Introduction: Anesthetics have been used to aid medical procedures for decades; however, despite careful monitoring, many people still experience explicit and sometimes traumatic conscious awareness under anesthesia (Orser, 2008). Current methods of monitoring rely on imperfect correlates of unconsciousness, making it impossible to distinguish between states with perfect accuracy. Low frequency (<7 Hz) power in electroencephalograms is commonly used to monitor the sleep-state, however, it is not present in all situations (Alkire, 2005). Newer methods that detect stability of models of cortical signals have recently been used to discriminate between awake and anesthetized states in non-human primates where other methods could not (Solovey, 2015). An open question remains; can a combination of these analyses into one model more accurately track conscious awareness? To address this, we examined brain activity recorded with electrocorticography (ECoG) from patients undergoing epilepsy surgery in both awake and anesthetized states.

Methods: While patients were being administered anesthetics, they were asked to respond to simple motor commands, to give a behavioral estimate of conscious awareness. We examined ECoG data recorded during this task for low frequency power, and (for the first time in humans) stability. We then analyzed the differences between states in each measure separately, and combined them into a novel model for predicting conscious awareness.

Results: Stability proved to be a successful indicator of consciousness in humans in most cases, despite lower doses and reduced cortical coverage compared to previous work, however the analysis showed complicated behavior across subjects. Additionally, when this was combined with low frequency power into one model, we could predict the conscious state with high accuracy.

Conclusions: With further development, this model has potential to lead to even more robust and generalizable prediction. Therefore, combining many current indicators of conscious awareness into one model could be developed into a powerful tool to aid physicians.

Patient Care: Despite the extensive use of anesthetics, patients still run a risk of experiencing traumatic conscious awareness during surgery. This research work towards discovering a more accurate correlate of that awareness, and therefore more accurate tracking of consciousness in clinical settings.

Learning Objectives: Examine whether stability analyses, which have predicted unconsciousness in nonhuman primates, do so for humans in clinical settings. Assess the efficacy of a model that combines several imperfect indicators of unconsciousness. Analyze which measure will contribute the most to predictions.

References: Alkire, M. T., & Miller, J. (2005). General anesthesia and the neural correlates of consciousness. Progress in Brain Research, 150, 229–244. https://doi.org/10.1016/S0079-6123(05)50017-7 Orser, B. A., Mazer, C. D., & Baker, A. J. (2008). Awareness during anesthesia. CMAJ?: Canadian Medical Association Journal = Journal de l’Association Medicale Canadienne, 178(2), 185–8. https://doi.org/10.1503/cmaj.071761 Solovey, G., Alonso, L. M., Yanagawa, T., Fujii, N., Magnasco, M. O., Cecchi, G. A., & Proekt, A. (2015). Loss of Consciousness Is Associated with Stabilization of Cortical Activity. Journal of Neuroscience, 35(30), 10866–10877. https://doi.org/10.1523/JNEUROSCI.4895-14.2015