Continuous regimens of cortico-motor integration calibrate levels of arousal during emergence from anesthesia

Abstract

31Background: Recovery to a conscious state when emerging from anesthesia requires 32 full cortical desynchronization, initiation of movement and behavioral reactivity to 33 sensory stimuli. However, the variety of cortical electroencephalogram (EEG) patterns 34 associated with specific anesthetics and the paucity of behavioral descriptions during 35 emergence from anesthesia have prevented EEG and behavior as feasible tracking 36 methods to assess emerging from anesthesia. We propose a detailed combined 37 analysis of motor and cortical activity to determine levels of arousal in rodents. 39Methods: Using decreasing anesthetic concentrations, we simultaneously recorded 40 local field potentials (LFPs) and movement in mice. We delineated cortical dynamics 41 and sub-states during emergence from anesthesia by applying a smoothed-Z score to 42 extract dominant frequencies from spectrogram. Then, we implemented KMeans to 43 obtain cortical sub-states. Finally, we used density estimation and an abrupt change 44 detection algorithm to segment cortical activity into periods. We used cortical sub-states 45 obtained during isoflurane traces to supervise sub-states in sevoflurane and a 46 pharmacologically induced-arousal model. This information together with examining 47 videos were used to categorize behavior. 49Results: We identified five cortical periods with restored motor behavior during 50 emergence from isoflurane anesthetic. Periods of structured sub-states denoted when 51 specific motor behaviors occurred. No significant differences were found when 52 comparing the combined cortical features and motor behavior using isoflurane, 53 3 sevoflurane and our arousal-rodent model. We describe graded regimens of cortico-54 motor activity during emergence from anesthesia to assess arousal levels. 55 56 Conclusion: We show cortical patterns denote gradual motor behaviors when emerging 57 from anesthesia. Restoring motor behavior is a dynamic process that begins tens of 58 minutes earlier than the righting reflex. Combining cortical activity and motor behavior 59 unveils novel biomarkers to accurately track emerging from general anesthesia in 60 rodents and likely other species. 61 62 Key words: Emergence from anesthesia, movement, behavioral arousal, cortical 63 biomarkers, cortical periods, spectral analysis, smoothed Z score algorithm, k-means 64 classifier, righting reflex 65 66 67 68 69 70 71 72 4Introduction: 73 Recovery to a conscious state in humans usually comprises full cortical 74 desynchronization, initiation of movement and behavioral reactivity to sensory stimuli 1 . 75 However, the combination of these parameters in rodents is rarely used when assessing 76 arousal when emerging from anesthesia 2 or when investigating the neuronal pathways 77 underlying reversing anesthetic states in animal models 3-6 . A clear discrepancy arises 78 when analyzing cortical activity 7 and behavior among anesthetics. Some data suggests 79 excluding these parameters. For example, the onset of movement during the emergence 8...

Robust alternative to the righting reflex to assess arousal in rodents

“…We applied local anesthetic as described before 35 . After opening an incision in the scalp, we created a subcutaneous pocket along the animal's dorsal flank and placed the body of the transmitter (model F20-EET;DSI) into the pocket making sure biopotential leads were oriented cranially.…”

“…Spontaneous ventilation was maintained throughout the experiment. Respiratory rate was continuously monitored by the investigators as described before 35 . Temperature was maintained at approximately 37 °C using a temperature regulator coupled to a rectal temperature probe (CWE Inc).…”

“…The anesthetic ramp was initiated upon exposing animals to isoflurane with a starting concentration of 1MAC (1.25%). Anesthetic concentration was reduced at intervals of 0.25% as previously described 35 . Each interval lasted for 30 min until reaching 0% anesthetic.…”

“…Detection of dominant frequency band. We defined the dominant frequency band in a cortical spectrum as the band that surpasses other bands in power and that we previously described 35 . Briefly, it is identified by first measuring the mean power of each 50 log-spaced frequency 42 between 2 and 150 Hz (denoted as x = x 1 , x 2 , .…”

“…Segmentation of cortical period. We delineated cortical dynamics through occurrence density of the classified cortical states and then segmented this into periods by applying an abrupt change detection algorithm as we described earlier 35 . To achieve density estimation, we implemented the locfit.m function in the Chronux toolbox in MATLAB.…”