Objective Endoscopic submucosal dissection (ESD) aims to achieve en bloc resection of non-pedunculated colorectal adenomas which might be indicated in cases with superficial submucosal invasive cancers (SMIC), but the procedure is time consuming and complex. The prevalence of such cancers is not known but may determine the clinical necessity for ESD as opposed to the commonly used piecemeal mucosal resection (endoscopic mucosal resection) of colorectal adenomas. The main aim was to assess the prevalence of SMIC SM1 (ie, invasion ≤1000 µm or less than one-third of the submucosa) on colorectal lesions removed by ESD.
Design A literature review was conducted using electronic databases (up to March 2017) for colorectal ESD series reporting the histology of the dissected lesions.
Results 51 studies with data on 11 260 colorectal dissected lesions were included. Most resected lesions (82.2%; 95% CI 78.8% to 85.3%) were adenomas (low- and high-grade dysplasia, 26.8% and 55.4%, respectively). Overall, 15.7% were submucosal cancers, but only slightly more than half (8.0%; 95% CI 6.1% to 10.3%) had an infiltration depth of ≤1000 µm, providing a number needed to treat (NNT) to avoid one surgery of 12.5. Estimating an oncologically curative (R0; G1/2; L0/V0) resection rate of 75.3% (95% CI 52.2% to 89.4%) for malignant lesions, the prevalence of curative resection lowered to 6% (95% CI 4.2% to 7.2%) with an NNT of 16.7.
Conclusion The low prevalence of SMIC SM1 in lesions selected for ESD as well as the even lower rate of curative resection limits the clinical applicability of endoscopic en bloc resection. This calls for caution over an indiscriminate use of this technique in the resection of colorectal neoplasia.
- colorectal cancer
- endoscopic submucosal dissection
- interventional endoscopy
Statistics from Altmetric.com
LF, AR, CH and DKR contributed equally.
Contributors Concept and design: LoF, AR, CH and DKR; analysis and interpretation of the data: LoF, AR, CH and DR; drafting of the article: LoF, AR, TR and CH; critical revision of the article for important intellectual content: TP, PB, PS, RJ, TT, FB, TR and DR; final approval of the article: all the authors; statistical expertise: LoF, LeF and CH; collection and assembly of data: LoF, DM, LeF and CB.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.