1173-P: CGM Detection of Undiagnosed Celiac Disease in Children with Type 1 Diabetes

Celiac disease (CD) results in enteropathy and malabsorption that can cause prandial insulin-glucose mismatch in patients with type 1 diabetes (T1D). This study leveraged continuous glucose monitoring (CGM) in children during the first year after T1D diagnosis to determine whether those with untreated CD have lower post-prandial glucose levels compared to those without CD. Children with T1D, positive CD serologies and biopsy-confirmed CD (CD+) were matched 1-to-4 to children with T1D but negative CD serologies (CD-) on age, sex, BMI and duration of diabetes. Meals were analyzed in the 30-day window before CD serologies and were computationally defined as paired CGM glucose peaks and troughs. Differences in the primary outcome, post-prandial trough, and other post-prandial glucose variables were assessed via log-linear mixed models. The cohort (N = 16 CD+, N = 64 CD-) was 69% female, with median age 10.7 years (interquartile range 8.9 - 12.8), T1D duration 4.1 months (2.3 - 7.7), and BMI z-score 0.29 (0.00 - 0.92). Over the 30-day window, mean post-prandial trough glucose and peak glucose were 111.7 mg/dL (SD 36.3) and 213.6 mg/dL (48.5) in the CD+ children, and 128.6 mg/dL (39.8) and 238.4 mg/dL (53.0) in the CD- children, respectively. In the repeated measures models, CD diagnosis reduced trough glucose by 12.8% (p = 0.059) and reduced peak glucose by 10.2% (p = 0.048). Other secondary outcomes including change in glucose or time from peak to trough did not significantly differ between groups. There is a quantifiable difference in post-prandial glucose when CD is present and untreated in patients with T1D. These findings can inform a future, CGM-based, predictive model of CD in T1D to prompt screening and aid in earlier diagnosis and treatment.