Yield of Gluten Challenge in Patients on Self-Prescribed Gluten-Free Diets

Patients on a gluten-free diet (GFD) whose celiac disease (CD) status is unknown may undergo gluten challenge (GC) to clarify their diagnosis. Though this is an established diagnostic practice, the proportion of patients undergoing GC who are diagnosed with CD is unknown.

We aimed to analyze which factors were predictive of having CD in a cohort of patients who underwent GC followed by upper endoscopy with duodenal biopsy.

We identified adult patients at a CD referral center who had been on a GFD and then underwent GC to determine a diagnosis of CD during the years spanning 2006 to 2020. We compared those patients found to have CD (defined as villus atrophy/Marsh 3) on duodenal biopsy with those who did not, using the chi square and Fischer exact tests.

We identified 206 patients who underwent GC. Of these 206, 30 (14%) were diagnosed with CD based on post-GC duodenal biopsy. 176 of the 206 (85%) patients reported various gastrointestinal symptoms, including bloating (39%), though these were more common in those without CD (any GI symptoms: 89% vs 67%, p 0.004; bloating: 43% vs 20%, p 0.019). Serology values, when normalized, including pre- and post-challenge TTG IgA (37% vs 1.7%, p 0.001; 23% versus 2.3%, p 0.001), DGP IgG and IgA (57% vs 2.8%, p 0.001; 37% vs 6.2%, p 0.001) were higher in the group of patients with CD.

Among patients undergoing GC for diagnostic purposes, only 14% had evidence of villus atrophy corresponding with CD on duodenal biopsy. The presence of any elevated pre-challenge serology was associated with CD. Bloating in combination with low serologies may help risk stratify patients as being less likely to have CD upon GC.