Perforation occurred in 17 cases in the control group, with 12 cases managed conservatively and five requiring emergency surgery. Rates of postoperative bleeding in the residual/locally recurrent group and control group were 0% (0/34) and 2.6% (10/384), respectively. Postoperative LY2109761 concentration bleeding in the control group could be managed conservatively using endoclips. No case of recurrence was observed in the control group, but one case was observed in the residual/locally recurrent group. This case had an unclear lateral margin (intramucosal cancer) on histological

evaluation. The recurrent case was detected by progressive T2 cancer 17 months after ESD. The patient underwent laparoscopic resection and did not display local lymph node or distant metastases. The present study compared residual/locally recurrent lesions with primary lesions in terms of the technical feasibility, safety and efficacy of colorectal ESD. En bloc resection was satisfactorily achieved for every patient in the residual/locally recurrent group (34/34, 100%), representing a higher rate than that reported in some previous studies (80–98.6%).4,15,28 Although the rate of R0 resection was higher in the control group than in the residual/locally recurrent group, the rate of curative resection was higher in

the residual/locally recurrent group than Isotretinoin in the control group, probably www.selleckchem.com/products/rxdx-106-cep-40783.html because indications for ESD in the residual/locally recurrent group were only adenoma or intramucosal cancer in histological evaluation during previous therapy. Previous histological evaluation is very important in the treatment of residual/locally recurrent lesions. With the advent of ESD, curative endoscopic treatment has become possible for

lesions, regardless of tumor size, if histological reports from previous therapy indicate adenoma or intramucosal carcinoma. However, in cases with submucosal cancer invasion treated with piecemeal EMR, submucosal cancer infiltration cannot be diagnosed precisely by histological evaluation at the previous endoscopic therapy. If previous histological reports suggest submucosal invasive cancer, surgical resection with lymphadenectomy is indicated instead of ESD. Despite the smaller resected specimen size, ESD for the residual/locally recurrent group is technically more difficult. This is reflected in the study results, with higher perforation rate and procedure duration in the residual/locally recurrent group than in the control group. We believe that this was attributable to severe fibrosis in these lesions. When a lesion shows severe fibrosis, direct identification of the submucosa is difficult due to insufficient injection, making differentiation from the muscularis propria difficult.