CGM and HbA 1c data were obtained quarterly in the diabetes group, and cognitive and brain MRI were repeated after 18 months in all. All subjects had clinical and glycemic assessments, structural brain MRI, and neurocognitive testing performed at baseline. Siblings of children with diabetes were included if they had normal fasting glucose, HbA 1c, and negative diabetes autoantibodies. A group of 72 healthy, age-matched control subjects without diabetes and with similar inclusion/exclusion criteria was recruited. Participants were not excluded from the study if they developed any other conditions after enrollment. At enrollment, participants were on stable insulin therapy and had no history of prematurity, other chronic illnesses, seizures, developmental delay, or psychiatric diagnosis. Children with type 1 diabetes ( n = 144) on insulin therapy for at least 1 month, ages 4–9 years at study onset, were recruited. Informed written consent from the parents/guardians and child assent were obtained as appropriate. Louis (IRB #201411074), and Yale University (IRB #1411014935), and by the study’s data safety monitoring board. Studies were approved by the institutional review boards (IRBs) of each of the five participating DirecNet centers, including Nemours Children’s Health System Jacksonville (IRB #588973), Stanford University (IRB #32179), University of Iowa (IRB #201501830), Washington University in St.

We hypothesized that observed changes in the specific brain regions noted above would persist or worsen over time in children with type 1 diabetes, eventually negatively affecting cognition. We designed these new studies to determine whether abnormalities in total and regional gray and white matter volumes and white matter microstructure persist or worsen over longitudinal follow-up and whether these changes are associated with measures of glycemic control and neurocognitive metrics as children with type 1 diabetes grow and progress into puberty. These results highlight the importance of longitudinal studies of brain growth in early-onset type 1 diabetes.

Vulnerable brain areas in type 1 diabetes appear particularly localized to frontoparietal networks as well as components of networks involved in the processing of complex sensory information, sensorimotor function, and cognition. Remarkably, the biggest association of these reported differences has been with measures of hyperglycemia, as reflected by HbA 1c and percent sensor glucose (% sensor glucose) above target as detected by CGM ( 7– 16).

Moreover, in children with diabetes, there was slower growth of the hippocampus, which was associated with hyperglycemia and greater glycemic variability ( 15) differences in trajectories of brain growth in those with early severe diabetic ketoacidosis (DKA) were also observed ( 16). Subtle differences in specific cognitive domains compared with age-matched control subjects without diabetes were also detected ( 13, 14). Results from the first two time points over 18 months showed significant cross-sectional and longitudinal anatomical differences between our diabetes and control cohorts as demonstrated by volumetric and voxel-based morphometric methods and by diffusion tensor imaging ( 7– 12). Unsedated structural brain MRIs as well as age-appropriate cognitive testing were performed the diabetes group also had continuous glucose monitoring (CGM). To address this question, our Diabetes Research in Children Network (DirecNet), a five-center National Institutes of Health–funded consortium, conducted structural and diffusion-weighted imaging studies in very young children with type 1 diabetes (aged 4 to 8 years ago. Whether dysglycemia in young children irreversibly compromises key neurodevelopmental processes is unknown. The human brain undergoes unique dynamic structural and functional changes during childhood and requires continuous delivery of glucose for brain function and growth ( 6).

Michael a mazaika