Relationship between Delta, Sigma, Beta, and Gamma EEG bands at REM sleep onset and REM sleep end

The aim of the present study was to analyze in detail the relationship of two newly introduced measures, related to the Beta and Gamma EEG bands during REM sleep, with Delta and Sigma activity at REM sleep onset and REM sleep end, in order to understand their eventual role in the sleep modulation mechanism.

For this purpose, power spectra of 1 EEG channel (C4, referred to A1) were obtained by means of the fast Fourier transform and the power of the bands ranging 0.75-4.50 Hz (Delta), 4.75-7.75 (Theta), 8.00-12.25 (Alpha), 12.50-15.00 (Sigma), 15.25-24.75 (Beta), 25.00-34.75 (Gamma 1), and 35.00-44.75 (Gamma 2) was calculated for the whole period of analysis (7 h), in 10 healthy subjects. Additionally, two other time series were calculated: the ratio between Beta and Gamma2, and between Gamma1 and Gamma2 (Beta and Gamma ratios). For each subject, we extracted 3 epochs of 30 min corresponding to the 15 min preceding and the 15 min following the onset of the first 3 REM episodes. Data were then averaged in order to obtain group mean values and standard deviation. The same process was applied to the 30-min epochs around REM sleep end.

The course of the Delta band around REM sleep onset was found to be characterized by a first phase of slow decline lasting from the beginning of our window up to a few seconds before REM onset; this phase was followed by a sudden, short decrease centered around REM onset, lasting for approximately 1.5-2 min. At the end of this phase, the Delta band reached its lowest values and remained stable up to the end of the time window. The Sigma band showed a similar course with stable values before and after REM sleep onset. The Beta and Gamma ratios also showed a 3-phase course; the first phase, in this case, was characterized by stable low values, from the beginning of our window up to approximately 5 min before REM onset. The following second phase was characterized by an increase which reached its maximum shortly after REM sleep onset (approximately 1 min). In the last phase, both Beta and Gamma ratios showed stable high values, up to the end of our time window. At REM sleep end, the Delta band only showed a very small gradual increase, the Sigma band presented a more evident gradual increase; on the contrary, both Beta and Gamma ratios showed a small gradual decrease.

The results of the present study show a different time synchronization of the changes in the Delta band and in Beta and Gamma ratios, at around REM sleep onset, and seem to suggest that the oscillations of these parameters might be modulated by mechanisms more complex than a simple reciprocity. All these considerations point to the fact that REM sleep can be considered as a complex phenomenon and the analysis of high-frequency EEG bands and of our Beta and Gamma ratios represent an additional important element to include in the study of this sleep stage.