Introduction

Recent advances in large language models, such as GPT-4.0 and GPT-4.5, have increased the accessibility of artificial intelligence (AI) in clinical settings. However, its role in decision-making remains unclear.

Aims & Methods

This study compares the appropriateness of esophagogastroduodenoscopy (EGD) prescriptions across ten clinical scenarios, as evaluated by gastroenterology specialists, residents, general practitioners, and five AI systems (ChatGPT-4.0, ChatGPT-4.5, Gemini, Claude, and OpenEvidence), using ESGE and ASGE guidelines as the reference standard. A cross-sectional survey was developed using 10 real-world clinical vignettes in gastroenterology, each with three response options: appropriate, inappropriate, or uncertain. The vignettes were developed based on real-world clinical practices by three expert gastroenterologists, and then reviewed by the authors.The final questionnaire was hosted on Google Forms and simultaneously distributed—without prior notice—to gastroenterology specialists, residents, and general practitioners across Italy to minimize external influence. The same clinical scenarios were then submitted to five AI systems (ChatGPT-4.0, ChatGPT-4.5, Gemini, Claude, and OpenEvidence) using the same multiple-choice format. AI responses were collected on April 22, 2025.

Results

A total of 122 healthcare professionals participated in the study (61% male; mean age 39.7 ± 13.6), including 41 gastroenterology specialists, 31 residents, and 50 general practitioners. Most participants were from northern Italy (76%), with additional representation from central (14%) and southern regions (10%), covering a total of 10 regions. The majority had less than 5 years of clinical experience (66%) and were employed in public hospitals (49%). Only 30% reported prior use of AI in clinical practice. All respondents considered the clinical scenarios realistic.
The average accuracy rate was highest among gastroenterologists (65.6% ± 11.6) and residents (64.8% ± 10.0), compared to general practitioners (56.8% ± 11.0), with statistically significant differences (p < 0.001 and p < 0.005, respectively).
Among AI systems, OpenEvidence and Gemini outperformed all human groups (80% and 80%, p < 0.001), while ChatGPT-4.0 (70%) and Claude (60%) showed comparable performance to clinicians. Notably, ChatGPT-4.5 (50%) underperformed relative to all human categories.

Conclusion

This study highlights significant variability in the appropriateness of upper GI endoscopy prescriptions among different medical professionals, with gastroenterologists performing better than general practitioners. Emerging AI models, particularly OpenEvidence and GEMINI, demonstrated promising performance in clinical decision-making, outperforming all human groups in certain scenarios. These AI systems could serve as valuable tools to support gastroenterologists and other healthcare professionals, enhancing adherence to evidence-based practices and facilitating more direct consultation of clinical guidelines. Further research is needed to explore the integration of AI into clinical workflows and its potential to refine decision-making in gastroenterology.

References

1) Chang PW, Amini MM, Davis RO, et al. ChatGPT4 Outperforms Endoscopists for Determination of Postcolonoscopy Rescreening and Surveillance Recommendations. Clin Gastroenterol Hepatol. 2024 Sep;22(9):1917-1925.e17. doi: 10.1016/j.cgh.2024.04.022. Epub 2024 May 9. PMID: 38729387.
2) Maida M, Celsa C, Lau LHS, et al. The Application of Large Language Models in Gastroenterology: A Review of the Literature. Cancers (Basel). 2024 Sep 28;16(19):3328. doi: 10.3390/cancers16193328. PMID: 39409948; PMCID: PMC11475222.

Introduction

Endoscopic submucosal dissection (ESD) is being increasingly adopted worldwide for the treatment of colorectal neoplastic lesions. However, its role in patients with inflammatory bowel disease (IBD) remains debatable, particularly due to the increased risk of submucosal fibrosis from chronic inflammation, which can theoretically impact the clinical outcomes of this technique. This study aimed to assess the efficacy and safety of ESD for colorectal neoplastic lesions in IBD patients, comparing Asian and Western settings.

Aims & Methods

A systematic review was conducted by searching MEDLINE and Web of Science until January 2025, in accordance to MOOSE guidelines. We included primary studies reporting ESD performance in IBD patients. We performed random-effects meta-analysis for several outcomes, including en-bloc, R0, and curative resection rates, frequency of adverse events (post-procedural bleeding and perforation), local recurrence and need for additional surgery after ESD. Certainty of evidence was assessed using GRADE.

Results

We included 20 studies (565 patients and 732 lesions). The meta-analytical frequencies of en-bloc, R0, and curative resection rates were 95.7% (95%CI=93.5-97.9%; I²=60.1%), 84.6% (95%CI=79.4-90.1%; I²=68.9%), and 83.9% (95%CI=76.3-92.4%; I²=75.9%), respectively. Bleeding and perforation occurred in 7.0% (95%CI=4.4-11.2%; I²=16.9%) and 7.9% (95%CI=5.3-11.9%; I²=2.4%) of cases, respectively. Local recurrence frequency was 5.3% (95%CI=3.5-7.9%; I²=0%). The meta-analytical need for additional surgery was 13.3% (95%CI=9.5-18.1%; I²=42.5%). No differences were found between Asian and Western studies for any outcome. The certainty of evidence was considered "low" in 4 outcomes and "very low" in the remaining outcomes.

Conclusion

ESD appears feasible and safe for treating IBD-associated colorectal neoplasias. Similar results were observed between Asian and Western studies.

Introduction

Lynch syndrome (LS) is associated with an increased risk of upper gastrointestinal (GI) cancers. While upper GI surveillance is recommended in several guidelines, there remains international variability and ongoing debate regarding its implementation, timing, and extent. This study evaluated the incidence of gastric (GC), duodenal (DC), and jejunal cancers (JC) in LS patients using data from the German Consortium for Familial Intestinal Cancer (GCFIC). In addition, the diagnostic yield of esophagogastroduodenoscopy (EGD) and push enteroscopy (PE), performed at a tertiary care center, was retrospectively analyzed.

Aims & Methods

A total of 960 endoscopic procedures (494 EGD, 466 PE) were retrospectively analyzed in 349 LS patients documented at a tertiary care center. Push enteroscopy (PE) was performed using a colonoscope to allow deeper insertion into the small intestine, enabling inspection of the proximal jejunum. Clinically relevant findings included intestinal metaplasia, Barrett’s esophagus, adenomas, and carcinomas. These findings were systematically compared between EGD and PE to evaluate diagnostic yield. The GCFIC database was evaluated to determine cancer incidence by genotype.

Results

In the GCFIC cohort (n=2859), the majority of GC (2.2%), DC (2.1%), and JC (1.9%) occurred in high-risk pathogenic variant carriers (PGV) (MLH1, MSH2, EPCAM). However, a small proportion of cancers was also observed in low-risk PGV carriers: GC in two MSH6 carriers, DC in five patients with MSH6 or PMS2, and JC in seven low-risk carriers (MSH6: 3, PMS2: 4). In the NZET cohort, 22% of procedures and 50 % of patients revealed relevant findings. Gastric adenomas were found in 3.7%, gastric cancer in 1.1%, and intestinal metaplasia in 28% of patients. Duodenal adenomas were found in 3.7% and cancer in 0.9%. Jejunal neoplasia was detected exclusively by PE (adenomas 1.7%, cancer 0.6%).

Conclusion

Upper gastrointestinal surveillance is essential in patients with Lynch syndrome. Push enteroscopy (PE) provides diagnostic value beyond conventional esophagogastroduodenoscopy (EGD), particularly in detecting jejunal neoplasia, and should be considered in high-risk surveillance protocols.

Disclosure

Robert Hüneburg was funded by Deutsche Krebshilfe (DECADE), and Wilhelm-Sander-Stiftung.
Jacob Nattermann was funded by Wilhelm-Sander-Stiftung.

Introduction

Minocycline is considered the most potent semi-synthetic tetracycline, demonstrates promising pharmacokinetic and antimicrobial properties against Helicobacter pylori (H. pylori). While previous studies suggested the efficacy of minocycline-containing bismuth quadruple therapy for H. pylori eradication, real-world evidence remains limited.

Aims & Methods

his study aimed to evaluate the clinical efficacy, safety, and cost-effectiveness of minocycline-containing bismuth quadruple therapy in routine clinical practice.This multicenter, real-world observational study was conducted at 25 hospitals in Fujian China between January and March 2025. Patients diagnosed with H. pylori infection received one of three 14-day bismuth-containing quadruple regimens with b.i.d. esomeprazole 20 mg and bismuth 220 mg, plus amoxicillin 1000 mg b.i.d. and clarithromycin 250 mg b.i.d. (PBAC), or minocycline 100 mg b.i.d. and clarithromycin 250 mg b.i.d. (PBMC), or minocycline 100 mg twice b.i.d. and levofloxacin 500mg q.d. (PBML). The primary outcome was the H. pylori eradication rate, secondary outcomes included the frequency of adverse events, patient compliance, and cost-effectiveness.

Results

A total of 347 patients were enrolled, with 110 patients completing treatment in each group. In the intention-to-treat (ITT) analysis, eradication rates were 77.4% (PBAC), 85.5% (PBMC), and 84.3% (PBML). In the per-protocol (PP) analysis, eradication rates were 80.9%, 90.9%, and 88.2%, respectively. Although the differences among the three regimens were not statistically significant, minocycline-based regimens achieved numerically higher eradication rates. Adverse event incidence was 32.73%, 20.91%, and 23.64% in the three groups, respectively. Compliance rates exceeded 94% across all groups. Cost-effectiveness analysis showed PBML had the lowest cost-effectiveness ratio (5.38 CNY per 1% eradication), followed by PBAC (5.41 CNY) and PBMC (6.06 CNY).

Conclusion

H. pylori eradication, with favorable patient compliance and pharmacoeconomic profiles. These regimens present a viable alternative to traditional amoxicillin-or tetracycline-based therapies. Further large-scale randomized controlled trials are warranted to confirm these findings.

Clinical Case Summary

A 57 years-old patient with a history of active smoking, moderate body overweight (BMI 29 Kg/mq) and mild systolic hypertension in treatment with Olmesartan complained history of chronic diarrea and abdominal pain since 2006, rarely associated with hematochezia. After replacement of antihypertensive therapy to eliminate a potential confouding factor, he underwent multiple colonoscopies through the years with nonspecific findings until 2017, when moderated discontinuos inflammation with multiple erosions was seen in the transverse colon and in the rectum. Histologic examination described presence of chronic inflammation with cryptic microabscesses and cryptitis suggetive for Crohn's disease, while intestinal ultrasound revealed no small bowel localizations. He was treated with 2.4 daily grams of oral mesalamine with benefit until the end of 2023, when he came up to our clinic complaining resumption of bloody diarrhoea and abdominal pain, with 3-4 liquid stools per day. Addition of topic mesalamine was not of benefit. The patient underwent another colonoscopy which confirmed presence of hyperemic mucosa with diffuse erosions in the rectum, with histology confirming active Inflammatory Bowel Disease. In march 2024, after neurological investigations for recurrent paresthesias and strength deficits, the patient was diagnosed with relapsing-remitting Multiple Sclerosis. After multidisciplinary consultation, negative cardiologic evaluation, chest X-ray and blood screening exams, in October 2024 the patient started the anti sphingosine-1 phosphate Ozanimod, which was well-tolerated at maintenance dose of 0.92 milligrams per day. Three months after, he came back for scheduled follow-up reporting 1 or 2 daily evacuations in absence of abdominal pain and rectal bleeding. Calprotectin levels droped up to 17 ug/g. He also described complete neurological remission and complete absence of drug side effects.

References

1) Feagan BG, Schreiber S, Afzali A, Rieder F, Hyams J, Kollengode K, Pearlman J, Son V, Marta C, Wolf DC, D'Haens GG. Ozanimod as a novel oral small molecule therapy for the treatment of Crohn's disease: The YELLOWSTONE clinical trial program. Contemp Clin Trials. 2022 Nov;122:106958.
2) Pérez-Jeldres T, Alvarez-Lobos M, Rivera-Nieves J. Targeting Sphingosine-1-Phosphate Signaling in Immune-Mediated Diseases: Beyond Multiple Sclerosis. Drugs. 2021 Jun;81(9):985-1002. Erratum in: Drugs. 2021 Aug;81(12):1451.
3) Harris S, Feagan BG, Hanauer S, Vermeire S, Ghosh S, Yan J, Wu C, Hu Y, Maddux R, Wolf DC, D'Haens G. Ozanimod Differentially Impacts Circulating Lymphocyte Subsets in Patients with Moderately to Severely Active Crohn's Disease. Dig Dis Sci. 2024 Jun;69(6):2044-2054.
4) Sands BE. Positioning Sphingosine-1 Phosphate Receptor Modulators in Inflammatory Bowel Disease. Gastroenterol Hepatol (N Y). 2024 Jul;20(7):417-420.
5) Cree BA, Selmaj KW, Steinman L, Comi G, Bar-Or A, Arnold DL, Hartung HP, Montalbán X, Havrdová EK, Sheffield JK, Minton N, Cheng CY, Silva D, Kappos L, Cohen JA. Long-term safety and efficacy of ozanimod in relapsing multiple sclerosis: Up to 5 years of follow-up in the DAYBREAK open-label extension trial. Mult Scler. 2022 Oct;28(12):1944-1962. doi: 10.1177/13524585221102584. Epub 2022 Jun 28. Erratum in: Mult Scler. 2022.

Introduction

Irritable bowel syndrome (IBS) is a chronic gut-brain disorder that can be responsabil for a variety of uncomfortable gastrointestinal symptoms (laid out in the Rome Criteria) that can reduce quality of life. Although the precise mechanism remains unknown, recent research suggests that an imbalance in intestinal microbiota and a dysfunctional gut-brain axis may be involved in the development of IBS. Because of this, methods to restore the microbiota have been explored as treatment for this condition.

Aims & Methods

Randomized placebo controlled trial 180 patients: active group (probiotic formula, 3g/day- 92, placebo group – 88. men and women aged 24 to 75 who met the Rome IV criteria. One sachet probiotic (3 g) contains 7.5 billion bacteria: Lactobacillus casei W56, Lactococcus lactis W19, Lactobacillus acidophilus W22, Bifidobacterium lactis W52, Lactobacillus paracasei W20 Lactobacillus plantarum W62, Bifidobacterium lactis W51, Bifidobacterium bifidum W23, Lactobacillus salivarius W24. Assessments were made at 4, 8, 12 weeks. All patients agreed to maintain their lifestyle behaviors during their participation in the study. Exclusion criteria: chronic gastrointestinal condition (other than IBS), including inflammatory bowel disease, lactose intolerance, celiac disease, diabetes, neoplasms.Patients have complete weekly data on stool frequency and consistency (Boston scale) and intensity of IBS symptoms (abdominal pain/discomfort, bloating, flatulence/borborygmi, and difficulty defecating) on a Likert scale of 0 (absent) to 7 (severe). Participants completed an IBS-specific quality of life questionnaire (IBS-QoL) at randomization (V1), fallow-up (V2) and at the end of the intervention (V3). Primary endpoint was mean weekly scores of gastrointestinal symptoms. Responders of abdominal pain were defined as subjects who experienced a reduction of at least 30% of the mean daily abdominal score reported at baseline (week 0). Secondary endpoints were the total AUC (W0-W12) of abdominal symptoms (abdominal pain, bloating and flatulence/borborygmi).

Results

​There is a significant reduction in abdominal pain and bloating scores trough all 12 weeks of supplementation versus placebo. A higher overall IBS-QOL score, indicative of better QOL was reported at the end of the intervention (specially for interference with activity P=0.05, body image P=0.040, food avoidance P=0.03)

Conclusion

​The probiotic treatment provides statistically significant and clinically relevant benefits for individuals with IBS. Improves gastrointestinal symptoms and quality of life in IBS patient

References

Rome IV criteria-Rome Foundation

Introduction

Approximately 50% of patients with Ulcerative Colitis (UC) suffer from daily symptoms of faecal urgency, significantly impacting their quality of life¹. Treatment in quiescent disease is challenging and the first line treatment, loperamide, is often used with caution due to the side effect of constipation.
Enterosorbents such as Enterosgel provide a potential alternative treatment. Enterosgel is thought to work through the adsorption of small molecules in the gastrointestinal (GI) tract, with minimal or no systemic absorption and metabolism², without causing constipation. Enterosgel has shown clinical effectiveness in reducing urgency in patients with diarrhoea predominant irritable bowel syndrome (IBS-D), suggesting potential therapeutic value in patients with UC³, however the mechanism by which this may occur is unknown.

Aims & Methods

This study aimed to evaluate the in vitro adsorption capacity of Enterosgel for three key inflammatory mediators implicated in UC pathogenesis (IL-1β, MMP9 and TNF-α) to explore its potential as a therapeutic option in this patient group.
Three experimental paradigms were performed by diluting recombinant cytokines and enzymes in water-based Enterosgel solutions. After incubation, gel-free supernatants were harvested to analyse residual levels of metabolite via ELISA to identify the quantity of adsorption by Enterosgel. Conditions were conducted with different:

1. Incubation times (15,30,60,120 min)
2. Metabolite concentrations (5 concentrations ranging from 0 - 3000pg/mL)
3. Gel concentrations (0.056, 0.1125, 0.225g/mL)

Results

Enterosgel demonstrated measurable concentration-dependent adsorption of IL-1β, MMP9 and TNF-α with higher analyte input concentrations resulting in greater absolute adsorption. However for TNF-α adsorption saturated at 1000pg/mL, indicating that Enterosgel may not fully adsorb analytes at levels simulating extreme luminal inflammation. Charcoal consistently resulted in lower residual concentrations, although this did not reach statistical significance on permutation testing (p = 0.14).
In time dependent analysis, Enterosgel induced a rapid initial reduction in cytokine levels within 15 minutes, (IL-1β: 98.7%, MMP9: 99.4%, TNF-α: 95.8% adsorbed) with a plateau by 120 minutes (IL-1β: r²=0.74, MMP9: r²=0.64, TNF-α: r²=0.45). In contrast, charcoal demonstrated a near-linear monotonic decline with a slightly greater concentration adsorbed by 120 minutes (IL-1β: 99.9% vs. 99.8%; MMP9: 99.9% vs. 99.1%; TNF-α: 97.2 vs. 96.3%). Increasing gel concentrations improved adsorption efficiency; however, at the highest gel concentration small amounts of residual cytokines remained (IL-1β: 0.25%, MMP9: 0.28%, TNF-α: 3.5%). Doubling of gel concentration led to an average 30% decrease in residual cytokines.

Conclusion

Our findings suggest that Enterosgel demonstrates excellent adsorption of three key pro-inflammatory cytokines relevant to UC, with the greatest reduction at 1 hour and the highest gel concentration—though less than that achieved by charcoal. Evidence of saturation at high analyte levels suggests a limit to Enterosgel’s binding capacity under conditions simulating severe inflammation. These results highlight Enterosgel's clinical potential, however further validation with biological replicates and clinical studies assessing urgency in treated UC is required.

References

(1) Theresa Hunter Gibble, April N. Naegeli, Michael Grabner, Keith Isenberg, Mingyang Shan, et al. Identification of inadequate responders to advanced therapy among commercially-insured adult patients with Crohn’s disease and ulcerative colitis in the United States. 2023; 63 (23): 10.1186/s12876-023-02675-w.
(2) Howell CA, Mikhalovsky SV, Markaryan EN, Khovanov AV. Investigation of the adsorption capacity of the enterosorbent Enterosgel for a range of bacterial toxins, bile acids and pharmaceutical drugs. Scientific Reports. 2019; 9 (1): 10.1038/s41598-019-42176-z.
(3) Howell CA, Kemppinen A, Allgar V, Dodd M, Knowles CH, Mclaughlin J, et al. Double-blinded randomised placebo controlled trial of enterosgel (polymethylsiloxane polyhydrate) for the treatment of IBS with diarrhoea (IBS-D). BMJ; 2022.
(4) Shane O'Sullivan, John F Gilmer, Carols Medina. Matrix Metalloproteinases in Inflammatory Bowel Disease: An Update. Mediators of Inflammation. 2015; 2015 (1): 10.1155/2015/964131.
(5) Qingdong Guan JZ. Recent Advances: The Imbalance of Cytokines in the Pathogenesis of Inflammatory Bowel Disease. Mediators of Inflammation. 2017; 2017 (1): 10.1155/2017/4810258.