Nonlinear Dynamics Analyses of EEG Signals Capture Brain States at Different Levels of Consciousness  Bruce MacIver , Divya Chander, MD, PhD1*; Sarah L. Eagleman, PhD2; Christina Reynolds, PhD3,4; Nicholas T. Ouellette, PhD5 (Anesthesia, Stanford University, Stanford, CA )   C20

Anesthetic transition periods around loss and recovery of consciousness are not always captured by behavioral assessment or standard linear analyses of EEG. Propofol is one of the most widely used anesthetics for routine surgical anesthesia. Propofol administration alone produces EEG spectral characteristics similar to most hypnotics; however, inter-individual variation often makes spectral measures inconsistent. Complexity measures of EEG signals could offer universal measures to better capture anesthetic depth as brain activity exhibits nonlinear behavior at several scales. Patients undergoing propofol general anesthesia for various surgical procedures were identified as having changes in states of consciousness by the loss and recovery of response to verbal command after induction and upon recovery from anesthesia, respectively. We demonstrate that complexity measures, derived from nonlinear dynamic techniques, and captured in attractors, reliably and with sensitivity, distinguish such states. Notably, nonlinear dynamics analyses showed more significant differences between consciousness states than most spectral measures. Additionally, we found these measures are dependent on analysis features and show tight correlation with spectral measures during consciousness transition states. Thus, these measures could provide a means for reliably capturing depth of consciousness based on subtle EEG changes at the beginning and end of anesthesia administration. In addition, complexity is able to more fully describe how different these brain states are. For example, the attractors generated through time-delayed embeddings during anesthesia exhibit significantly different shapes, which have implications for network connectivity and information processing in the brain. This work supports existing theories on neural correlates of consciousness showing a diminished information carrying capacity of the brain during decreasing levels of consciousness.