Gastroenterology

AI-assisted EGD substantially increases detection of upper GI neoplasia, especially subcentimeter lesions

This meta-analysis of 11 RCTs (57,512 adults) compared AI-assisted EGD with conventional EGD for upper GI neoplasm detection. AI-EGD increased per-patient neoplasm detection (RR 1.57; 95% CI 1.23–2.01) and per-lesion detection (RR 1.55; 95% CI 1.33–2.18). The effect was greatest for lesions ≤10 mm (RR 2.24; 95% CI 1.72–2.91), with no clear advantage for larger lesions. Detection improved across LGIN, HGIN, and carcinoma, with particularly higher yield for LGIN. 1

Six‑month ileal stricture status under biologics predicts Crohn’s hospitalization and surgery

This post hoc analysis followed 170 patients with ileal or ileocolonic Crohn’s starting or switching biologics (anti‑TNF, ustekinumab, vedolizumab). At six months, strictures resolved in 30.9% and progressed in 15.7%, with no significant differences between biologic classes. Baseline strictures were not prognostic, but strictures present at six months increased risks of hospitalization and surgery, especially in non-terminal ileum. Clinical, biochemical, and endoscopic indices improved overall, indicating structural response may lag inflammatory control. 2

Advanced therapies show typical IBD efficacy but notable infections in PSC-associated IBD

This systematic review and meta-analysis included 19 studies with 864 patients with PSC-associated IBD treated with biologics or small molecules. Clinical response was 52% and clinical remission 43%, while endoscopic remission reached 26%, comparable to non-PSC IBD experience. Endoscopic remission was higher with anti-TNF agents (29%) than with non–anti-TNF biologics (15%), though heterogeneity was substantial. Infections occurred in 22% and acute cholangitis in 14%, underscoring the high complication burden. 3

International consensus standardizes terminology and imaging criteria for fatty pancreas

This multi-society consensus statement recommends "fatty pancreas" as the unified term for all pancreatic fat accumulation. It provides imaging-based diagnostic criteria using radiologic and endoscopic modalities to standardize detection and reporting. The report synthesizes data linking fatty pancreas with alcohol use, smoking, acute and chronic pancreatitis, exocrine insufficiency, diabetes, and surgical outcomes. Potential associations with IPMN, pancreatic cancer, and metabolic dysfunction are summarized but remain incompletely defined. 4