Abstract / Summary
Purpose: We explored the associations between vitamin C intake and the risk of islet autoimmunity (IA) and/or type 1 diabetes in genetically at-risk children. Furthermore, we explored associations between vitamin C metabolism-related single nucleotide polymorphisms (SNPs) and type 1 diabetes outcomes.
Methods: The current study within Environmental Determinants of Diabetes in the Young (TEDDY) cohort included 8478 children followed up every 3-6 months for relevant autoantibodies and diet. Dietary vitamin C intake was assessed longitudinally throughout childhood using age-specific dietary assessment methods and analysed using repeated measurements. Cox regression was used for the primary analyses and Bayesian joint longitudinal-survival models as sensitivity analyses.
Results: A total of 777 (9.2%) children developed IA, including 292 (3.4%) with IAA-first and 337 (4.0%) with GADA-first autoimmunity; 319 (41.0%) progressed to type 1 diabetes. Mean (SD) vitamin C intake was 60.7 (38.8) mg/1000 kcal. Higher vitamin C intake was associated with an increased risk of IAA-first autoimmunity (adjusted hazard ratio 1.04; 95% confidence interval 1.00-1.07, per 10 mg/1000 kcal increase) while lowest and highest tertiles of intake were associated with increased risk of GADA-first autoimmunity as compared to mid-tertile [(1.65; 1.20-2.27) and (1.42; 1.02-1.98), respectively]. Sensitivity analyses using Bayesian joint longitudinal-survival models supported the primary findings and suggested nonlinear associations between vitamin C intake and the risks of IA and multiple IA. Associations between vitamin C-related SNPs and study outcomes did not withstand correction for multiple testing.
Conclusion: Vitamin C intake was not consistently associated with IA or progression to type 1 diabetes. However, the observed nonlinear association, with lowest risk for moderate intakes, merits further investigation.
Trial registration: NCT00279318, 06/09/2004.