In a study conducted at 29 international centers, mostly in Europe, experts performed 10,812 colorectal endoscopic submucosal dissections (ESDs) over a 6-year interval, of which 315 (2.9%) involved ≥90% of the circumference and were considered circumferential.
The median lesion size was 95 mm, and 85.4% of circumferential lesions were rectal lesions. Full (100%) involvement of the circumference was present in 36.8% of lesions, and the remainder had 90% to 99% involvement.
Of 314 completed endoscopic procedures, 85.1% were en bloc resections (81.8% R0), with a median procedure time of 187 minutes. Steroid prophylaxis was used in only 18% of cases. Superficial submucosal invasion was present in 9.9% of lesions. Intraprocedural perforation occurred in 13.7% of patients, delayed bleeding in 10.2%, and delayed perforation in 0.64%. Recurrences occurred in 9.5%, which were managed with additional endoscopic resection.
There were 284 patients evaluable for stenosis at follow-up. Stenosis occurred in 34.2% of these patients, with a higher incidence in those with rectal lesions (38.2%) than with colonic lesions (11.6%). Prophylactic measures to prevent stenosis appeared to be ineffective. Most patients (86.6%) with stenosis had symptoms, with a median time to onset of 34 days (interquartile range, 23-90). Stenoses were effectively treated in all cases with balloon or bougie dilation, requiring a median of 2 or 3 sessions, respectively.
In multivariable analyses, the predictors of stenosis were 100% circumferential involvement or resection (odds ratio [OR], 2.44), rectal location (OR, 4.40), and long-axis lesion size (OR, 1.01/mm).
Douglas K. Rex, MD, MASGE
Bio and Disclosures
Sferrazza S, Calabrese G, Maida M, et al; CIRCLE-ESD (Circumferential Colorectal ESD) Study Group. Stenosis development after circumferential colorectal endoscopic submucosal dissection: a multicenter analysis of predictive factors and outcomes.
Gastrointest Endosc 2025 Dec 26. (Epub ahead of print) (
https://doi.org/10.1016/j.gie.2025.12.276)