Mo1261 THE EFFECT OF FIBROSIS ON EFFICACY OF ENDOSCOPIC SUBMUCOSAL RESECTION IN BARRETT’S ESOPHAGUS

Abstract

Endoscopic submucosal dissection (ESD) is an established endoscopic resection technique used to treat superficial esophageal cancer. ESD is predominantly used in Asia where en bloc resection is often successful for esophageal squamous cell cancer or gastric adenocarcinoma due to good tissue plane delineation. It is in these lesions that early training in ESD is recommended. In comparison, most Western ESD procedures occur in the setting of Barrett’s esophagus (BE), a known inflammatory condition often treated with ablation therapy or endoscopic mucosal resection prior to ESD, making fibrosis more common. To determine the effect of submucosal fibrosis on ESD in BE. A prospectively maintained database of esophageal ESD performed in the BE unit for neoplasia was reviewed for cases of fibrosis. All procedures were performed by a single endoscopist using an Olympus H190 (Center Valley, PA) gastroscope and a Fujifilm Clutch Cutter DP2618DT device (Tokyo, Japan) at Mayo Clinic (Rochester, MN) from 2017-2018. ESD was performed after injection of epinephrine, hydroxymethylcellulose, and methylene blue to elevate the mucosa. Primary outcomes were intraprocedural or post-procedural complications up to 7 days following ESD, and R0 curative and en bloc resection. A total of 142 ESDs were performed among 97 patients (74.2% male, mean age 69.1±10.4 y). Refer to Table 1 for lesion characteristics. En bloc resection was achieved in 124 (87.3%) cases. Mean resection size was 24.0±10.8 mm, with the defect measuring a mean 27.8±14.9% (range 10-80%) of the luminal circumference. For adenocarcinoma or squamous cell carcinoma-related resections, R0 was achieved in 16 of 50 resections (32%). Fibrosis was encountered in 24 (16.9%), and of these, 2 were noted as F1, 13 as F2 per standard classification. Notably, patients with fibrosis were more likely to have had prior endoscopic resection (p 0.01) and were more likely to go on to failed en bloc ESD (p <0.0001). Mean procedure time did not differ (p 0.44). Complications were identified in 2 cases without fibrosis; 1 involved intraprocedural bleeding from a 3 mm arterial branch that was cauterized with the Clutch Cutter device and also resulted in use of topical hemostatic spray, while 1 involved need for blood transfusion within 24 hours of the procedure, followed by topical hemostatic spray and epinephrine to address bleeding along the resection margin at 5 days post-procedure. Patients were on antiplatelet/anticoagulants for 52.8% of procedures, including 30.3% on aspirin, 2.1% clopidogrel, 9.9% warfarin, 9.2% direct oral anticoagulants. ESD in BE with fibrosis is more likely to result in failed en bloc resection though complications were not increased. Fibrosis is common in patients with BE and may make lesions less ideal for training for Western endoscopists.