Aims Long-standing ulcerative colitis (UC) has an increased risk for developing colorectal
dysplasia and neoplasia. Dye-based chromoendoscopy (DCE) and virtual chromoendoscopy
(VCE) increase detection of neoplastic lesions. However, limited data are available
on the impact of i-scan VCE for UC neoplasia detection. We undertook a prospective
randomized controlled trial to compare the neoplasia detection between DCE and i-scan
in patients with long-standing UC.
Methods In 4 European hospitals, 136 patients with long-standing UC (mean disease duration
19.88 (DCE) vs 18.49 years (i-scan)) were randomized (1:1) to either DCE with methylene
blue 0.1% (n=71) or i-scan (n=65). Biopsies were taken from visible lesions and surrounding
mucosa. Neoplasic lesions included any type of dysplasia, polyp or carcinoma. Statistical
analysis was performed using t-test for continuous data and Fishers’ exact for proportions.
Results The neoplasia detection rate was not significantly different between the DCE (18.3%)
vs VCE (26.2%) group, respectively (OR 0.63, 95%CI 0.27 – 1.37, p=0.305). However,
the per lesion neoplasia detection was significantly better with i-scan than with
DCE (14.5% vs 33.9%, p=0.033). The mean number of neoplastic lesions per colonoscopy
was 0.24 for DCE and 0.32 for i-scan (p=0.432). Both withdrawal and total procedural
time were on average 10.1 and 9.8 minutes shorter in the i-scan group (p<0.001).
Conclusions This multicenter prospective randomized trial showed that in long-standing UC patients,
no significant difference in neoplasia detection was found between DCE and i-scan.
However, i-scan had a lower false positive rate and a significant shorter procedure
time. I-scan could therefore be a valid replacement for DCE.
