Development of colorectal endoscopic submucosal dissection

摘要

In Japan, where gastric cancer incidence is remarkably high, many efforts had been made for its early detection and treatment. Since conventional endoscopic mucosal resection (EMR) was not reliable enough, Hosokawa developed a prototype of insulated tip knife in the middle 1990s and Ono et al.1 established a technique of safe mucosal incision around the lesion in the late 1990s at the National Cancer Center. Meanwhile, Yamamoto et al.2 at the Jichi Medical University developed a new technique of making safe mucosal incision even with a sharp needle knife by creating a thick submucosal fluid cushion using sodium hyaluronate. Our group at the University of Tokyo developed a method of using a slightly protruded a tip of thin-type snare as a knife, since we thought that a short knife with a blunt tip could be used for making safe mucosal incision without perforation.3 The resection technique with mucosal incision and submucosal dissection was first established in the stomach by the early 2000s. At the same time, Gotoda et al. investigated a large number of surgically treated early gastric cancer cases and clarified the low-risk conditions of early gastric cancer of lymph node metastasis. However, in the colon, the risk of perforation was considered extremely high due to the thin and soft colonic wall, and treatment by incision and dissection was regarded as contraindication. In addition, the colonic mucosa is thin enough to provide good mucosal lifting and soft protrusions are easier to snare than those of the stomach. And most of the lesions were adenomas or mucosal cancers, therefore, piecemeal resection was widely accepted even for large colorectal lesions at that time. Furthermore, unlike the stomach, there is almost no functional impairment after colonic surgery, and since laparoscopic surgery can be performed smoothly, most people agreed that there is no problem with laparoscopic surgery if it cannot be removed by conventional EMR. In spite of the above mentioned situation, we felt that it was unacceptable to send the patients with apparent intramucosal cancer or adenoma to surgery just because the snaring is not available, thus we started our challenge to change this situation. Yamamoto et al.4 and Gotoda et al.5 had already reported the resection of rectal tumor by incising the surrounding mucosa with a knife in 1999, but there were no reports of incision and dissection in the colon. However, we thought that we could perform the procedure in the colon without any problem if we used a thin-type snare that could keep the tip short, so we started with resection of a little over 1 cm of 0-IIc intramucosal cancer in the sigmoid colon and gradually expanded the resection target. The turning point was the resection of a large elevated lesion of more than 5 cm in the rectosigmoid colon.6 Since the maneuverability was poor due to the bended location, I made a perforation, but the resection was completed after clipping. A perforation in the colon meant emergency surgery at that time, but since we had been treating the lesion in a clean condition after washing the area with a water jet, and since the perforation was completely closed, our surgeon agreed to treat the patient conservatively. After that experience, we began to actively treat even larger lesions by incision and dissection. However, it was still all headwinds and was strongly criticized by the academic community. But, our group and Yamamoto's group at Jichi Medical University thought that we should aim for complete en bloc resection from the beginning, because preoperative diagnosis is not 100% even with the progress of magnifying endoscopy and piecemeal resection increases a risk of uncertain histological diagnosis and local recurrence. And of course, retreatment becomes very difficult after recurrence, thus we thought that we should aim for a complete en bloc resection even in the colon. Most of the criticism of colorectal ESD had been against its risks. This was though