Dynamics of Recovery from Anesthesia-Induced Unconsciousness

Post by Stephanie Williams

What's the science?

Many research groups have attempted to understand how anesthetics alter local and global neural activity during the loss and recovery of consciousness. Significant progress has been made in understanding how the brain changes during a loss of consciousness, but the relationship between dynamics that occur during anesthesia-induced loss of consciousness and during the recovery from anesthesia-induced unconsciousness is not well understood. This week in Brain, Patel, and colleagues recorded neural activity from primate cortex to characterize the dynamics that distinguish recovery from propofol-induced unconsciousness from loss of propofol-induced unconsciousness. 

How did they do it?                             

The authors recorded neural activity from the primary somatosensory cortex, secondary somatosensory cortex, and ventral premotor area of 2 adult male monkeys during the induction, maintenance, and cessation of propofol anesthesia. The authors designed a task that required 1) a behavioral response from the monkeys, and also consisted of 2) multisensory stimulation that did not require a response. The monkeys were required to press a button within 1.5 seconds of hearing a sound, and keep their hand on the button until the end of the trial in order to receive a juice reward. During the task, the authors randomly presented sound stimuli or a combination of sound and air-puff stimuli. The button press portion of the task allowed the authors to track behavioral state (as monkeys lost, and then subsequently recovered, consciousness). The sensory stimulation portion (sound, air puff) allowed the authors to track neural dynamics in response to sensory stimulation across all behavioral states, The authors focused on two behavioral metrics: 1) the monkey’s task engagement — whether the monkey made any attempt to initiate a response during a trial even if it was incorrect, and 2) the monkey’s task performance — the probability of a correct response. Separating these two behavioral endpoints allowed the authors to distinguish both whether the monkey was conscious and whether the monkey was performing the task well.

After the monkey had performed the task, the authors began a propofol infusion and recorded local field potentials and spiking activity. They also performed a series of spectral analyses to understand how power at different frequencies changed as the animals transitioned between different behavioral states. The authors performed clustering analyses on the local field potential data - they analyzed the velocity of the transition between different clusters to understand state transitions. Next, the authors investigated how communication between brain regions changed as the animals regained consciousness by calculating local and regional coherence. The authors analyzed single neuron responses across different stages of consciousness from 1) neurons that responded to both air puffs and sound with increased firing rates, 2) neurons that responded to air puffs and sound with decreased firing rates, 3) neurons that only responded to air puffs and 4) neurons that only responded to sound. 

What did they find? 

As the monkeys regained consciousness, the authors observed an abrupt shift marked by increases in beta oscillations and disruptions in slow-delta oscillations in the somatosensory and premotor cortex. Clustering analysis showed two distinct high-density clusters and a high-velocity transition between the clusters. The clusters, which corresponded to anesthesia and wakefulness, were distinguishable by the level of the monkey’s task engagement.  Coherence analysis revealed that beta oscillations were immediately coherent both locally and inter-regionally when the animals regained consciousness and began performing the task again. The authors observed that interregional coherence significantly increased in power as the animals transitioned to a behavioral state in which they were performing the task at a pre-anesthesia level. 

The average neuronal firing rate decreased during the propofol anesthesia and began increasing again after the end of the propofol infusion, consistent with previous reports. The recovery of firing rates tended to differ between the regions the authors recorded. Recovery of firing rate in ventral premotor area neurons preceded recovery in somatosensory areas, suggesting that there are region-specific firing responses during changes in consciousness. When the authors analyzed the response of individual neurons to puffs of air or noises across different behavioral states, they found that neurons that had previously responded to puffs of air continued to do so. In contrast, neurons that had previously responded to sounds no longer showed a significant response. The authors found the recovery of single-neuron responses to the pre-anesthesia baseline as correlated with behavioral performance recovery, not simply the duration of the recovery time. 

What's the impact?

This study characterizes the neural dynamics of recovery from propofol induced-unconsciousness and complements our understanding of how local field potentials and single neural dynamics relate to specific behavioral endpoints. Importantly, the authors suggest the changes that occur during the recovery of consciousness are not merely an inverse of the change that occurs during the loss of consciousness, and instead can be described as an abrupt shift of dynamics that include the return of inter-regionally coherent beta oscillations.

Patel et al. Dynamics of Recovery from Anesthesia-Induced Unconsciousness Across Primate Neocortex. Brain. (2020). Access the original scientific publication here.