This is a prospective single-arm, observational study conducted at Chiba University Hospital. The study complied with the Declaration of Helsinki, and the study protocol was approved by the Institutional Review Board of Chiba University Hospital (approval no. 3741). All study participants signed an informed consent statement before study enrollment.

UC patients with clinical remission who underwent colonoscopy with evaluation　using WLI and IEE from March 2021 to September 2023 were enrolled in the study. Of the patients, those with a follow-up period of > 3 months after colonoscopy were included in the analysis. All patients were diagnosed with UC using endoscopic, histologic, and clinical criteria at our institution. Clinical remission was defined as partial Mayo score ≤ 2 [13]. The exclusion criteria were (1) patients who were under 18 years old, (2) patients in whom biopsy specimens could not be obtained due to comorbidities or impossibility of discontinuation of anticoagulant or antiplatelet medications, and (3) patients who refused to participate in the study.

All patients underwent colonoscopy at Chiba University Hospital. Bowel preparation was performed with oral polyethylene glycol (Moviprep; EA Pharma, Tokyo, Japan). All examinations were performed with a PCF-H290ZI or CF-HQ290 scope connected to an EVIS X1 system (Olympus Corporation, Tokyo, Japan). In this study, disease activity was assessed using WLI, TXI, and RDI in the six areas: cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum. After insertion into the cecum, still images of WLI, TXI, and RDI were taken sequentially at each segment while the endoscopy withdrawing.

TXI is an image enhancement technique that optimizes structure, color, and brightness, leading to improved observability of lesions with subtle color changes [11]. TXI has two modes (TXI mode 1 and 2), and in this study TXI mode 1, of which enhancement of color contrast is superior to that of mode 2 [14, 15], was used. On the other hand, under RDI for highlighting the luminal vessels, deep vessels are shown to be green in color, and superficial vessels are shown to be brown in color. There are three modes in RDI (RDI mode 1–3). In this study, we used RDI mode 3, which emphasizes both thick blood vessels in deep tissues and relatively thin blood vessels in the mucosa [16, 17].

A biopsy sample was collected from the area with the most severe inflammation which was evaluated with MES. If the endoscopic activity was even throughout the colon, the biopsy sample was obtained from the rectum.

MES and UCEIS were evaluated according to the previous reports [5, 8]. The TXI and RDI were evaluated with the scores used in the previous reports. TXI scores were classified into three grades, TXI score 0, 1, and 2, based on redness enhancement and vessel visibility; TXI score 0 = no accentuated redness, TXI score 1 = accentuated redness, and TXI score 2 = accentuated redness and poor visibility of deep vessels [11] (Fig. 1).

TXI score. TXI score 0 = no accentuated redness. TXI score 1 = accentuated redness. TXI score 2 = accentuated redness and poor visibility of deep vessels

On the other hand, RDI scores were classified into four grades according to the visibility of deep and superficial vessels. RDI score 1 was defined as a typical vascular pattern in both the superficial (brown) and deep (green) vessels. RDI score 2 was defined as the patchy or complete obliteration of the brown vessels with the clear reorganization of the green vessels. RDI score 3 was defined as patchy obliteration of the green vessels, and RDI score 4 was defined as the complete obliteration of the green vessels [12] (Fig. 2).

RDI score. RDI score 1 = a typical vascular pattern in both the superficial (brown) and deep (green) vessels. RDI score 2 = the patchy or complete obliteration of the brown vessels with the clear reorganization of the green vessels. RDI score 3 = patchy obliteration of the green vessels. RDI 4 = the complete obliteration of the green vessels

These endoscopic inflammatory activity scores were scored using still images taken during endoscopy by each endoscopist who performed the endoscopy, and the endoscopic score was further confirmed independently by another endoscopists who was blind to the scoring results of the endoscopist who performed endoscopy. All of the endoscopists were a board certified gastroenterologist of the Japanese Society of Gastroenterology, a board certified fellow of the Japan Gastroenterological Endoscopy Society, or a board certified trainer of the Japan Gastroenterological Endoscopy Society. They had been engaged in endoscopy for > 7 years, and had experience in endoscopic evaluation of IBD using IEE in more than 30 cases. If the endoscopic scores were not consistent between the two evaluators, the final endoscopic scores were determined by discussion between them. Each of the six segments of the colon (cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum) was evaluated with each score, and the scores in the part of the colorectum with the maximum MES were used for analysis.

Fecal calprotectin was measured in feces that were collected before bowel preparation. The fecal samples were sent to the testing laboratory (SRL, Tokyo, Japan), and the value of fecal calprotectin was measured using enzyme linked immunosorbent assays (ELISA).

An expert and experienced pathologist, who was blind to the endoscopic results, assessed all biopsy specimens and scored pathological inflammatory activities using the Geboes score. The Geboes score characterizes disease severity across six grades with each grade further divided into either four or five subcategories (e.g., 0.0, 0.1, 0.2 and 0.3, and 5.0, 5.1, 5.2, 5.3 and 5.4). This creates 29 ordered, categorical scores between 0 and 5.4, with higher scores denoting greater levels of disease activity [18, 19]. In this study, the six grades were assessed separately for each biopsy sample in line with the original sources [18]; Grade 0 = architectural changes, Grade 1 = chronic inflammatory infiltration; Grade 2A = eosinophil in lamina propria; Grade 2B = neutrophil in lamina propria; Grade 3 = neutrophils in epithelium; Grade 4 = cryptic destruction; Grade 5 = erosions and ulcerations.

The primary endpoint of this study was to evaluate the cumulative relapse of UC according to MES, UCEIS, and the TXI and RDI scores. Relapse of UC was defined as the need for the change of medications due to worsening clinical symptoms with the increase in partial Mayo score ≥ 2 according to the definitions in previous reports [20, 21]. Because biomarkers including fecal calprotectin were not always measured for patients with relapse and none of those patients required surgery or hospitalization, the abovementioned definition was applied in this study.

The secondary endpoints were to examine the correlation between the novel IEE scores (TXI score and RDI score), and the WLI scores (MES and UCEIS), the value of fecal calprotectin and pathological findings.

Relationships between endoscopic scores and fecal calprotectin or histological scores were tested using Spearman's rank correlation coefficient. p values of < 0.05 were considered statistically significant. Kaplan–Meier analysis and log-rank test were used to analyze the cumulative relapse-free rate. The hazard ratio (HR) and its 95% confidence interval (CI) for the risk of clinical relapse were calculated with the univariate Cox regression model. These statistical analyses were performed using SPSS software, version 29 (IBM Corp., Armonk, New York, United States).