Random Biopsies During Colonoscopy in Patients with Primary Sclerosing Cholangitis and Ulcerative Colitis: Does it Impact Clinical Outcome?

Abstract

Purpose: Patients with primary sclerosing cholangitis (PSC) and ulcerative colitis (UC) are at increased risk of dysplasia and/or colorectal cancer. We hypothesize that random biopsies may have low yield in patients without endoscopic features of previous severe inflammation. We wanted to evaluate the yield and clinical impact of random biopsies taken during colonoscopic surveillance of patients with both UC and PSC. Methods: In total, 54 patients with UC/PSC who underwent colonoscopy at the Cleveland Clinic were included in this study. Retrospective analysis of 142 colonoscopies (both for surveillance and symptomatic indication) in patients with UC and PSC performed from 2001 to 2010. Colonoscopy and pathology reports were reviewed to assess the yield and clinical impact of random biopsies. The main outcome measures are neoplasia yield per-colonoscopy and clinical impact per-patient of random biopsies. Results: Of the 142 colonoscopies, 111 were performed for surveillance(in 54 patients) during which 2,379 random biopsies were taken (median 16) calculated from the pathology report. Overall, neoplasia was detected in 16 colonoscopies (13 patients): in 8 colonoscopies(50%) by targeted biopsies only, in 4 (25%) by both targeted and random biopsies. Neoplasia was detected in random biopsies only in four (25%) colonoscopies in four patients. One patient had unifocal low-grade dysplasia (LGD) that could not be confirmed in two subsequent colonoscopies, while the other two patients with unifocal LGD did not have any evidence of dysplasia in the colectomy specimen. Only one patient had unifocal LGD on random biopsy and proctocolectomy confirmed the presence of multifocal LGD. Neoplasia was detected by only random biopsies in four surveillance colonoscopies (per-colonoscopy yield 25%), while the clinical impact was seen in three colonoscopies (per-colonoscopy yield 18.75%). However the real clinical significance was only seen in one colonoscopy (per-colonoscopy yield 6.25%) This patient had evidence of previous inflammation in the form of pseudopolyps. Increasing age (p=0.03), presence of pseudopolyps (p=0.004), active treatment for underlying UC (p=0.004) and target biopsies done during colonoscopy (p<0.001) independently predicted the presence of dysplasia during colonoscopy on multivariate analysis. (Table 1)Table 1: Multivariate adjusted odds ratio (aOR) for dysplasia during colonoscopyConclusion: Presence of endoscopic features of previous inflammation, increasing age and presence of suspicious lesions predict associated neoplasia in UC/PSC patients. Neoplasia was detected in only random biopsies in four patients of whom only one had significant lesion in the proctocolectomy specimen. Random biopsies may be of higher yield only in patients with previous inflammation on endoscopy even in this high risk population of UC/PSC.