Cap-assisted EMR for large colorectal laterally spreading tumors: a cautionary tale from real-life experience

We have read with great interest the study by Conio et al1Conio M. Manta R. Filiberti R.A. et al.Cap-assisted EMR versus standard inject and cut EMR for treatment of large colonic laterally spreading tumors: a randomized multicenter study (with videos).Gastrointest Endosc. 2022; 96: 829-839.e1Google Scholar about the use of cap-assisted EMR (EMR-C) for large colorectal laterally spreading tumors (LSTs). The authors have investigated for the first time its performance in a randomized controlled trial, showing a similar rate of radical resection to that of standard EMR (EMR-S) but a clear advantage in terms of procedure duration and adverse events. We appreciate the concept and the design of the study, but we would caution against considering EMR-C the treatment of choice for colorectal LSTs. We are proficient in the use of EMR-C to treat large LSTs, and we believe that this technique is underused in routine endoscopic practice, favoring novel modified EMR techniques such as underwater EMR.2Tziatzios G. Gkolfakis P. Papadopouluos V. et al.Modified endoscopic mucosal resection techniques for treating precancerous colorectal lesions.Ann Gastroenterol. 2021; 34: 757-769Google Scholar In our center, from 2012 to 2020 we performed 123 EMR-C procedures versus 63 EMR-S procedures for colorectal LSTs ≥20 mm. We observed a similar rate of R0 resection (P = .73) and of residual/recurrent disease (P = 1.00) but a longer procedure duration with EMR-C (median 128 min vs 104 min with EMR-S; P < .05). Furthermore, EMR-C was associated with a higher rate of intraprocedural bleeding (23.6% vs 7.93% with EMR-S; P < .05) despite the adoption of “controlled suction” as pointed out by Conio et al.1Conio M. Manta R. Filiberti R.A. et al.Cap-assisted EMR versus standard inject and cut EMR for treatment of large colonic laterally spreading tumors: a randomized multicenter study (with videos).Gastrointest Endosc. 2022; 96: 829-839.e1Google Scholar These discrepancies might be explained considering that Conio et al excluded from the analysis many LSTs that, in our experience, would have instead needed EMR-C (ie, lesions with inadequate submucosal lifting, depressed morphology, or previous attempts of resection). For this reason, although we support the use of EMR-C in resecting large colorectal LSTs, we recommend caution in the real-life setting, because this technique may be effective and safe only when performed after strict selection of patients and in a center with long-standing experience. All authors disclosed no financial relationships.