Tracking brain states under general anesthesia by using global coherence analysis

Abstract

Time and frequency domain analyses of scalp EEG recordings are widely used to track changes in brain states under general anesthesia. Although these analyses have suggested that different spatial patterns are associated with changes in the state of general anesthesia, the extent to which these patterns are spatially coordinated has not been systematically characterized. Global coherence, the ratio of the largest eigenvalue to the sum of the eigenvalues of the cross-spectral matrix at a given frequency and time, has been used to analyze the spatiotemporal dynamics of multivariate time-series. Using 64-lead EEG recorded from human subjects receiving computer-controlled infusions of the anesthetic propofol, we used surface Laplacian referencing combined with spectral and global coherence analyses to track the spatiotemporal dynamics of the brain's anesthetic state. During unconsciousness the spectrograms in the frontal leads showed increasing α (8-12 Hz) and δ power (0-4 Hz) and in the occipital leads δ power greater than α power. The global coherence detected strong coordinated α activity in the occipital leads in the awake state that shifted to the frontal leads during unconsciousness. It revealed a lack of coordinated δ activity during both the awake and unconscious states. Although strong frontal power during general anesthesia-induced unconsciousness--termed anteriorization--is well known, its possible association with strong α range global coherence suggests highly coordinated spatial activity. Our findings suggest that combined spectral and global coherence analyses may offer a new approach to tracking brain states under general anesthesia.

Fig. 1.

Overview of the experiment. (A) Target effect-site concentrations of the anesthetic propofol delivered to each subject. (B) EEG electrode sites viewed from the top of the head. (C) Timeline of the auditory tasks, presented at 0, 4, 8, 12, and 16 s within each trial (Materials and Methods). Forty-two trials were presented per level.

Fig. 2.

Spectrograms of the radial current density estimated at the electrode sites in Fig. 1B. The five different propofol levels are demarcated by the black vertical lines in each spectrogram. x axis, time (0–∼84 min); y axis, frequency (0–30 Hz).

Fig. 3.

(A–C) Behavioral curves computed as the number of correct responses from a possible 5 on each trial (brown curve, scale on the right), the average delta (0- to 4-Hz range) power (blue curve) and the average alpha (8- to 12-Hz range) power (red curve) per trial computed from the radial current density estimated at a single occipital site. Behavioral and power curves were smoothed with a median filter.

Fig. 4.

(A–C) Time course of the global coherence computed at each frequency using the electrode sites in Fig. 1B. Global coherence values (color coded, scale on the right) close to 1 (red) suggest highly coordinated activity among the electrode sites whereas global coherence values close to 0 (blue) suggest an absence of coordinated activity. Global coherence values >0.6 (yellow–red) observed in the alpha range during the conscious state (levels 0 and 1) and during the unconsciousness (levels 4 and 5). The behavioral curve (brown) is defined in Fig. 3.

Fig. 5.

Snapshots of the row weights for each electrode site (Fig. 1B) during high global coherence at 11 Hz. The row weights were high over the occipital sites when the subjects were awake (A, C, and E), level 1, and high over the frontal sites when the subjects were unconscious (B, D, and F), level 5. Color scale (0–0.4) for the row weights are on the right.

Fig. 6.

(A–C) Time course of the cumulative sum of the row weights at each trial showing the contribution of each electrode site (Fig. 1B) to the leading eigenvector at 11 Hz. The row weights sum to 1 by definition. The color scheme identifies the spatial location of the electrode sites (Inset) with the occipital sites being colored pink-purple and the frontal sites being colored yellow-green. The color scheme has left–right symmetry. The behavioral curve (brown) is defined in Fig. 3.