Frequency-specific microstate correlates of ciprofol-induced alterations of consciousness

Ciprofol is a novel GABAergic intravenous anesthetic with a rapid onset and favorable safety profile; however, the neural mechanisms underlying the induction and reversal of unconsciousness remain unclear. Electroencephalogram (EEG) microstates provide millisecond-scale markers of large-scale brain network dynamics and may reveal frequency-specific signatures of ciprofol' s effects on consciousness.

Sixty-channel EEG was recorded during wakefulness, ciprofol-induced loss of consciousness (LOC), and recovery of consciousness (ROC). Power spectra were computed across the delta, theta, alpha, beta, and gamma bands. Microstate analysis was performed separately for each band, yielding seven microstates (A-G). Duration, coverage, and occurrence were compared across conscious states using one-way analysis of variance (ANOVA) with Bonferroni correction. Support vector machine classifiers were trained on broadband and frequency-specific microstate features to distinguish the conscious states.

Ciprofol increased delta-alpha power during LOC and maintained elevated beta-gamma activity during early ROC. Microstate parameters showed clear state-dependent changes across frequencies: microstate D decreased in the delta-theta bands during LOC; alpha-band microstates displayed reduced duration and increased occurrence; and microstates A, C, and E in the beta-gamma bands showed significant alterations in occurrence, coverage, or explained variance. Classification improved with sub-band features, with alpha band achieving the highest accuracy (0.841) compared with broadband features (0.754). Moreover, classification models integrating features of all sub-frequency bands yielded superior performance in distinguishing conscious states, achieving an accuracy of 0.971.

Ciprofol induces a distinct frequency-specific reorganization of cortical microstates, revealing multiscale network signatures of unconsciousness and recovery. Frequency-resolved microstate metrics may serve as sensitive markers for characterizing anesthetic-induced brain state transitions.