The iam of the study is  compare the performance of virtual chromoendoscopy (VCE) using Linked Color Imaging(LCI) and dye-chromoendoscopy (DCE) with methylene blue for the detection of colonic neoplasticlesions during surveillance colonoscopy in patients with inflammatory bowel disease (IBD), since curently data on LCI performances  ford etecting dysplasia specifically among IBD patients undergoing surveillance remain scarce compare to other virtual chromoendoscopy technology including Narrow Band Imaging (NBI, Olympus, Japan) or  i-SCAN (Pentax, Japan). 

This is a single-centre retrospective cohort study included patients with colonic IBD undergoingcolorectal cancer (CRC) surveillance colonoscopy between June 2018 and December 2024. The primary outcome was the detection of total neoplastic lesions, defined as the mean number ofneoplastic lesions detected per colonoscopy, including colitis-associated dysplasia,adenocarcinoma, sporadic adenoma, serrated lesions, by targeted biopsies or polypectomy witheither DCE or LCI in subjects with long-standing colonic IBD. The secondary outcome was theneoplastic lesions detection rate, defined as the proportion of colonoscopies with at least oneneoplastic lesion over the total number of colonoscopies

Propensity-score matchingand multivariable Poisson and logistic regression were used to adjust for baseline differences andidentify independent predictors.

A total of 404 patients underwent surveillance colonoscopy during the study period, 135 with DCEand 269 with LCI. After exclusion of 93 patients because of inadequate bowel preparation orendoscopically active disease (23 in the DCE group, 70 in the LCI group, Figure 3), 311 wereincluded in the final analysis (112 DCE, 199 LCI). Due to the retrospective design, baseline characteristicswere not fully balanced between the two groups. Compared with the LCI group, patients in the DCE group had a longer disease duration (19.58 vs 17.12 years, p=0.021), a higher prevalence ofextensive (E3) UC, a higher frequency of concomitant primary sclerosing cholangitis, and a greaterproportion with a personal history of colonic lesions. A first-degree family history of colorectalcancer (CRC) was also more common in the DCE arm. After the 1:1 propensity-score matching, 208 procedures (104 DCE and 104 LCI) were available foranalysis.

Neoplasia detection did not differ significantly between DCE and LCI. The mean number ofneoplastic lesions identified per colonoscopy was 0.116 with DCE and 0.085 with LCI (p=0.472).Similarly, the neoplasia detection rate was comparable between techniques (8.03% for DCE vs7.53% for LCI; OR 1.07, 95% CI 0.45-2.54, p=0.874). After 1:1 propensity score matching, thechromoendoscopy modality (LCI vs DCE) was not associated with the total number of neoplasticlesions (IRR 0.77; 95% CI 0.34–1.78; p=0.543) and with the neoplastic lesions detection rate (OR0.58; 95% CI 0.21–1.65; p=0.310).

LCI-based VCE provides a neoplastic yield comparable to traditional DCE in patients with long-standing colonic IBD undergoing surveillance colonoscopy. As with other VCE modalities, LCImay be adopted as an acceptable alternative to DCE for dysplasia surveillance in IBD.