Missed meal boluses and poorer glycemic control impact on neurocognitive function may be associated with white matter integrity in adolescents with type 1 diabetes

Abstract

Background: The notion that pediatric type 1 diabetes impacts brain function and structure early in life is of great concern. Neurological manifestations, including neurocognitive and behavioral symptoms, may be present from childhood, initially mild and undetectable in daily life. Despite intensive management and technological therapeutic interventions, most pediatric patients do not achieve glycemic control targets for HbA1c. One of the most common causes of such poor control and frequent transient hyperglycemic episodes may be lifestyle factors, including missed meal boluses.

Objective: The aim of this study was to assess the association between specific neurocognitive accomplishments-learning and memory, inhibition ability learning, and verbal and semantic memory-during meals with and without bolusing, correlated to diffusion tensor imaging measurements of major related tracts, and glycemic control in adolescents with type 1 diabetes compared with their healthy siblings of similar age.

Study design and methods: This is a case-control study of 12- to 18-year-old patients with type 1 diabetes (N = 17, 8 male patients, diabetes duration of 6.53 ± 4.1 years) and their healthy siblings (N = 13). All were hospitalized for 30 h for continuous glucose monitoring and repeated neurocognitive tests as a function of a missed or appropriate pre-meal bolus. This situation was mimicked by controlled, patient blinded manipulation of lunch pre-meal bolus administration to enable capillary glucose level of <180 mg/dl and to >240 mg/d 2 hours after similar meals, at a similar time. The diabetes team randomly and blindly manipulated post-lunch glucose levels by subcutaneous injection of either rapid-acting insulin or 0.9% NaCl solution before lunch. A specific neurocognitive test battery was performed twice, after each manipulation, and its results were compared, along with additional neurocognitive tasks administered during hospitalization without insulin manipulation. Participants underwent brain imaging, including diffusion tensor imaging and tractography.

Results: A significant association was demonstrated between glycemic control and performance in the domains of executive functions, inhibition ability, learning and verbal memory, and semantic memory. Inhibition ability was specifically related to food management. Poorer glycemic control (>8.3%) was associated with a slower reaction time.

Conclusion: These findings highlight the potential impairment of brain networks responsible for learning, memory, and controlled reactivity to food in adolescents with type 1 diabetes whose glycemic control is poor.

Keywords: adolescents; brain domains; cognitive performance; diffusion tension imaging; glycemic control; type 1 diabetes.

Figure 1

Day–night emotional Stroop stimuli.

Figure 2

Visual Update Stimulus example.

Figure 3

Object Recall examples.

Figure 4

Example of a healthy participant’s FA profile of the right SLF.

Figure 5

New Object Recall significant differences.

Figure 6

Voxel-based group comparison revealed a higher grey matter (GM) index in the type 1 diabetes (T1D) group. This included brain regions connected by white-matter (WM) fiber tracts, such as the SLF and corona radiata. A higher GM probability index was found in the frontal inferior triangular and operculum regions, known as Wernicke, and in parietal regions, such as the inferior parietal and postcentral gyrus.

Figure 7

Analysis of variance by group in the superior frontal FA and RD corpus callosum segment center.

Figure 8

FA group comparison plots. Control group and diabetes mean FA ± 1 SD is plotted along 100 nodes on tract center. The upper graph shows significantly higher FA values in the diabetes group in 32 locations of the posterior parietal callosal segment. The panel beneath it shows significantly lower FA values in the diabetes group in 12 locations of the superior frontal callosal segment.

Figure 9

Right SLF white matter. Left: axial view of one diabetes patient (c016), right SLF pathway as traced by AFQ–Center: diagram showing the lateral surface of the left cerebral hemisphere. SLF is at the center, in red. Right: dissection of the human brain from the lateral aspect by the Klingler technique. Resecting the superficial digitations exposes the superior longitudinal fasciculus (SLF) superior to the sylvian fissure and insula (In).