Digestion最新文献

Background: The Fecal Immunochemical Test (FIT) is a wide available fecal biomarker that could evaluate the disease activity in IBD. The aim is to assess the correlation between the FIT and fecal calprotectin (FC) for evaluating IBD activity.

Methods: Unicentric, transversal cohort study. Consecutive patients with IBD were included and FIT and FC were determined. The clinical activity was assessed with Truelove-Witts and Yamamoto-Furusho index for UC patients while Harvey-Bradshaw and CDAI for CD patients. Spearman's rank correlation test was used to assess the correlation between FIT and FC. Sensitivity, specificity, and positive and negative predictive values for FIT and FC were calculated. Receiver operator curves were constructed.

Results: A total of 206 patients were included. One hundred forty-eight (72%) patients had diagnosis of UC and 58 (28%) with CD. The median of FIT was 2.8 g/g (range, 2.6 - 2394 g/g) and the median for FC level was 265.5 g/g (range, 22 - 6285 g/g). There was a very good correlation between FIT with and FC in UC patients (rs= 0.745, P < 0.01) and moderate in CD patients (rs = 0.574, P < 0.01). A FIT cutoff of 2.6 g/g identified endoscopic activity in UC patients with a sensitivity of 78%, specificity of 86%, PPV of 91% and NPV of 67% with an AUC of 0.852 (95% IC 0.758-0.946).

Conclusion: FIT can be an alternative fecal biomarker to assess the disease activity in UC patients.

Introduction European and national Celiac Disease (CeD) guidelines offer an easy pathway to diagnose CeD. The German CeD Registry aimed to assess symptoms and clinical findings before diagnosis, diagnostic delay, care during the diagnostic process, and factors associated with persistence of symptoms. Methods Individuals with CeD provided demographic, clinical and healthcare-related information. Participants were divided into four subgroups according to age at diagnosis (>18 or <18 years) and year of diagnosis (before and since 2012). Factors associated with symptoms after at least 1-year on a gluten free diet (GFD) were assessed using multivariate logistic regression. Results From 11/2019 to 10/2021, 2333 participants were enrolled. After exclusion of 169 (7.2%), 2164 remained for analysis, thereof 796 (36.8%) were diagnosed <18 years, and 1283 (59.3%) since 2012. Most common symptoms before diagnosis included abdominal pain (83%), bloating (82%), fatigue (78%), and diarrhoea (71%). Diagnostic delay after 2012 was longer in adults than children (median 4.4 years [IQR 1.2-13.0] versus 1.1 [IQR 0.5 - 2.2], respectively) (p<0.001). Guideline-conform diagnoses increased over time. After diagnosis, only 60% received professional dietary counselling. Factors associated with symptoms despite GFD included female gender (OR 1.79 [95%CI 1.34; 2.40], p<0.001), same symptom before diagnosis (OR 3.45 [2.45; 4.96], p<0.001), insufficient information provided at diagnosis (OR 1.25 [1.00; 1.57], p=0.046), and age at diagnosis (per decade) (OR 1.11 [1.04;1.18], p<0.001), but not time since diagnosis. Conclusions Our findings revealed deficits in awareness, the diagnostic process, and post-diagnostic care that are linked to decreased clinical improvement over time.

Introduction: The Mayo endoscopic score (MES) is used widely in ulcerative colitis (UC) for severity assessment and therapeutic decision-making. Deep learning (DL) models developed to determine MES currently lack explainability. We aimed to develop explainable models for the MES in patients with UC and examine the human-artificial intelligence interactions with the models.

Methods: This was a retrospective multicenter study conducted across four large tertiary institutions in China. A total of 2,600 white-light images were used for training. Two approaches were adopted: traditional blackbox or explainable AI (XAI). The trained models were evaluated with three external test datasets (#1 Changshu & Jintan hospitals, n = 100; #2 HyperKvasir, n = 100; #3 Yongding hospital, n = 260), and the performance was compared with endoscopists. The primary outcome was the performance of 4-way classification. For explainability, moreover, Grad-CAM was for computer vision, while local interpretation, variable importance, and partial dependence plots were for the classifier within XAI.

Results: In the test #1 dataset, a Xception-backboned XAI showed accuracy of 0.910, Matthew's correlation coefficient 0.880 and Cohen's kappa 0.960 [95% CI, 0.940 - 0.990]. The metrics were better than other models, as well as the two endoscopists. With the AI-assistance, the performance of endoscopists were improved (senior's accuracy from 0.890 to 0.930 and junior's accuracy from 0.810 to 0.880). Similar trend was observed in the test #2 and #3 datasets.

Conclusion: The use of an explainable framework empowers AI models to achieve improved performance with transparency. XAI can also improve endoscopist performance in interpretation of MES in UC.

Introduction: In the surveillance of esophageal squamous cell carcinoma (ESCC), advanced lesions may still be detected despite regular screening with esophagogastroduodenoscopy (EGD). In this study, we investigated the endoscopic characteristics and prognosis of ESCC cases that progressed to pT1a-MM or deeper despite undergoing surveillance EGD.

Methods: We retrospectively analyzed 225 consecutive superficial ESCC lesions invading beyond the muscularis mucosa that were resected by endoscopic submucosal dissection (ESD) from 215 patients at Hiroshima University Hospital between April 2010 and March 2023. Among them, 28 patients (29 lesions) were classified as the post-EGD ESCC (PEESCC) group, defined as cases where surveillance EGD performed 24 months before diagnosis did not detect neoplasia or carcinoma. The remaining 188 patients (196 lesions) were the screening group. Subsequently, endoscopic findings and prognosis were compared.

Results: From the multivariate analysis, the presence of Lugol-voiding lesions (69.0% vs. 39.3%), cervical esophageal location (17.2% vs. 3.1%), small tumor diameter (21.6±12.3 mm vs. 34.0±18.3 mm), and submucosal tumor (SMT)-like elevation (20.7% vs. 7.7%) were significantly identified as characteristic endoscopic findings of PEESCC. The PEESCC group exhibited lower 5-year disease-specific survival (93.1% vs. 98.5%, p=0.039) and recurrence-free survival (69.0% vs. 83.7%, p=0.025).

Conclusion: PEESCC lesions are associated with distinct endoscopic features and a poorer prognosis than are non-PEESCC lesions.

Introduction: Particle radiotherapy (PRT) is a new option for the treatment of unresectable pancreatic cancer (PC). While gastrointestinal bleeding (GIB) is a feared adverse event, real-world evidence in this setting is limited.

Methods: We conducted a single-center retrospective study to evaluate the frequency and outcomes of GIB and to elucidate the risk factors for GIB after PRT for PC.

Results: Thirty-four patients were included. Twenty-nine received PRT to the pancreatic primary, while five received PRT for metastases. Concurrent chemotherapy was given to 26 patients (76%). Eleven patients (32%) experienced acute GIB symptoms after PRT. Median time from PRT to GIB was 13.2 months. Endoscopic signs of hemorrhage were observed in eight patients (24%), and endoscopic hemostasis was performed in six (18%). Three cases presented with ruptured pseudoaneurysms, of which two were treated with transarterial embolization. Hemostasis was ultimately achieved in all cases, and no deaths occurred directly as a result of GIB. However, overall survival after GIB was short (median: 1.9 months). Median overall survival after PRT tended to be longer in bleeders than in non-bleeders (26.8 vs. 22.7 months, P = 0.03). Concurrent chemotherapy was associated with a lower risk of GIB in univariate analysis (P = 0.05).

Conclusion: Gastrointestinal bleeding after particle radiotherapy may not be as rare as previously believed, particularly in the terminal stages of pancreatic cancer.

Background: In recent years, several studies have described the clinicopathological characteristics of Helicobacter pylori (H. pylori) -uninfected gastric cancer. This entity is now recognized as one of the major topics in gastric cancer research and clinical practice.

Summary: Currently, H. pylori-uninfected gastric epithelial neoplasms (HpUGENs; excluding adenocarcinomas of the esophagogastric junction and gastric neuroendocrine tumors) are classified into seven subtypes in our research results: raspberry-type gastric epithelial neoplasm (GEN; foveolar-type adenoma), whitish flat elevated-type GEN (GEN with gastric phenotype), gastric adenocarcinoma of fundic-gland type (GA-FG), gastric adenocarcinoma of the fundic-gland mucosa type (GA-FGM), other GEN with a gastric phenotype (complex type of GEN with gastric phenotype), GEN with an intestinal or gastrointestinal mixed phenotype arising in the pyloric gland region, and signet-ring cell carcinoma (SRCC).

Key messages: This study outlines our analysis of current cases, detailing the endoscopic and clinicopathological characteristics of HpUGENs, and provides practical insights for their endoscopic and pathological diagnosis. Since many of these neoplasms histologically show low-grade atypia, they are sometimes diagnosed as gastric adenoma or gastric dysplasia rather than adenocarcinoma in the World Health Organization classification, highlighting the need for standardized histopathological diagnostic criteria of gastric epithelial neoplasms with low-grade atypia. Moreover, as no clinical practice guidelines have yet been established for HpUGENs, future research should aim to elucidate the relationship between early and advanced lesions, perform comprehensive analyses of H. pylori-uninfected advanced gastric cancer, and conduct molecular biological studies to achieve a better understanding of the entire disease spectrum and to establish evidence-based clinical guidelines.

Introduction: Early use of biologics improves outcomes for Crohn's disease (CD). Evidence now indicates that initiating therapy within 6 months of diagnosis - the "very early" window - may yield additional benefits over the traditional ≤2-year target. We therefore compared 1-year outcomes after very early (<6 months) versus early (6-24 months) biologic initiation in routine practice.

Methods: In this retrospective cohort (March 2018 to June 2025), biologic-naïve adults with CD and ≥52 weeks of follow-up were stratified by time from diagnosis to first biologic: very early (<6 months) or early (6-24 months). The primary endpoint was steroid-free clinical remission at week 52. Multivariate logistic regression identified variables independently associated with remission.

Results: Ninety-six patients were analyzed (very early = 52; early = 44). Baseline characteristics were comparable except for a higher proportion of corticosteroid use in the very early group (67.3% vs. 43.2%; p = 0.018). At week 52, very early initiation was associated with a lower mean CD Activity Index (64.82 ± 6.79 vs. 96.10 ± 13.03; p = 0.038) and a higher steroid-free clinical remission rate (71.2% vs. 45.5%; p = 0.011). Concomitant corticosteroid use fell to 11.4% in the very early group versus 30.6% in the early group (p = 0.033). Very early initiation remained the strongest independent predictor of steroid-free remission (adjusted OR 3.537, 95% CI: 1.417-8.824; p = 0.007).

Conclusions: Initiating biologic therapy within 6 months of CD diagnosis significantly increases 1-year steroid-free clinical remission and reduces corticosteroid dependence compared with initiation at 6-24 months. These real-world data support adopting a standardized "very early" biologic treatment strategy to optimize clinical outcomes in newly diagnosed, biologic-naïve CD.

Introduction: Sedation protocols for balloon-assisted enteroscopy (BAE) are not yet standardized. The aim of this study was to compare efficacy and safety between continuous infusion and intermittent bolus administration of midazolam for sedation during BAE. The study hypothesis was that continuous infusion would provide a greater proportion of time under conscious sedation than would intermittent bolus administration.

Methods: We conducted a multicenter, prospective, double-blind, randomized controlled trial at 15 institutions of the National Hospital Organization in Japan. Patients scheduled for diagnostic or therapeutic BAE were randomly assigned to receive continuous infusion or intermittent bolus administration of intravenous midazolam. The primary endpoint was the proportion of time under conscious sedation, defined as a Ramsay Sedation Scale (RSS) score of 3-4. Secondary endpoints included body movements causing procedure interruption, endoscopist and patient satisfaction, total drug dosage, adverse events, and termination of the procedure.

Results: Of 76 enrolled patients (39 continuous infusion group, 37 intermittent bolus group), one from each group discontinued before treatment, leaving 74 patients (38 continuous group, 36 intermittent bolus group) for analysis. The proportion of time under conscious sedation was comparable between groups (mean±SD: 0.637±0.315 vs. 0.609±0.272, p=0.721). However, the continuous infusion group showed a lower incidence of body movements causing procedure interruption (7.9% vs. 25.0%, p=0.091). The total midazolam dose was higher in the continuous infusion group, whereas the incidence of adverse events was comparable between the two groups.

Conclusions: Continuous infusion of midazolam did not demonstrate superiority over intermittent bolus administration with regard to the proportion of time under conscious sedation. However, continuous infusion suppressed body movements during BAE without increasing adverse events; thus, it could be one of the feasible sedation options for BAE in clinical practice.

Introduction: The incidence of inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn's disease (CD), is increasing rapidly worldwide. Although multiple advanced therapies are now available, selecting the optimal treatment remains challenging due to the expanding options and diverse healthcare systems.

Methods: We conducted a questionnaire survey among physicians in nine Asian countries prior to the 18th International Gastrointestinal Consensus Symposium (IGICS) to assess the current status of advanced therapies for IBD. The survey included questions regarding therapeutic agent selection, biomarkers, and imaging modalities for monitoring.

Results: Of the 210 respondents, 173 physicians treating IBD were analyzed. Anti-TNFα antibodies remain the most commonly selected advanced therapy for both UC and CD. Elderly patients with UC were more likely to receive anti-α4β7-integrin antibodies or anti-IL-12/23p40 monoclonal antibodies, reflecting safety considerations. Janus kinase inhibitors were used more frequently as a second-line option in severe cases. Comorbidities, drug costs, and lifestyle factors also influenced treatment choice. CRP is the most common biomarker used for monitoring, and endoscopy is the most frequently used imaging modality.

Conclusion: This questionnaire survey revealed the current status of advanced therapies for IBD in nine Asian countries and regions. Region-specific evidence-based algorithms for selecting advanced therapies for IBD should be established.

Background: Barrett's esophagus (BE) is a recognized precursor to esophageal adenocarcinoma (EAC), yet its endoscopic diagnosis remains inconsistent worldwide. This review summarizes current challenges and recent advancements in the endoscopic diagnosis of BE, including updates from international consensus statements and emerging technologies such as image-enhanced endoscopy (IEE) and artificial intelligence (AI).

Summary: This narrative review integrated international guidelines, multicenter studies, expert consensuses, including the Kyoto International Consensus and Asian Barrett Consortium data, and recent trials of diagnostic imaging and quality indicators (QIs) regarding BE surveillance.

Key messages: Discrepancies in defining the gastroesophageal junction (GEJ) - notably between palisade vessels and gastric folds - contribute to the global variability of the BE diagnosis. The Kyoto International Consensus recommends using the distal end of the palisade vessels as a more stable and histologically consistent landmark. Additionally, the Prague C & M criteria offer a standardized approach to measuring the BE length; however, limitations for ultra-short-segment BE exist. IEE modalities such as linked color imaging and red dichromatic imaging enhance GEJ visualization, whereas AI systems have the potential for automated BE classification. QIs such as the neoplasia detection rate, inspection time, and adherence to biopsy protocols have been proposed to improve diagnostic consistency and outcomes. Standardizing the endoscopic definition of BE and adopting quality-based surveillance strategies are essential to improving detection and reducing variability. Incorporating IEE- and AI-based tools into routine practice may support a more reliable and efficient diagnostic pathway for BE, thus facilitating early EAC detection and prevention worldwide.