Sa1893 EFFICACY AND SAFETY OF A MAGNET TRACTION DEVICE FOR ENDOSCOPIC SUBMUCOSAL DISSECTION: EX-VIVO STUDY IN A PORCINE MODEL

Abstract

Many traction devices have been developed for safer and easier endoscopic submucosal dissection (ESD). However, traction caused by most devices depends on the endoscopic longitudinal direction with ineffective traction attained in most cases. Magnetic anchor-guided ESD has been reported and requires both an internal and external magnet. This system allows traction to occur independent from the endoscopy. However, the strength of the magnets varied by location of the lesion and thickness of the abdominal wall. Thus, we developed a magnet traction device that consisted of two internal magnets. The aim of this study was to evaluate the efficacy and safety of an internal magnet traction device (MTD) for ESD in ex vivo porcine stomachs. The MTD consisted of a small neodymium magnet and a 2-0 suture (2cm length) attached to a hemoclip. A circumferential mucosal incision was completed around a 30mm round template that served as the target lesion. The first MTD was deployed at the edge of the lesion and the second MTD was deployed on the opposing wall to the lesion which resulted in both magnets connecting. This created traction of the target lesion towards the opposing wall facilitating resection (Figure). Submucosal dissection was performed using a needle knife. After resection of the lesion, all MTDs and hemoclips attached to the lesion were retrieved en bloc using a snare. The primary outcome measure was the comparison of submucosal dissection times between conventional ESD (C-ESD) and MTD assisted ESD (MTD-ESD). Secondary outcomes measured included rate of en bloc resection, visualization of submucosal layer during dissection, number of muscular injuries, and ease of use of the MTD. Visualization and ease of use were evaluated using a visual analog scale ranging from 1 (poor) to 5 (excellent). A total of 20 lesions were created in five porcine stomachs along the anterior wall, posterior wall or greater curvature. Ten lesions each were resected using C-ESD or MTD-ESD. En bloc resection was successfully completed without perforation in all cases. All MDTs were retrieved successfully. The submucosal dissection time in MTD-ESD was significantly shorter than C-ESD (median: 6.4minutes [IQR: 4.6 – 8.7] vs.14.4 minutes [IQR: 11.8 – 18.0], p<0.05). The visualization score of the submucosal layer during dissection in MTD-ESD was significantly higher than in C-ESD (median: 5 [IQR: 4 - 5] vs.3 [IQR: 2 - 4], p<0.05) and the number of muscular injuries was significantly lower in MTD-ESD than C-ESD (median: 0 [IQR: 0 - 0] vs.1 [IQR: 0 - 2], p<0.05). The ease of use score of MTD was excellent (median: 5 [IQR: 5-5]) (Table). The MTD-ESD was both effective and safe compared to C-ESD with significantly shorter procedure time, improved visualization of the submucosal layer and absent deep muscle-based trauma.TableMTD-ESD (n=10)C-ESD (n=10)P valueRate of en bloc resection100%100%-Submucosal dissection time (minute); medium (IQR)6.4 (4.6 – 8.7)14.4 (11.8 – 18.0)<0.05Specimen size (mm); medium (IQR)38 (36 - 40)39 (36 - 40)0.908Visualization score of submucosal layer; medium (IQR)5 (4 - 5)3 (2 - 4)<0.05Number of muscular injuries; medium (IQR)0 (0 - 0)1 (0-2)<0.05Ease of use score of MTD; medium (IQR)5 (5-5)-- Open table in a new tab