The prevalence of celiac disease (CeD) in children with type 1 diabetes mellitus (T1DM) is significant, and celiac serology screening is routinely done in T1DM patients. The sequence of CeD and T1DM diagnosis and the natural history of transient celiac serology in T1DM is poorly defined. We aimed to investigate the temporal relationship between T1DM and CeD diagnoses, and the long-term natural history of celiac serology and CeD diagnosis in children with T1DM. A retrospective cohort study of 1,799 children with T1DM, treated at a single tertiary center (2009-2024), was evaluated for clinical, laboratory, and histologic data. Patients with positive anti-tissue-transglutaminase IgA antibodies (anti-TTG) were stratified based on CeD diagnosis or long-term serological behavior on a gluten-containing diet. Among 136 patients with T1DM and positive anti-TTG, 113 were diagnosed with CeD, 15 had persistent seropositivity (without confirmed CeD), and eight had a single borderline result (< & times;3 upper limit of normal). CeD was diagnosed after T1DM in 81.4% of cases, and prior to TIDM in 18.6%. Patients diagnosed with CeD prior to T1DM exhibited more gastrointestinal symptoms, higher EMA positivity, and poorer glycemic control (p < 0.05). HbA1c values at follow-up were significantly higher in patients with CeD, compared to those with a single borderline anti-TTG, p < 0.01. In conclusion, most, but not all, patients with T1DM and elevated celiac serology, eventually develop biopsy-confirmed CeD. CeD can precede T1DM in a significant number of cases, emphasizing the need for continued monitoring for celiac serology in children diagnosed with T1DM and monitoring for developing T1DM in patients with CeD.
Parents of children with type 1 diabetes experience substantial stress, anxiety, and sleep difficulties. We evaluated the effectiveness of Body Emotional Maps (BEM), an experiential group intervention delivered during a diabetes camp, compared with a traditional educational camp. In this multicenter randomized trial (NCT07098689), 139 parents were enrolled. The primary endpoint was the between-group difference in Parenting Stress Index-Short Form (PSI-SF) change at three months (T3). Secondary outcomes included depressive symptoms (Patient Health Questionnaire, PHQ-9), anxiety (Generalized Anxiety Disorder questionnaire, GAD-7), subjective sleep quality (Brief Pittsburgh Sleep Quality Index, B-PSQI), parental self-efficacy (Parental Self-Efficacy for the Management of Diabetes Mellitus, [PSESDM]), coping (Coping Orientation to Problems Experienced, COPE), and children's glucometric indices at T3 and six months (T6). Mixed models for repeated measures were used. At T3: PSI-SF total score improved significantly in the BEM group (p = 0.03) but not in controls (p = 0.08). At T6, both groups showed significant improvement (p < 0.001) with no between-group difference. The parental distress domain improved earlier and more consistently in the BEM group. Anxiety decreased significantly in both groups. Subjective sleep quality improved only in the BEM arm at both assessments, with a significant advantage in sleep efficiency at T6 (p = 0.03). Children's glycemic parameters remained stable in both groups. In conclusion, both camp models reduced parental stress, but BEM produced earlier improvements and additional benefits in parental distress and sleep quality.
Open-source automated insulin delivery shows promise for managing type 1 diabetes, though pediatric efficacy and safety data remain limited. This study aimed to compare real-world glycemic outcomes of open-source automated insulin delivery systems with hybrid closed-loop systems in youth with type 1 diabetes. This retrospective cohort study included individuals aged <= 19 years with type 1 diabetes for at least one year who used either open-source automated insulin delivery systems (Loop, iAPS, or AndroidAPS) or hybrid closed-loop systems (Tandem t:slim) for at least six months between January 2015 and June 2024. Data were collected from one month before system initiation to 18 months after initiation or the last clinical visit. Glycated hemoglobin, time in range, and glucose management indicator were evaluated using regression analyses and adjusted linear mixed-effects models. A total of 198 participants (143 open-source automated insulin delivery and 55 hybrid closed-loop users) were included (mean age 13.8 years; 57% male). Over 18 months, glycated hemoglobin decreased from 7.40% to 6.96% in the open-source automated insulin delivery group and from 7.35% to 7.28% in the hybrid closed-loop group. Adjusted analyses showed no significant differences in glucose management indicator or time in range between groups across follow-up. Severe hypoglycemia occurred in one participant per group, and no diabetic ketoacidosis events were observed. Open-source automated insulin delivery systems demonstrated similar long-term glycemic outcomes and safety compared with hybrid closed-loop systems, supporting their use as a safe, non-inferior alternative.
