Wheat challenge in self-reported gluten sensitivity: a comparison of scoring methods

Abstract

Background: The condition non-coeliac gluten sensitivity (NCGS) is clinically similar to coeliac disease, but lack objective diagnostic criteria. Symptom relief on gluten-free diet followed by gluten containing food challenge may confirm the condition in clinical settings.

Aim: To describe the results of an open bread challenge in patients with suspected NCGS, and to compare the results with recently suggested cut-offs for symptom change.

Material and methods: Fifty-six patients (12 males) self-instituted on gluten-free diet with negative coeliac disease diagnostics were examined for NCGS by an open bread challenge. Symptoms were reported by Gastrointestinal Symptom Rating Scale, IBS-version (GSRS-IBS) and visual analogue scale (VAS). Results were retrospectively compared to the Salerno and Monash cut-offs for symptom change.

Results: Forty-seven patients were diagnosed with NCGS. Total GSRS-IBS score and overall symptoms by VAS increased significantly in NCGS (p < .001), but not in non-NCGS patients (p < .12 and p = .08, respectively). Total GSRS-IBS challenge score and overall symptoms by VAS were significantly higher in NCGS than in non-NCGS patients (53 vs. 37, p = .004 and 76 vs. 39 mm, p = .02, respectively). Applying the Salerno and Monash cut-offs, 63 and 75% would be classified with NCGS, respectively. According to total GSRS–IBS absolute agreement was lowest between clinician’s diagnosis and Salerno cut-off (63%) and highest between Salerno and Monash cut-offs (88%).

Conclusion: Clinician diagnosed 85% with NCGS. The proportion of NCGS was lower according to the Salerno and Monash cut-offs. The Salerno cut-off should be the starting point for a common definition of symptom change.

Keywords:

• Non-coeliac gluten sensitivity
• gluten challenge
• diagnostics
• symptom score
• Salerno Experts’ Criteria

Acknowledgments

Thanks to all the patients who gave written consent to use information from their patient record.

Disclosure statement

The authors declare no conflicts of interest.