LAUSR

BACKGROUND AND AIMS Cold snare defect protrusions (CSDPs) include muscularis mucosa (MM) and submucosa (SM) tissue. CSDPs are thought to result from fragmentation of the specimen during shallow excision. Our aim in this study was to clarify whether CSDPs are in fact associated with polyp fragmentation. METHODS We retrospectively analyzed 1026 neoplastic colorectal polyps resected by cold snare polypectomy from which the presence or absence of CSDPs was assessed from the endoscopic image. All prepared specimens were reviewed and assessed for the presence or absence of polyp fragmentation and the proportion of MM on the stump was measured. Additionally, the risk factors for CSDP occurrence were evaluated. RESULTS CSDPs occurred in 116 of the 1026 polyps (11.3%). Polyp fragmentation was significantly associated with CSDP occurrence on univariate (odds ratio (OR), 3.74; P<0.001) and multivariable (OR, 3.13; P<0.001). The proportion of MM >50% was significantly lower in the CSDP than in the non-CSDP (51.5% vs 70.9%, P<0.001) group. CSDPs were significantly associated with a large polyp size (OR, 1.32; P=0.007) and a large specimen size (OR, 1.24; P<0.001) on multivariable analysis. CONCLUSIONS CSDP occurrence was associated with less MM on the stump and fragmentation of the specimen. Clinically, presence of a CSDP is a good indicator of polyp fragmentation.