control study that included 24 patients with intraprocedural perforation and 240 matched patients without perforation as a control group. Endoscopic closure by using through-the-scope endoclips was attempted in all cases of intraprocedural perforations immediately after perforation was recognized during the procedure. RESULTS: Perforation occurred in 25 cases (2.7%), including 24 intraprocedural perforation and 1 delayed perforation. All but 1 patient with intraprocedural perforation was conservatively managed by endoscopic closure. One patient with unsuccessful endoscopic closure required emergency surgery. Analysis of clinical courses revealed statistically significant differences (P < .01) between the patients with perforation and the case-controlled, nonperforation patients in total procedure time, white blood cell count, and level of serum C-reactive protein on the day after the procedure, admission period, and fasting period. Both location (P = .027) and submucosal fibrosis (P = .04) of the lesion were significantly associated with perforation. Multivariate analysis revealed that fibrosis was a significant risk factor associated with perforation (odds ratio 2.86; 95% confidence interval, 1.03-7.90). CONCLUSIONS: Endoscopic closure allows effective nonsurgical management in cases of intraprocedural perforation during ESD.
BACKGROUND AND AIMS: Despite advances in endoscopic submucosal dissection (ESD), perforation can still occur. The purpose of this study is to determine the clinical course and effectiveness of endoscopic closure in addition to the clinicopathologic features related to perforation. METHODS: A total of 935 lesions in 900 consecutive patients between February 1998 and February 2013 underwent ESD for colorectal tumors at our institution. We studied the clinical course and histologic features of perforation through a matched case-control study that included 24 patients with intraprocedural perforation and 240 matched patients without perforation as a control group. Endoscopic closure by using through-the-scope endoclips was attempted in all cases of intraprocedural perforations immediately after perforation was recognized during the procedure. RESULTS: Perforation occurred in 25 cases (2.7%), including 24 intraprocedural perforation and 1 delayed perforation. All but 1 patient with intraprocedural perforation was conservatively managed by endoscopic closure. One patient with unsuccessful endoscopic closure required emergency surgery. Analysis of clinical courses revealed statistically significant differences (P < .01) between the patients with perforation and the case-controlled, nonperforation patients in total procedure time, white blood cell count, and level of serum C-reactive protein on the day after the procedure, admission period, and fasting period. Both location (P = .027) and submucosal fibrosis (P = .04) of the lesion were significantly associated with perforation. Multivariate analysis revealed that fibrosis was a significant risk factor associated with perforation (odds ratio 2.86; 95% confidence interval, 1.03-7.90). CONCLUSIONS: Endoscopic closure allows effective nonsurgical management in cases of intraprocedural perforation during ESD.