Beyond conventional endoscopy: Image-enhanced techniques in quiescent ulcerative colitis assessment

Table 1 Key performance metrics for each image-enhanced endoscopy modality compared to Mayo Endoscopic Score/Ulcerative Colitis Endoscopic Index of Severity

IEE modality	Interobserver agreement (κ or description)	Predictive HR (relapse)	Correlation with MES/UCEIS	Ref.
NBI	Good to excellent interrater reliability	Detects incomplete vascular regeneration (not specified)	Strong concordance with MES and histology	[20-23]
TXI	Moderate to good (κ = 0.597-0.823)	4.16 for TXI score 2 (high risk of relapse)	Strong correlation with MES and UCEIS	[24,25]
RDI	Superior to MES with strong expert/non-expert agreement	3.56 for RDI scores 3-4 (increased relapse risk)	Significant correlation with MES and UCEIS	[25-28]
LCI	Excellent interobserver agreement	Strong predictor, numeric HR not given	Strong correlation with MES and histology	[30-34]
iSCAN/PICaSSO	Excellent reproducibility	Comparable to MES/UCEIS in relapse prediction	Strong correlation with MES and UCEIS	[36-41]

IEE: Image-enhanced endoscopy; NBI: Narrow-band imaging; TXI: Texture and color enhancement imaging; RDI: Red dichromatic imaging; LCI: Linked-color imaging; PICaSSO: Paddington International virtual ChromoendoScopy ScOre; MES: Mayo Endoscopic Subscore; HR: Hazard ratio; UCEIS: Ulcerative Colitis Endoscopic Index of Severity.

Table 2 Representative research on evaluating ulcerative colitis histological activity using image-enhanced endoscopy

Ref.	Study design	Endoscopic technology	Histologic score	Outcomes
Kudo et al[20]	Single-center prospective	Conventional NBI	NA	Acute inflammatory cell infiltrates (26% vs 0%) and goblet cell depletion (32% vs 5%) were more commonly found in segments with an obscure MVP compared to those with a clear MVP, while basal plasmacytosis was less frequent in the former (2% vs 21%)
Sasanuma et al[22]	Single-center prospective	WLE, magnified NBI	Definitions for histological findings followed those of the Japanese Ministry of Health, Labour, and Welfare	Over 60% of mucosal areas with a normal appearance under white light exhibited BV-H on NBI, while two-thirds of mucosal areas showing scars on white light displayed BV-BB on NBI; a significant correlation was observed between magnified NBI findings and histological results
Iacucci et al[37]	Single-center retrospective cohort	WLE, iSCAN	ECAP	Mucosal and vascular pattern scores showed a significant correlation with most of the parameters in the proposed ECAP score
Iacucci et al[38]	Single-center prospective cohort	WLE, iSCAN	RHI and ECAP	Overall iSCAN score showed a strong correlation with the ECAP score; its accuracy in detecting abnormalities based on ECAP was 80%, with a sensitivity of 78% and specificity of 100%; Both the iSCAN vascular and mucosal scores were also significantly correlated with ECAP; correlation between the overall iSCAN score and RHI was moderate, with an accuracy of 68% for detecting abnormalities by RHI (sensitivity 78%, specificity 50%); notably, most patients with MES 0 still exhibited abnormalities on iSCAN
Iacucci et al[41]	Multicenter prospective	HD-WLE, iSCAN	RHI, NHI, ECAP, GS, Villanaci	PICaSSO score demonstrated a strong correlation with multiple histological indices, showing significantly higher correlation coefficients compared to those of MES and UCEIS with histology scores
Takagi et al[33]	Single-center prospective observational	WLE, LCI	GS	There was no significant difference in the proportion of the LCI group nor the relapse rate between UC patients with histologically active disease and those with inactive disease, as determined by GS
Matsumoto et al[34]	Single-center prospective	WLE, LCI	GS	LCI findings showed a significant correlation with GS
Kanmura et al[32]	Single-center prospective	WLE, LCI	GS	LCI was significantly correlated with the histological mucosal inflammation score
Naganuma et al[27]	Single-center retrospective cohort	WLE, RDI	GS	Correlation between the RDI score and the histological grade was similarly significant for both expert and non-expert investigators
Hashimoto et al[28]	Single-center prospective	WLE, RDI	NHI	RDI score showed a positive correlation with the NHI and exhibited a stronger correlation with the NHI than both UCEIS and MES
Hayashi et al[24]	Single-center prospective	WLE, TXI	Matts	When pathologic remission was defined as a Matts grade of ≤ 2, rate of remission significantly decreased as TXI scores increased
Mamiya et al[25]	Single-center prospective	WLE, TXI, RDI	GS	TXI and RDI scores showed significant correlations with each GS, with a particularly strong association observed for Geboes grade 2B across all scores

NBI: Narrow-band imaging; WLE: White light endoscopy; HD-WLE: High definition white light endoscopy; LCI: Linked-color imaging; RDI: Red dichromatic imaging; TXI: Texture and color enhancement imaging; NA: Not available; ECAP: Extent, Chronicity, Activity, Plus score; RHI: Robarts Histopathology Index; NHI: Nancy Histological Index; GS: Geboes score; MVP: Mucosal vascular pattern; BV-H: Honeycomb-like blood vessels; BV-BB: Blood vessels shaped like bare branches; PICaSSO: Paddington International virtual ChromoendoScopy ScOre; UC: Ulcerative colitis; UCEIS: Ulcerative Colitis Endoscopic Index of Severity; MES: Mayo Endoscopic Subscore.