During colonoscopic surveillance for dysplasia in longstanding extensive ulcerative colitis (UC), multiple non-targeted "random" biopsies of colonic mucosa are advised, based on historical data suggesting dysplasia may only be detectable microscopically.
In this study, researchers from Harrow, England, assessed what proportion of dysplastic lesions were macroscopically evident at colonoscopy.
|Overall, 66% of the dysplastic biopsies were from macroscopically visible lesions.|
The team reviewed all cases of colonoscopically detected dysplasia in a major UC surveillance program, between 1988 to 2002.
The team obtained patient details from a colonoscopy database, case notes, and histology reports.
They assumed all dysplasia to be macroscopically invisible unless stated otherwise at the time.
During the study period, 300 patients underwent 2189 colonoscopies.
Of these, 56 patients had 1 or more biopsies showing colorectal dysplasia. In total, 92 colonoscopies yielded 106 positive biopsies.
The team found that 76% of dysplastic biopsies were from macroscopically visible lesions, and that 24% of dysplasia sites were macroscopically invisible.
They determined that 33 lesions were endoscopically and histologically tubular adenomas. When these lesions were excluded the team found that there were 73 dysplastic biopsies, from 38 patients undergoing 65 colonoscopies.
Overall, 66% of the dysplastic biopsies were from macroscopically visible lesions.
The research team found that 82% of patients had macroscopically detectable dysplasia during the study period. However, 18% of patients with dysplasia macroscopically invisible lesions.
Dr Rutter's team concluded, "Over 80% of patients with dysplastic lesions in ulcerative colitis will develop a colonoscopically visible lesion".
"Even after excluding tubular adenomatous lesions, the majority of dysplasia is colonoscopically visible."
"Colonoscopists should concentrate on careful mucosal scrutiny for dysplastic lesions, rather than relying solely on detection by random biopsies."