Possible Role of Image-Enhanced Endoscopy in the Evaluation of Mucosal Healing of Ulcerative Colitis

Background: Mucosal healing (MH) is recognized as a therapeutic target in ulcerative colitis (UC) because of evidence that it is associated with favorable clinical outcomes. Current endoscopic assessment of MH by conventional white-light endoscopy is subject to several important clinical issues including the subjective nature of assessment, intra- and interobserver variability, and persistent microscopic inflammation, even in mucosa it was observed as quiescent on conventional endoscopy. Summary: Advances in image-enhancement technologies enable the provision of high-contrast images that emphasize the mucosal structures, blood vessel patterns, and color tones of the intestinal mucosa, and recently, several image-enhanced endoscopy (IEE) techniques have become available for the assessment of MH in UC. Narrow-band imaging and dual-red imaging facilitate visualization of mucosal vascular structures, which is useful for detecting minor inflammation and predicting relapse because of the capturing of information on incomplete vascular regeneration in patients with UC. Linked-color imaging (LCI) is optimized to emphasize the redness of the mucosa and blood vessels, and is superior for depicting subtle color changes arising from mucosal inflammation. LCI could possibly be used to stratify UC patients with MH on conventional endoscopy. Autofluorescence imaging and i-scan can also depict subtle histological changes underlying the healing of mucosa in UC, revealing them as simple color changes. Key Messages: Accumulating evidence suggests that IEE techniques could overcome current unmet needs in the endoscopic assessment of MH in UC and contribute to improving therapy based on treat-to-target strategies.

© 2022 S. Karger AG, Basel

Introduction

Ulcerative colitis (UC) is a chronic idiopathic inflammatory disease that affects the colon and is characterized by relapsing and remitting mucosal inflammation. With recent advances in the treatment of inflammatory bowel disease (IBD), the importance of choosing treatment strategies based on a specific therapeutic target, “treat-to-target (T2T),” has been advocated to improve the long-term outcomes of patients with IBD. Mucosal healing (MH) is recognized as a therapeutic target in UC because increasing evidence suggests that MH is associated with better clinical outcomes such as sustained clinical remission, lower rates of hospitalization, decreased need for surgery, and lower risk of colorectal cancer [1-5]. However, there are currently no standardized criteria for evaluating disease activity and no validated definition of MH. The latest update of the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) program, STRIDE-II, proposes that a Mayo endoscopic subscore (MES) of 0 or an Ulcerative Colitis Endoscopic Index of Severity (UCEIS) ≤1 is considered as endoscopic healing and are treatment targets for UC [6]. However, these scoring systems are subjective and result in intra- and interobserver variations in the endoscopic assessment of UC. Furthermore, even patients with quiescent UC with an MES of 0 have a relapse rate of approximately 10% during the first 6 months of follow-up [7]. Persistent microscopic inflammation is associated with a higher relapse rate in endoscopically quiescent disease [8]. Recent studies have also shown that histologic risk factors are associated with clinical relapse and long-term prognosis in patients with quiescent UC [9-11]. These findings imply that conventional endoscopy is insufficient for evaluating MH. Additionally, because histological specimens only demonstrate disease activity within a limited portion of the colon, it remains unclear whether histological evaluation can represent the overall inflammatory condition. Currently, image-enhanced endoscopy (IEE) is being considered as a modality to overcome these unmet needs in the assessment of MH. IEE provides high-contrast images of lesions using optical or electronic technologies, including contrast enhancement of the mucosal surface, blood vessels, and color tone [12]. IEE has potential for more objective assessment of MH and the detection of minute differences reflecting mucosal inflammation that cannot be recognized on conventional endoscopy. In this article, we review the current literature on IEE for the assessment of MH in patients with UC.

Narrow-Band Imaging

Narrow-band imaging (NBI) is an optical endoscopic technology that allows clear visualization of microvascular structures in the surface layer of the gastrointestinal mucosa. NBI uses narrow-band illumination created with two NBI filters, blue light at 415 ± 30 nm and green light at 540 ± 30 nm, with 415 nm being the hemoglobin absorption band [13]. NBI combined with magnifying endoscopy is commonly used to identify early-stage carcinoma in the oropharynx, hypopharynx, esophagus, stomach, and large intestine. NBI systems are also expected to be useful for evaluating MH in patients with UC because the regeneration and reconstruction of blood vessels occur in the healing process after mucosal injury (Fig. 1).

Fig. 1.

The magnified NBI findings of colonic mucosa with an MES of 0 in patients with quiescent UC. Based on the arrangement of the mucosal vascular pattern, the NBI findings in the healed colonic mucosa can be classified into several types, including a honeycomb-like blood vessels (left) and bare branches-like blood vessels (right), under magnifying observation. NBI, narrow-band imaging; MES, Mayo endoscopic subscore; UC, ulcerative colitis.

/WebMaterial/ShowPic/1471537

Esaki et al. classified the magnifying NBI findings of patients with mildly active UC into two types: the crypt opening type with whitish round crypts and the villous type with a villous surface structure [14]. Histologic analysis revealed that atrophic or distorted crypts and goblet cell depletion were more frequently found in mucosa with the villous type than in mucosa with the crypt opening type. In inactive UC, they divided the mucosal vascular pattern (MVP) of rectal mucosa into two types: a honeycomb-like MVP and irregular MVP based on the arrangement of the MVP under NBI observation. They reported that these NBI findings closely correlated with histologic findings, including cryptal distortion, goblet cell depletion, and basal plasmacytosis, suggesting that NBI observation is helpful for in vivo assessment of histologic severity in patients with mildly active and inactive UC. Isomoto et al. [15] also performed a similar study to explore the association of magnifying endoscopic findings on NBI with clinical relapse in patients with inactive UC. They classified the rectal mucosal MVP on NBI-guided magnifying observation into either MVP regularly arranged in a honeycomb-like structure (MVP-regular) or MVP with an irregular tortuous structure (MVP-irregular), similar to the previous classification proposed by Esaki et al. [14]. The cumulative nonrelapse rate of the MVP-regular group was significantly higher than that of the MVP-irregular group.

Sasanuma et al. [16] evaluated the magnified NBI findings of colonic mucosa with an MES of 0 or 1 in patients with quiescent UC and their relationships with histologic activity and prognosis. In their study, they categorized magnified NBI findings into three groups: honeycomb-like blood vessels (BV-H), blood vessels shaped like bare branches (BV-BB), and blood vessels shaped like vines, according to the vascular structures on the colonic mucosal surface. Active histologic inflammation was defined as positive according to at least one of the following criteria: diffuse infiltration of inflammatory cells, erosion, and a crypt abscess/cryptitis. Their histologic assessment revealed that BV-H and BV-BB groups rarely showed active histologic inflammation (BV-H: 4%, 12/292 and BV-BB: 3%, 8/299, respectively), whereas the blood vessels shaped like vines group showed high histologic activity (81%, 27/33), indicating a significant correlation between magnified NBI findings and histologic activity (p < 0.001). In addition, the authors reported that 1.4% (2/144) of the BV-H group and 16.7% (27/162) of the BV-BB group were histologically active at 1-year follow-up. The odds ratio for the BV-BB group showing histologically active inflammation at 1-year follow-up was 14.2 (95% CI: 3.3–60.9). These data suggest that patients with different NBI vascular patterns have a different potential risk of relapse, even within patients with UC showing endoscopic remission.

Considering the important role of Peyer’s patches (PPs) in the initiation of the intestinal mucosal immune response, Hiyama et al. [17] evaluated the endoscopic features of PPs on NBI with magnifying endoscopy (NBI-ME) and investigated the association of NBI-ME images with the clinical course in patients with UC. They collected NBI-ME images of PPs from 67 consecutive patients with UC who underwent ileocolonoscopy and divided the PPs of the patients into two groups: a low (L)-type (score 0 or 1) and a high (H)-type (score 2 or 3), according to the Villi index [18]. The Kaplan-Meier analysis showed that the clinical relapse rate was significantly higher in patients with H-type PPs than in those with L-type PPs (relapse rates over 12 months: 54 vs. 17%; 24 months: 73 vs. 25%; p < 0.01). In contrast, there was no significant difference in cumulative relapse rates between patients with different MESs (p = 0.76). Moreover, a multivariate analysis revealed that the H-type of the PPs was an independent risk factor for clinical relapse (HR, 3.3; 95% CI: 1.4–8.2; p = 0.0047).

As NBI is superior to conventional white-light endoscopy for depicting microvascular structures, it may be useful for detecting minor inflammation and predicting relapse by capturing incomplete vascular regeneration in patients with UC. However, because UC is characterized by contiguous and diffuse lesions in the colon, there remains a challenge of how to apply the evaluation of localized lesions on magnified NBI observation to the evaluation of the entire lesion.

Linked-Color Imaging

Linked-color imaging (LCI) is a novel image-enhancing technology available for the LASEREO laser endoscope system developed by Fujifilm Corporation (Tokyo, Japan). LCI provides an expanded color range obtained by pre-processing narrow-band irradiation and post-processing color separation, such that blue, green, and red are reallocated to amplify color differentiation, thereby making it easier to recognize slight differences in the color of the mucosa. In particular, LCI can display images with enhanced contrast of the redness of the mucosa and blood vessels by superimposing a violet light of 410 nm, resulting in reddish and whitish colors becoming redder and whiter, respectively. Therefore, LCI facilitates improved visualization of inflammatory changes in comparison with conventional white-light imaging (WLI), such as those arising from inflammation or atrophy (Fig. 2).

Fig. 2.

The representative WLI (left) and LCI images (right) of colonic mucosa in UC patients. The LCI visualizes the mucosal redness more clearly than conventional WLI even in a distant view due to sufficient brightness. WLI, white-light imaging; LCI, linked-color imaging; UC, ulcerative colitis; MES, Mayo endoscopic subscore.

/WebMaterial/ShowPic/1471535

Using these characteristics of LCI, Uchiyama et al. [19] proposed a new LCI classification in which mucosal redness recognized on LCI is divided into three patterns: no redness (LCI-A), redness with visible vessels (LCI-B), and redness without visible vessels (LCI-C), according to the visibility of blood vessels in the background mucosa. The interobserver agreement (kappa value) of the LCI classifications was comparable to or even higher than that of the MES on conventional WLI and was excellent between an expert observer and nonexperts, with no statistically significant difference. In comparisons between MES and LCI classifications, the areas with an MES of 0 (i.e., no mucosal redness diagnosed on conventional WLI) were recognized as LCI-B (41.8%) or LCI-C (4.6%), indicating the identification of mucosal redness on LCI. Among the areas diagnosed with an MES of 1, 60.5% were recognized as LCI-C, 34.6% as LCI-B, and 4.9% as LCI-A. According to these data, inactive colonic mucosa with an MES of 0 or 1 on WLI could be further subdivided by LCI, suggesting that the LCI classification was independent of the MES. Furthermore, Takagi et al. [20] subdivided the LCI classification of colonic mucosa diagnosed with an MES of 0 on conventional WLI into two classes, LCI-A and LCI-B, and compared the relapse rate and histologic activity between the two groups. They showed that the relapse rate of patients with LCI-A was significantly higher than in those with LCI-B over a 48-month follow-up period, even in patients with quiescent UC with an MES of 0 on conventional WLI (100%, 9/9 vs. 52.9%, 9/17, respectively, p = 0.018). However, there was no significant difference in the proportion of the LCI-A group and the relapse rate between UC patients with histologically active disease and those with inactive disease determined by a Geboes score. These data suggest that the LCI classification could stratify patients with MES 0 or MES ≤1, which is currently considered as MH, and is useful for identifying those patients with long-term favorable outcomes, even in quiescent UC.

Matsumoto et al. [21] also reported the clinical utility of LCI for the assessment of endoscopic activity and prediction of relapse in patients with quiescent UC. They classified the endoscopic findings of the entire rectum into three groups: WLI−/LCI− (group A, WLI nonredness/LCI nonredness), WLI−/LCI+ (group B, WLI nonredness/LCI redness), and WLI+/LCI+ (group C, WLI redness/LCI redness), based on the presence or absence of obvious planar redness on a combination of WLI and LCI observation. All areas (100%, 109/109) recognized as redness on WLI were identified as redness on LCI observation, whereas 36% (47/129) of the areas recognized as nonredness on WLI were identified as redness on LCI. The histologic analysis revealed that 89% (73/82) of the area with LCI nonredness (group A) had a Geboes score of 0 or 1. During follow-up of more than 12 months after initial colonoscopy, 71.4% (10/14) in group C, 28.6% (4/14) in group B, and none in group A had a clinical relapse. The nonrelapse rate of the patients significantly correlated with the WLI/LCI classification but not with the MES.

Kanmura et al. [22] also studied the efficacy of LCI for the evaluation of colonic mucosal inflammation in UC. They quantified the color tones of regions of interest on WLI and LCI using the L*a*b* color value (L*, black-white axis; a*, red-green axis; b*, yellow-blue axis), and investigated the correlations between L*a*b* color values (WLI-l, WLI-a, WLI-b, LCI-l, LCI-a, LCI-b) and histologic healing defined as a Geboes score ≤2. They showed that only LCI-a values were significantly correlated with the score for histologic healing. The LCI color difference between histologic inflammation and noninflammation was more than twice the average difference on WLI. These data suggest that LCI is more useful than WLI for the assessment of mucosal inflammation in UC. Furthermore, Kanmura et al. [23] showed that over a 1-year follow-up period, the relapse rate of patients with scarlet-colored lesions on LCI was significantly higher than that of patients without scarlet-colored lesions (37.0%, 10/27 vs. 6.5%, 2/31, p = 0.02). However, there was no significant difference in the relapse rate at 1-year follow-up between MES 0 and MES 1 patient groups according to WLI observation.

LCI has superior brightness to other IEE technologies such as NBI and blue-laser imaging, which can help illuminate the wide lumen in the colorectum. Therefore, an advantage of LCI is that it can provide a distant view sufficient to evaluate the mucosal inflammation of the entire colon.

i-Scan

The i-scan method is a new image-enhancing endoscopic technique developed by HOYA/PENTAX (Tokyo, Japan) [24] that is classified as a digital contrast method [12]. The i-scan method used three different algorithms for image enhancement: surface enhancement, contrast enhancement, and tone enhancement (TE) [24]. TE divides the individual red-green-blue (RGB) components of normal endoscopic images and performs a selective modification of the color frequencies of each component, which is then followed by image reconstruction in real-time. Thus, TE is designed to emphasize minute mucosal structures, MVPs, and subtle color changes. Utilizing this strength of the TE mode of i-scan, Iacucci et al. [25] reported that the subtle histologic abnormalities underlying the healed mucosa in patients with UC of MES 0 could be detected more effectively using HD colonoscopy with i-scan. Neumann et al. [26] showed that visualization of MVPs and mucosal abnormalities in the colon by virtual chromoendoscopy with the i-scan significantly improved the diagnosis of the severity and extent of mucosal inflammation in quiescent or mildly active IBD in comparison with high-definition white-light endoscopy. However, the assessment of MH in these reports was subjective and not quantitative.

Honzawa et al. [27] retrospectively reviewed the data from quiescent 52 UC patients who had undergone routine colonoscopy with the i-scan TE-c system. Using the TE-c enhanced images of normal and inflamed colonic mucosa captured with the improved color contrast, the degree of inflammation was quantified for the entire screen by correlating the value with the reference value for each pixel in the HSV (hue, saturation, value) color space. Then, the authors defined the Mucosal Analysis of Inflammatory Gravity by i-scan TE-c Image (MAGIC) score as the mean value of the quantified values for each pixel and investigated the associations between the MAGIC score, MES, and histologic activity (Geboes score). This analysis showed that the MAGIC score of the MES 1 group was significantly higher than that of the MES 0 group (779.8 ± 488.4 vs. 487.2 ± 378.2, p = 0.0034), and that it was significantly correlated with the Geboes score (r = 0.468, p = 0.015). Similar to the previous study by Iacucci et al. [25], their data indicated wide variations in the MAGIC score among patients with the same MES of 0 or 1. These data suggest that the i-scan TE-c system can be used to stratify patients with MES 0 or MES ≤1 according to histologic activity, without requiring a biopsy procedure (Fig. 3).

Fig. 3.

The representative i-scan TE-c enhanced images (middle) and MAGIC scores (bottom) in the MES 0 or MES 1 group. As the degree of mucosal inflammation increased, the calculated score based on the mapping images increased. MES, Mayo endoscopic subscore.

/WebMaterial/ShowPic/1471533Autofluorescence Imaging

Autofluorescence imaging (AFI) is a novel endoscopic technique that involves detection of the autofluorescence emitted following the excitation of endogenous fluorescent substances (fluorophores) in intestinal tissues, with these fluorophores being mainly released from type-I collagen. Since the intensity of fluorescence on AFI is influenced by various in vivo conditions, such as differences in the constitution of fluorophores, tissue architectures, and epithelial permeability, AFI video-endoscopy is expected to be useful for distinguishing between neoplasia and non-neoplasia and for assessing the severity of histologic inflammation. Fujiya et al. [28] categorized AFI images of colonic mucosa in patients with UC into four types: green, green with purple spots, purple with green spots, and purple, and showed that these AFI findings correlated with histologic severity. Osada [29] evaluated correlations between the red-green-blue color component of AFI and WLI findings in 572 endoscopic images from a total of 286 areas in patients with UC. They showed that the green color component of AFI corresponded more closely with mucosal inflammation (MES) than the red or blue color components. In addition, as the severity of the mucosal inflammation increased, the abundance of the green color component on AFI decreased. Furthermore, Moriichi et al. [30] reported that the fluorescence index (F index), defined as the ratio of the reverse gamma value of green (autofluorescence) divided by that of red (reflection), can be calculated using image analysis software and used to quantify the intensity of the AFI signal and predict the histological severity with high diagnostic accuracy (84.7 vs. 78.5% on conventional endoscopy, p < 0.01).

Thus, although the major advantage of AFI is its simplicity in the evaluation of simple color changes, it can also be evaluated more objectively by using image analysis software to quantify the color changes. However, it is sometimes difficult to distinguish between mucosal inflammation and ulcer scars (which are depicted as purple) when reflecting the histologic abnormalities in the submucosal layer. The intensity of the AFI signal also varies according to different conditions, such as bowel dilatation and observation direction.

Dual-Red Imaging

Dual-red imaging (DRI) is a novel IEE technology that uses three wavelengths for illumination (540, 600, and 630 nm). The 600-nm and 630-nm light improve the visibility of blood vessels in submucosal tissues and bleeding points. Naganuma et al. [31] investigated the relationships between endoscopic severity (MES), histologic activity (Geboes score), and DRI score in patients with UC. Patients were classified into four groups on the basis of the DRI score: DRI 1, 2, 3, and 4, according to the degree of the damage to the surficial and/or deep vessels in the colonic mucosa. They showed that the DRI score significantly correlated with the MES in analyses by both experts and nonexperts (r = 0.695–0.778; p < 0.001 for each investigator). Interestingly, the interobserver agreement (k value) for the DRI score was significantly higher than the agreement for the MES (mean k value: 0.76 vs. 0.53, p < 0.001). The DRI score significantly correlated with histologic activity, and the Kaplan-Meier analysis revealed that the expected time until clinical recurrence was significantly longer in patients with lower DRI scores.

Usually, the vascular pattern in the surface of the colonic mucosa is partially or completely absent in the active phase of UC, meaning that it is hard to assess MVPs using conventional WLI. However, compared with WLI, DRI can enhance the vascular pattern and allow easy identification of the blood vessels in deeper tissues (Fig. 4). Therefore, DRI may be more useful for evaluating colonic inflammation and predicting the prognosis in patients with endoscopically mild to moderately active UC.

Fig. 4.

The representative pictures of WLI (left) and DRI (right) of colonic mucosa in UC patients. The DRI can detect the blood vessels in deeper tissues more clearly than conventional WLI. WLI, white-light imaging; DRI, dual-red imaging; UC, ulcerative colitis.

/WebMaterial/ShowPic/1471531Conclusions

As the importance of IBD treatment based on T2T strategies is recognized, it is widely accepted among IBD physicians that MH is a treatment target in UC. However, no standardized definition of MH has been established to date. While an MES of 0 and UCEIS of ≤1 assessed using conventional WLI are currently used as criteria for MH in most clinical trials, these are subjective measures, and intra- and interobserver variations remain important clinical challenges in the endoscopic assessment of UC. Furthermore, it needs to be noted that microscopic inflammation can exist even in colonic mucosa that are endoscopically quiescent under WLI observation. Considering the accumulating evidence that IEE findings correlate with histologic findings that are difficult to assess by conventional WLI, the IEE observation of UC could stratify UC patients with MH on conventional endoscopy and contribute to improving T2T-based therapeutic strategies.

Acknowledgment

We thank Karl Embleton, PhD, Edanz (https://jp.edanz.com/ac) for editing a draft of the manuscript.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

No external funding was supported for the preparation of the manuscript.

Author Contributions

Minoru Matsuura and Tadakazu Hisamatsu contributed to the conception and design of the manuscript. Minoru Matsuura drafted manuscript. All authors contributed to data interpretation. Daisuke Saito, Jun Miyoshi, and Tadakazu Hisamatsu contributed to the critical revision of the manuscript. All authors gave final approval for the manuscript to be published.

Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.

留言 (0)

沒有登入
gif