Second-look endoscopy findings after endoscopic submucosal dissection for colorectal epithelial neoplasms

Short‐term outcomes of patients undergoing endoscopic submucosal dissection for colorectal lesions

Objectives

Endoscopic submucosal dissection (ESD) of colorectal lesions was invented in Japan, but postoperative management including hospital stay has not been reconsidered due to the Japanese insurance system. To explore appropriate postoperative management after colorectal ESD, we reviewed short‐term outcomes after ESD in non‐selected consecutive patients.

Methods

Patients who underwent colorectal ESD from April 2013 to September 2020 in one institution were reviewed. The primary outcome measure was the occurrence of adverse events stratified by the Clavien‐Dindo classification with five grades. A logistic regression model with the Firth procedure was applied to investigate predictors of severe (grade III or greater) adverse events.

Results

A total of 330 patients (female 40%, male 60%; median 72 years; IQR 65–80 years) with colorectal lesions (median 30 mm, IQR 23–40 mm; colon 77%, rectum 23%; serrated lesion 4%, adenoma 47%, mucosal cancer 30%, invasive cancer 18%) was evaluated. The en bloc resection rate was 97%. The median dissection time was 58 min (IQR: 38–86). Intraprocedural perforation occurred in 3%, all successfully treated by endoscopic clipping. No delayed perforations occurred. Postprocedural bleeding occurred in 3% on days 1–10 (median day 2); all were controlled endoscopically. Severe adverse events included only delayed bleeding. In analyzing severe adverse events in a multivariate logistic regression model with the Firth procedure, antithrombotic agent use (p = 0.016) and rectal lesions (p = 0.0010) were both significant predictors.

Conclusions

No serious adverse events occurred in this series. Four days of hospitalization may be too long for the majority of patients after ESD.

Endoscopic submucosal dissection techniques and technology: European Society of Gastrointestinal Endoscopy (ESGE) Technical Review

ESGE suggests conventional endoscopic submucosal dissection (ESD; marking and mucosal incision followed by circumferential incision and stepwise submucosal dissection) for most esophageal and gastric lesions. ESGE suggests tunneling ESD for esophageal lesions involving more than two-thirds of the esophageal circumference. ESGE recommends the pocket-creation method for colorectal ESD, at least if traction devices are not used. The use of dedicated ESD knives with size adequate to the location/thickness of the gastrointestinal wall is recommended. It is suggested that isotonic saline or viscous solutions can be used for submucosal injection. ESGE recommends traction methods in esophageal and colorectal ESD and in selected gastric lesions. After gastric ESD, coagulation of visible vessels is recommended, and post-procedural high dose proton pump inhibitor (PPI) (or vonoprazan). ESGE recommends against routine closure of the ESD defect, except in duodenal ESD. ESGE recommends corticosteroids after resection of  > 50 % of the esophageal circumference. The use of carbon dioxide when performing ESD is recommended. ESGE recommends against the performance of second-look endoscopy after ESD. ESGE recommends endoscopy/colonoscopy in the case of significant bleeding (hemodynamic instability, drop in hemoglobin > 2 g/dL, severe ongoing bleeding) to perform endoscopic hemostasis with thermal methods or clipping; hemostatic powders represent rescue therapies. ESGE recommends closure of immediate perforations with clips (through-the-scope or cap-mounted, depending on the size and shape of the perforation), as soon as possible but ideally after securing a good plane for further dissection.