Gluten intake and metabolic health: conflicting findings from the UK Biobank

Abstract

Purpose: The impact of gluten intake on metabolic health in subjects without celiac disease is unclear. The present study aimed to assess the association between gluten intake and body fat percentage (primary objective), as well as a broad set of metabolic health markers. Methods: Gluten intake was estimated in 39,927 participants of the UK Biobank who completed a dietary questionnaire for assessment of previous 24-h dietary intakes. Multiple linear regression analyses were performed between gluten intake and markers of metabolic health with Holm adjustment for multiple comparisons. Results: Median gluten intake was 9.7 g/day (male: 11.7 g/day; female: 8.2 g/day; p < 0.0001). In multiple linear regression analysis, association between gluten intake and percentage body fat was negative in males (β = − 0.028, p = 0.0020) and positive in females (β = 0.025, p = 0.0028). Furthermore, gluten intake was a negative predictor of total cholesterol (male: β = − 0.031, p = 0.0154; female: β = − 0.050, p < 0.0001), high-density lipoprotein cholesterol (male: β = − 0.052, p < 0.0001; female: β = − 0.068, p < 0.0001), and glomerular filtration rate (sexes combined: β = − 0.031, p < 0.0001) in both sexes. In females only, gluten intake was positively associated with waist circumference (β = 0.041, p < 0.0001), waist-to-height ratio (β = 0.040, p < 0.0001), as well as body mass index (β = 0.043, p < 0.0001), and negatively related to low-density lipoprotein cholesterol (β = − 0.035, p = 0.0011). A positive association between gluten intake and triglycerides was observed in males only (β = 0.043, p = 0.0001). Conclusion: This study indicates that gluten intake is associated with markers of metabolic health. However, all associations are weak and not clinically meaningful. Limiting gluten intake is unlikely to provide metabolic health benefits for a population in total.