Endoscopic submucosal dissection for colorectal neoplasia during the clinical learning curve

The efficacy of colorectal endoscopic submucosal dissection (ESD) has been reported mainly from Japanese referral centers. However, ESD is technically difficult and associated with a higher risk of adverse events than endoscopic mucosal resection, especially for novices performing colorectal ESD with little experience in gastric ESD. The current study evaluated the results of colorectal ESD during the clinical learning curve by retrospectively examining the results of colorectal ESD performed by four endoscopists who had experience with fewer than five cases of gastric ESD.

The study retrospectively investigated the first 20 cases managed by each endoscopist, for a total of 80 cases. The main outcome measurements were procedural time, en bloc resection rate with tumor-free margins (R0 resection rate), and adverse events rate. From among clinicopathologic characteristics, factors that affected main outcome measurements were identified.

Of the 80 cases (56 colonic and 24 rectal lesions; 44 granular laterally spreading tumors (LSTs) and 23 nongranular LSTs, 5 depressed, and 8 protruding), 54 cases (67.5%) had resection using a standard tip-type knife, and 26 cases (32.5%) had resection using a small scissors-type knife. The mean tumor diameter was 34.9 ± 14.1 mm, and the mean procedural time was 108.8 ± 53.4 min. The resection in 75 cases (93.8%) was performed en bloc, and the R0 resection rate was 75% (60/80). Perforation occurred in six cases (7.5%) and postoperative hemorrhage in three cases (3.8%). Multivariate analyses showed that colonic lesions and larger lesions (≥40 mm) were significantly associated with prolonged procedural time (≥90 min). Use of the scissors-type knife was significantly associated with a higher R0 resection rate. Perforation occurred only in colonic lesions.

For novices in colorectal ESD, beginning with rectal and smaller lesions may be advisable. Also, using scissors-type knives may increase the R0 resection rate.