Characteristics of EEG Microstate Sequences During Propofol-Induced Alterations of Brain Consciousness States

Monitoring the consciousness states of patients and ensuring the appropriate depth of anesthesia (DOA) is critical for the safe implementation of surgery. In this study, a high-density electroencephalogram (EEG) combined with blood drug concentration and behavioral response indicators was used to monitor propofol-induced sedation and evaluate the alterations in consciousness states. Microstate analysis, which can reflect the semi-stable state of the sub-second activation of the brain functional network, can be used to assess the brain's consciousness states. In this research, the EEG microstate sequences were constructed to compare the characteristics of corresponding sequences. Compared with the baseline (BS) state, the microstate sequences in the moderate sedation (MD) state exhibited higher complexity indexes of the multiscale sample entropy. With respect to the transition probability (TP) of microstates, most microstates tended to be converted into microstate C in the BS state. In contrast, they tended to be converted into microstate F in the MD state. The significant difference between the expected TP and observed TP could lead to the conclusion that hidden layers were present when there were changes in the consciousness states. According to the hidden Markov model, the accuracy of distinguishing the BS and MD states was 80.16%. The characteristics of microstate sequence revealed the variations in the brain states caused by alterations in consciousness states during anesthesia from a new perspective and presented a new idea for monitoring the DOA.