The effects of propofol on local field potential spectra, action potential firing rate, and their temporal relationship in humans and felines

Abstract

Propofol is an intravenous sedative hypnotic, which, acting as a GABAA agonist, results in neocortical inhibition. While propofol has been well studied at the molecular and clinical level, less is known about the effects of propofol at the level of individual neurons and local neocortical networks. We used Utah Electrode Arrays (UEAs) to investigate the effects of propofol anesthesia on action potentials (APs) and local field potentials (LFPs). UEAs were implanted into the neocortex of two humans and three felines. The two human patients and one feline received propofol by bolus injection, while the other two felines received target-controlled infusions. We examined the changes in LFP power spectra and AP firing at different levels of anesthesia. Increased propofol concentration correlated with decreased high-frequency power in LFP spectra and decreased AP firing rates, and the generation of large-amplitude spike-like LFP activity; however, the temporal relationship between APs and LFPs remained relatively consistent at all levels of propofol. The probability that an AP would fire at this local minimum of the LFP increased with propofol administration. The propofol-induced suppression of neocortical network activity allowed LFPs to be dominated by low-frequency spike-like activity, and correlated with sedation and unconsciousness. As the low-frequency spike-like activity increased and the AP-LFP relationship became more predictable firing rate encoding capacity is impaired. This suggests a mechanism for decreased information processing in the neocortex that accounts for propofol-induced unconsciousness.

Keywords: consciousness; local field potential; microelectrode array; power spectrum; propofol.

Figure 1

Boluses of propofol decreased the high-frequency power in the LFP spectra spectrogram and low-pass filtered traces (200 Hz) from one representative electrode for Patient A (A) and Feline A (B). Patient A received one bolus of at ∼200 s, while Feline A received two boluses of propofol at ∼100 and ∼850 s. Colored rectangles above the spectrograms correspond with the level of anesthesia. Red represents the baseline or awake state. Blue represents the lightly anesthetized state. Green represents the deeply anesthetized state. White rectangles indicate the time periods and frequency bands chosen for the average power calculations across the array.

Figure 2

Target-controlled infusion of propofol gradually decreased the high-frequency power in the LFP spectra. Spectrogram and low-pass filtered traces (200 Hz) from one representative electrode in Feline C. Top plot show the predicted propofol plasma concentration using the target-controlled infusion system (red) and the TCI pump speed (blue) over the duration of the experiment. Induction and emergence from anesthesia occurred at nearly the same propofol concentrations as noted by the black arrows. Colored rectangles below the spectrogram correspond with the level of anesthesia. Red represents the awake state. Blue represents the lightly anesthetized state. Green represents the deeply anesthetized state. Purple represents the isoelectric state. White rectangles indicate the time periods and frequency bands chosen for the average power calculations across the array. With very high propofol concentrations, power in the LFP decreased across all frequency bands. After the peak propofol concentration is reached and the propofol concentration decreases, the power in the LFP increases.

Figure 3

Boluses of propofol decreased firing rate across array. Raster plots and firing rate of Patient A (A) and Feline A (B). Patient A received one bolus of at ∼200 s, while Feline A received two boluses of propofol at ∼100 and ∼850 s. Propofol boluses resulted in decreased AP firing rate as seen in the raster plots and firing rate histograms. The blue line represents the firing rate histogram and the gray dashed lines represent an interval of 4 standard errors wide centered at the mean. Colored rectangles above the plots correspond with time periods chosen for the average firing rate calculations. Red represents the baseline or awake state. Blue represents the lightly anesthetized state. Green represents the deeply anesthetized state. With the emergence from anesthesia, the firing rate begins to increase in Feline A after both boluses of propofol.

Figure 4

Target-controlled infusion of propofol decreased firing rate across the array for Feline C. Controlled infusion of propofol resulted in decreased AP firing rate as seen in the raster plot and firing rate histogram. The blue line represents the firing rate histogram and the gray dashed lines represent an interval of 4 standard errors wide centered at the mean. Colored rectangles above the plots correspond with time periods chosen for the average firing rate calculations. Red represents the awake state. Blue represents the lightly anesthetized state. Green represents the deeply anesthetized state. Purple represents the isoelectric state. With the emergence from anesthesia, the firing rate begins to increase 175 min into the recording.

Figure 5

Action potential-aligned LFP plots for one representative channel from Patient A (A), Feline A (B), Feline B (C), and Feline C (D) in three distinct brain states are shown. AP-aligned LFP exhibited a negative-going spike-like potential proximal in time to the APs. As the anesthesia increased, the amplitude of the spike-like LFP increased. Dashed lines represent the control cases in which randomly generate AP times were used to align the LFP. Red represents the awake state. Blue represents the lightly anesthetized state. Green represents the deeply anesthetized state.

Figure 6

Probability density of the temporal separation between APs and the minimum LFP value for each AP on all electrodes for Patient A (A), Feline A (B), Feline B (C), and Feline C (D) at three different levels of anesthesia. The integral of the probability density for the 50-ms around AP firing (shaded region) yielded the probability that the minimum value of the LFP and AP firing occurred with 50 ms of each other, i.e., they had a consistent temporal relationship. With propofol administration, the probability of an AP occurring at the local minimum of the LFP increased. Dashed lines represent the control cases in which randomly generated AP times were used to align the LFP. Red represents the awake or baseline state. Blue represents the lightly anesthetized state. Green represents the deeply anesthetized state.