EEG power and glucose fluctuations are coupled during sleep in young adults with type 1 diabetes

To determine the coupling between brain activity and glucose variations during sleep in young adults with type 1 diabetes mellitus (T1DM).

27 participants, age 18-30, wore a continuous glucose monitoring system (CGMS) and underwent in-laboratory overnight polysomnography (PSG). Quantitative electroencephalogram (qEEG) metrics were determined from the PSG and included Delta, Theta, Alpha, Sigma, Beta and Gamma Band power at 5-min intervals. Wavelet Coherence Analysis was employed to determine the time varying and frequency specific coupling between glucose and EEG Band power. ANOVA was used to compare differences across fluctuation speeds and EEG bands.

There was a high degree of time varying and frequency specific coupling between glucose variations and EEG power in all EEG Bands during sleep. The average number of intervals of statistically significant coherence was highest for fluctuations periods between 10 and 30min in all Bands (p<0.0001 for each). Mean significant coherence was negatively correlated with hemoglobin A1c, a marker of glycemic control.

The relationship between glucose and EEG power during sleep is time varying and frequency dependent in young adults with T1DM.

Understanding the time varying mutual relationship between glucose changes and brain activity during sleep may have implications for disease management in T1DM.