Endoscopic Submucosal Dissection (Esd) Method For Colorectal Laterally Spreading Tumor Larger Than 20 Mm In Diameter

Recently endoscopic submucosal dissection (ESD) for early gastric cancer has been performed commonly at many institutions in Japan. As colorectal wall is very thin, ESD tends to cause the high incidence of complication such as perforation, however, several institutions try to apply ESD also for colorectal neoplasia. The Aim of this study is to clarify the clinical usefulness and safety of ESD for large colorectal tumor. Materials and Methods: We performed ESD for 33 cases of large colorectal tumor (so-called LST: laterally spreading tumor; Kudo et al, Endoscopy 1993) at Hiroshima University Hospital until August in 2004. For ESD, Hook, Flex or IT knife and transparent disposable distal attachment were used. Glycerin (10%) was injected to surrounding submucosa of the lesions for incision around the lesion. Sodium hyaluronate with indigo carmine was used for submucosal injection to practice submucosal dissection. As high-frequency generator, ICC 200 (ERBE, Tubimgen, Germany) was used, which is furnished not only with end-cut modes for incision but also with multiple coagulation modes. We use the generator as follows: endo-cut mode effect 2, 120 W for mucosal incision; forced mode, 40 W for submucosal dissection and soft mode, 50 w for hemostasis. Using these lesions, en bloc resection rate, operation time, and complication were analyzed. Results: Cases consisted of 11 carcinomas (m 9 cases, sm 2 cases) and 12 adenomas. Seventeen cases located at rectum and 16 cases at colon. Average diameter of lesions was 29.5 ± 12.5 mm. Average operation time was 78.2 ± 51.2 min (35∼160 min). Rate of histological complete en bloc resection was 78.8% (26/33). Reasons of incomplete or piecemeal resection were complication or existence of submucosal fibrosis. Complication was perforation occurred in 6 (18.2%) of 33 cases. Two were distinct perforation at import period of ESD method using IT knife at Hiroshima University Hospital, which were surgically operated. Four cases were all microperforation using Hook knife and could be treated conservatively. Bleeding after ESD was not seen in all cases. No residual tumor was detected in follow-up examination in all cases. Conclusions: ESD is a useful new method for endoscopic en bloc resection of large lesion. Incidence of perforation in ESD should be reduced by adequate choice of device and technical/instrumental progress. Also, we have to clarify the correct indication of ESD for colorectal lesion in the near future.