Endoscopic submucosal dissection (ESD) allows en bloc resection of early colorectal neoplasia, providing precise histopathological evaluation and lower recurrence. However, colorectal ESD remains technically demanding due to prolonged procedure times and higher complication rates, especially in cases with severe fibrosis. Severe submucosal fibrosis may arise from prior interventions or inherent lesion characteristics, and it often obscures the dissection plane. Gel-immersion ESD (GI-ESD), using Viscoclear (Otsuka Pharmaceuticals, Tokushima, Japan), a highly viscous and transparent gel approved in Japan, has emerged as a potential solution. This gel maintains luminal clarity through its retention effect, stabilizes scope operability, and provides a durable submucosal lift via buoyancy even in fibrotic lesions. These effects are superior to saline. This study aimed to evaluate the clinical outcomes of GI-ESD for colorectal lesions with severe submucosal fibrosis compared with conventional ESD (C-ESD).

This single-center retrospective cohort study included all consecutive colorectal ESDs at a high-volume Japanese center from January 2022 to June 2025. Only cases with intraoperatively confirmed severe fibrosis were analyzed. Procedures were performed by expert endoscopists or supervised trainees. All lesions underwent preprocedural evaluation using high-definition magnification. ESD was performed using Fujifilm or Olympus colonoscopes with waterjet function and this water jet function was used for gel injection. Submucosal lifting used sodium hyaluronate or alginate mixed with indigo carmine. Dissection utilized scissors-type, needle-type, or mixed-knife strategies, with traction applied at the operator’s discretion. C-ESD was performed before April 2024; thereafter, GI-ESD was introduced, involving suction of intraluminal gas and gel instillation to maintain visualization. Primary outcome was procedural safety (delayed bleeding, perforation, post-ESD coagulation syndrome [PECS]). Secondary outcomes included en bloc and R0 resection rates, procedure duration, and operator conversion.

A total of 85 colorectal ESDs with severe fibrosis were analyzed (GI-ESD: n=22; C-ESD: n=63). Baseline demographics were comparable (mean age: 68.7 vs 66.5 years, female rate: 27.3% vs 33.3%), right-colon location (54.5% vs 49.2%; p=0.199), median lesion size [IQR] (25 ± 10 mm vs 30 ± 25 mm), polypoid morphology (23.8% vs. 47.6%; p=0.480), and expert operator (81.8% and 73.0%; p=0.953). Regarding primary endpoints, no adverse events occurred in GI-ESD, whereas 14 events (22.2%) occurred with C-ESD: perforations (n=5; 7.9%), delayed bleeding (n=2; 3.2%), and PECS (n=7; 11.1%). En bloc resection was achieved in all completed cases; R0 rates were higher with GI-ESD (85.7% vs 72.8%; p=0.191). Median procedure time was shorter in GI-ESD (56 vs 87 min; p=0.015), and mean time favored GI-ESD (74.2 vs 101.2 min; p=0.075). Procedure discontinuation was similar (4.5% vs 6.3%). In multivariable regression, GI-ESD was not independently associated with procedure duration of severe fibrotic cases (p=0.500); lesion size remained the only independent predictor (p<0.001).

In this high-risk cohort, GI-ESD was associated with favorable safety and procedural efficiency compared with conventional ESD. No adverse events occurred in GI-ESD. The findings suggest that a viscous gel environment can mitigate challenges associated with severe fibrosis without compromising resection quality, potentially offering clinical and economic benefits by reducing complications and associated resource use.