Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

P264 Revised Predictive Values for IBD Colitis-associated Neoplasia in the Modern Era

N. Krugliak Cleveland*1, R. Colman2, D. Rodriquez2, A. Hirsch2, J. Hart2, D. Rubin2

1University of Illinois at Chicago, College of Medicine, Chicago Il, United States, 2University of Chicago, Inflammatory Bowel Disease Center, Chicago, United States

Comparison of Colonoscopy Findings and Colectomy Findings in 35 IBD Patients

Index Grade of NeoplasiaPrimary Lesion at ColonoscopyIndex Lesion Found at ColectomyUpstage or Downstage Index Lesion?Synchronous Lesion at ColectomyGrade of Synchronous Lesions at Colectomy
LGD24/3511/240/117/244 LGD, 3 HGD, 0 CA
HGD4/353/41/3 upstage (CA)1/40 LGD, 0 HGD, 1 CA
Cancer (CA)7/355/71/5 downstage (HGD)2/71 HGD, 1 CA

Background

Due to historical reports of synchronous adenocarcinoma (CA) with low-grade dysplasia (LGD) (19%) and high-grade dysplasia (HGD) (42-60%), colectomy has been advocated when neoplasia of any grade was confirmed in an inflammatory bowel disease (IBD) patient with colitis. More recently, it is understood that most neoplasia in colitis is visible with improved technologies, and that the prognosis of such colitis-associated neoplasia may be different than previously described. This study is a modern assessment of the predictive value of IBD colitis-associated neoplasia found during colonoscopy.

Methods

We performed a retrospective review of all IBD patients at our Center who were found to have pathologist-confirmed neoplasia on surveillance colonoscopy ("index lesion") and underwent a subsequent colectomy between 2005-2014 (the dates of our high definition colonoscopes and monitors). Data included the location/grade of the index lesion at colonoscopy, and the location/grade of any macroscopic or microscopic lesion found on the subsequent standard protocol dissected colectomy specimen. The index lesion found on colonoscopy was compared to the lesion found in the same location at colectomy. Simple statistical analysis was performed.

Results

35 IBD patients met criteria and underwent colectomies for confirmed neoplasia (20 UC (57%, n=13 extensive colitis), 14 CD (40%, n=8 pancolitis) and 1 indeterminate pancolitis). The 36 index lesions included 24 LGD, 4 HGD, 7 CA, and 1 indefinite dysplasia. Of the 24 LGD index lesions (19 white light, 5 chromo), 11 (46%) were confirmed at colectomy (0 were upstaged) and 7/24 (29%) pts had synchronous lesions found at colectomy (4/7 LGD, 3/7 HGD, 0/7 CA). Of the 4/36 HGD index lesions, 3 were found at colectomy (1 of the 3 was upstaged as CA) and the 4th lesion had been resected entirely in polypectomy. 0/4 HGD patients had synchronous lesions. Of the 7/36 index CAs, 5/7 of them were found at colectomy (2 were resected completely prior to colectomy); 1/5 (20%) was downstaged to HGD. 2/7 patients were found to have synchronous lesions: 1 HGD and 1 CA.

Conclusion

With modern visualization techniques and mostly white light exams, we have revised the predictive value of IBD colitis-associated neoplasia. We found no CAs when colectomy was performed for LGD, and only one additional cancer in the index HGD and Ca patients. These findings provide further evidence for active surveillance or subtotal colectomy rather than proctocolectomy in many IBD patients found to have dysplasia.