TOPLINE

A crossover clinical trial found that every patient with irritable bowel syndrome (IBS) showed visible mucosal reactions either before or after exposure to food on confocal laser endomicroscopy (CLE), but a targeted diet based on those reactions produced no more improvement than an unrelated sham diet. Notably, healthy volunteers with no IBS symptoms showed identical reactions when tested, suggesting that the technique may not detect a disease-specific abnormality.

METHODOLOGY

• CLE has been used to visualize acute food-induced mucosal reactions in the duodenum of patients with IBS, with uncontrolled studies suggesting an improvement in symptoms after the exclusion of these 'trigger foods' from the diet.
• Researchers conducted a clinical trial to examine whether a CLE-guided elimination diet could benefit patients with predominantly nonconstipated IBS and a baseline IBS-Symptom Severity Score (IBS-SSS) of 175 or higher.
• A total of 33 patients with IBS (median age, 33 years; 85% women) and 15 age- and sex-matched healthy volunteers underwent CLE, during which foods were introduced one at a time in a random order onto the duodenal mucosa. A blinded reviewer classified each exam as reactive or nonreactive on the basis of fluorescein leakage and cell shedding at multiple sites; testing was stopped once a reaction was observed, yielding 54 exams in the IBS group.
• After CLE, patients entered a randomized double-blind crossover dietary phase: 4 weeks of avoiding up to 2 foods flagged as reactive on CLE and 4 weeks of avoiding up to 2 nonreactive foods for a sham comparison, with a 2-week gap between phases.
• The primary endpoint was the proportion of clinical responders — defined as at least a 50-point reduction in the IBS-SSS by the end of a 4-week diet phase.

TAKEAWAY

• Altered mucosa was visualized in 100% of examinations in patients with IBS — either at baseline (11%) or after food administration — with milk, fish, and egg white most commonly triggering reactions.
• Roughly 4 in 10 patients improved on the targeted diet vs about a third on the sham diet (41.9% vs 35.5%), a gap that was not statistically meaningful (P = .6); overall symptom relief was likewise similar across both diets (P = .7).
• Healthy volunteers reacted just as often as patients did — 100% showed alterations, either right away (13.3%) or after food exposure.
• After acute mucosal reactions, gut permeability, systemic and local tryptase release, and eosinophil-derived neurotoxin release all remained unchanged compared with baseline.

IN PRACTICE

"In conclusion, we observed high rates of mucosal alterations in the duodenum of IBS patients without any evidence of local mast-cell or eosinophil activation nor altered epithelial permeability. Observed alterations on food administration were not instructive of exclusion diets with poor sensitivity and specificity. Finally, we found the same pattern of alterations in healthy, asymptomatic individuals, suggesting that acute alterations, as visualized using CLE, carry no pathophysiological relevance," the authors wrote.

SOURCE

The study was led by Lukas Michaja Balsiger, MD, PhD, Translational Research in Gastrointestinal Disorders, KU Leuven, Leuven, Belgium. It was published online in Gastroenterology.

LIMITATIONS

The number of foods excluded varied from patient to patient, and not every patient was tested against every food during endomicroscopy. Food was delivered straight to the duodenum, skipping gastric acid digestion, which may have changed how it interacted with the gut lining. The trial was also stopped early once an interim analysis showed the diet was unlikely to outperform sham, leaving a smaller sample than originally planned.

DISCLOSURES

The study was internally funded by KU Leuven through a Methusalem grant. One author received funding from the Swiss National Science Foundation and the Novartis Foundation. Some other authors reported grants from the Flanders Research Foundation.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.