Objectives: Multiple biopsy has been a recommended procedure for cancer surveillance in patients with ulcerative colitis (UC). The aim of this study was to investigate the accuracy of chromoscopic findings in surveillance for patients with UC.
Methods: During the period 1995-2002, we performed 117 surveillance colonoscopies in 57 patients with pancolitis for more than 5 yr. Multiple biopsy specimens were uniformly obtained from flat mucosa in each segment of the colorectum, and, when necessary, from areas specified by chromoscopy. The specified area was classified as polypoid lesion or visible flat lesion. In each specimen, histology was graded according to the classification of dysplasia.
Results: Among 818 specimens, 28 (3.4%) were positive for dysplasia. There were 20 low grade dysplasias and eight high grade dysplasias. Dysplasia was more frequently positive in visible flat lesions (37.1%, p < 0001) and in polypoid lesions (16.9%, p < 0.0001) than in flat mucosa (0.4%, p < 0.0001). Furthermore, it was more frequently positive in visible flat lesions than in polypoid lesions (p < 0.05). High-grade dysplasia was found in 4.4% of polypoid lesions and in 14.8% of visible flat lesions, but it was not detected in flat mucosa. Overall, dysplasia was detected in 12 patients. Positive dysplasia was confined to visible flat lesions in four patients and to flat mucosa in one patient.
Conclusions: Our results suggest that biopsy from flat visible lesions under chromoscopy might improve the accuracy of cancer surveillance in UC.