This Week in Gastroenterology — Aug 18, 2026

Welcome to This Week in Gastroenterology. This week we're covering 10 notable papers spanning functional gut disorders and dietary therapy, celiac disease epidemiology and diagnostic accuracy, inflammatory bowel disease positioning and safety, and the fast-moving world of liver fibrosis assessment and treatment. Let's dive in.

We start with the functional gut, where Gastroenterology published a systematic review and meta-analysis from Staudacher and colleagues on fiber supplementation in irritable bowel syndrome — the first substantial synthesis in years [1]. Thirty randomised trials covering just over nineteen hundred patients were pooled. Across the sixteen trials reporting dichotomous clinical response, just over half OF PATIENTS on fiber responded versus about forty-four percent on placebo, a statistically significant but modest advantage of roughly a fifth in relative terms. The signal was driven almost entirely by psyllium, which increased the chance of response by about half. When the analysis shifted to integrative symptom scores, there was no significant overall effect — with one exception, beta-galacto-oligosaccharides, though that rests on only two trials. Wheat bran, the most commonly studied fiber with seven trials, does not emerge as the winner. The practical message is narrower than the phrase "try more fiber": if you are going to supplement, use psyllium, and be aware that subtype-specific data are largely absent, so we still cannot tell you reliably whether your constipation-predominant patient will respond differently from your diarrhoea-predominant one. That uncertainty sits alongside a major epidemiological paper in Gut, where Sperber and colleagues report the Rome V global epidemiology and validation survey — an internet-based population survey of nearly twenty-nine thousand adults across fifteen countries [3]. The headline is reassuring rather than surprising: about forty-one percent of adults met criteria for at least one disorder of gut-brain interaction under Rome V, essentially identical to the roughly forty percent under Rome IV, with the same pattern of higher prevalence in women and declining prevalence with age. Functional dyspepsia was the most common gastroduodenal disorder at about eight percent, chronic constipation and irritable bowel syndrome each hovered around nine percent, and proctalgia fugax was the leading anorectal disorder. Two new entities were quantified for the first time: abdominal migraine at about five percent, and inability to belch syndrome at just over one percent. The consistency across two independent criteria sets strengthens confidence that these prevalence figures are real, and it means the new Rome V labels are not going to inflate your clinic population.

Turning to celiac disease, two papers this week converge on the same uncomfortable theme — we are both under-detecting and, historically, over-diagnosing. In Gastroenterology, Karimzadhagh and colleagues pooled 87 studies covering nearly four hundred thousand people [2]. Global seroprevalence was about one and a half percent — roughly one in seventy — while biopsy-confirmed prevalence was about zero point eight percent, or one in a hundred and twenty. Prevalence was around fifty percent higher in females and nearly doubled in children. Two findings deserve attention in the clinic: using multiple serologic tests raised seropositivity by about forty percent, and in immunoglobulin A deficiency, adding deamidated gliadin peptide immunoglobulin G more than doubled detection. Meanwhile more than a third OF SEROPOSITIVE INDIVIDUALS never had a biopsy, and about forty-one percent of biopsy-confirmed cases were asymptomatic — which undercuts any purely symptom-driven testing strategy. The mirror image comes from Clinical Gastroenterology and Hepatology, where Hjort and colleagues used the Norwegian HUNT4 population study, screening more than fifty-two thousand adults and sequencing human leukocyte antigen DQ alleles [7]. Every single newly screen-detected case carried DQ2 or DQ8, as genetics predicts. But among those with a pre-existing established diagnosis, twenty-one patients carried neither — and on histological reassessment, the diagnosis held up in only five of those twenty-one. In other words, a meaningful slice of legacy celiac diagnoses appears to be wrong. There was a clear gene-dose effect, with DQ2.5 homozygotes carrying by far the highest risk. The actionable point: in a patient carrying a long-standing celiac label whose story does not quite fit, human leukocyte antigen typing remains a cheap and powerful way to exclude the diagnosis and potentially return them to a normal diet.

In inflammatory bowel disease, two papers address the questions patients actually ask — is this drug going to give me cancer, and does the order matter. On safety, Clinical Gastroenterology and Hepatology published a claims-based cohort from Ahuja and colleagues of nearly sixteen thousand patients starting an advanced therapy between 2016 and 2023, followed for a median of about two and a half years [4]. Two hundred and seventy-three cancers occurred. Three-year cumulative cancer incidence sat between two and three percent for tumour necrosis factor antagonists, vedolizumab, anti-interleukin agents, and Janus kinase inhibitors, and after propensity-score weighting there was no meaningful difference between classes. That consistency held across solid tumours, haematological malignancy, and melanoma. The caveat is that only 384 patients were on Janus kinase inhibitors, so the confidence around that class is genuinely wide, and median follow-up is short for oncological endpoints. Still, for routine counselling, this supports telling patients that class choice is not a cancer-risk decision. On sequencing, Alimentary Pharmacology and Therapeutics reports an ENEIDA registry analysis from Yagüe Caballero and colleagues [8]. Patients receiving anti-tumour-necrosis-factor therapy as second line after a biologic with a different mechanism of action were propensity-matched against first-line anti-tumour-necrosis-factor users and against those who had switched within the class. In Crohn's disease, where the prior agent was usually ustekinumab, durability was worse than first-line use — roughly a fifty percent higher discontinuation risk — but clinical effectiveness was not significantly different. In ulcerative colitis, where the prior agent was almost always vedolizumab, the picture was worse: durability was inferior both to first-line use and to anti-tumour-necrosis-factor after anti-tumour-necrosis-factor, and remission rates were significantly lower at both short and long-term follow-up. If you are planning to use an anti-tumour-necrosis-factor agent in ulcerative colitis, its best performance is early. And for a wider lens, The Lancet Gastroenterology and Hepatology traces the hundred-year trajectory of inflammatory bowel disease across Japan, South Korea, China, Hong Kong, and Malaysia [6]. All five regions sit in the acceleration-in-incidence stage, and east and southeast Asia is poised to become the first non-early-industrialised region to enter compounding prevalence — a warning about health-system capacity that will land within our careers.

Finally, the liver. Alimentary Pharmacology and Therapeutics carries a review from Ayoub, Tsutsumi and Rinella on fibrosis reversal in metabolic dysfunction-associated steatohepatitis [5]. The pattern they draw out is that drugs targeting metabolic dysfunction have outperformed those aimed directly at fibrogenic or inflammatory pathways: resmetirom achieved at least one-stage fibrosis improvement in up to about a quarter of patients versus fourteen percent on placebo at 52 weeks, and semaglutide in about thirty-seven percent versus twenty-two percent at 72 weeks. Fibroblast growth factor 21 analogues, dual and triple incretin agonists, and pan-PPAR agonists have encouraging phase 2 data. The honest caveat they emphasise is that the link between treatment-induced histological improvement and reduction in hard clinical outcomes remains unproven, and compensated cirrhosis needs a different framework. Complementing that, Hepatology published a study from Lee and colleagues of more than twenty-eight hundred patients with metabolic dysfunction-associated steatotic liver disease who underwent both shear wave elastography and vibration-controlled transient elastography in the same session [10]. Slotted into the American Gastroenterological Association and European Association for the Study of the Liver two-step algorithms, shear wave elastography stratified liver-related event risk just as well as vibration-controlled transient elastography, with discrimination around zero point seventy-seven to zero point eighty for both and no significant reclassification gain either way. The authors are careful to say this does not establish formal equivalence, but practically, for centres with ultrasound-based elastography rather than a dedicated device, existing FIB-4-based pathways appear usable.

If you only have time for one paper this week, make it the celiac human leukocyte antigen study from the HUNT cohort [7]. The finding that most patients carrying a legacy celiac diagnosis without a permissive DQ genotype turn out not to have the disease is immediately actionable in clinic, and it costs one blood test to act on.

Here are the key takeaways from this week in Gastroenterology. For irritable bowel syndrome, psyllium is the fiber with evidence behind it — wheat bran is not, and overall symptom scores show no clear benefit. About one in seventy people worldwide is celiac-seropositive and roughly two in five confirmed cases are asymptomatic, so symptom-triggered testing alone will miss a great deal; conversely, human leukocyte antigen typing can dismantle an incorrect long-standing diagnosis. Cancer risk appears low and comparable across all advanced therapy classes in inflammatory bowel disease, which should simplify counselling. Anti-tumour-necrosis-factor therapy loses effectiveness and durability when used after a different mechanism of action, most clearly in ulcerative colitis. And in fatty liver disease, shear wave elastography can substitute for vibration-controlled transient elastography within existing two-step algorithms, while fibrosis regression with resmetirom and semaglutide is real but not yet linked to hard outcomes.

That's your roundup for This Week in Gastroenterology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.