The influence of induction speed on the frontal (processed) EEG

Abstract

The intravenous injection of the anaesthetic propofol is clinical routine to induce loss of responsiveness (LOR). However, there are only a few studies investigating the influence of the injection rate on the frontal electroencephalogram (EEG) during LOR. Therefore, we focused on changes of the frontal EEG especially during this period. We included 18 patients which were randomly assigned to a slow or fast induction group and recorded the frontal EEG. Based on this data, we calculated the power spectral density, the band powers and band ratios. To analyse the behaviour of processed EEG parameters we calculated the beta ratio, the spectral entropy, and the spectral edge frequency. Due to the prolonged induction period in the slow injection group we were able to distinguish loss of responsiveness to verbal command (LOvR) from loss of responsiveness to painful stimulus (LOpR) whereas in the fast induction group we could not. At LOpR, we observed a higher relative alpha and beta power in the slow induction group while the relative power in the delta range was lower than in the fast induction group. When concentrating on the slow induction group the increase in relative alpha power pre-LOpR and even before LOvR indicated that frontal EEG patterns, which have been suggested as an indicator of unconsciousness, can develop before LOR. Further, LOvR was best reflected by an increase of the alpha to delta ratio, and LOpR was indicated by a decrease of the beta to alpha ratio. These findings highlight the different spectral properties of the EEG at various levels of responsiveness and underline the influence of the propofol injection rate on the frontal EEG during induction of general anesthesia.

Figure 1

Spectral properties of patients with slow and fast propofol induction. Differences in the spectral EEG features between patients with fast and slow induction. (A) Density spectral array (DSA) of the normalized power spectral density (PSD) for patients with slow (top) and fast (centre) induction. The bottom plot presents the difference in power for each time and frequency between patients with fast and slow induction. Only pixels with a significant difference are displayed. For the patients with slow induction, strong alpha and delta activity was already present before loss of responsiveness to painful stimuli (LOpR), whereas in the patients with fast induction these patterns developed after LOpR. The difference plot revealed higher relative power in the delta range in patients with fast induction, whereas patients with slow induction had higher relative beta band starting around LOpR and higher relative alpha power in the entire observation period. (B) PSD plot showing the normalized spectral power for patients with slow (blue) and fast (orange) induction in combination with the area under the curve (AUC) and 95% confidence interval (CI) for each frequency for the 10 s after LOpR. Black dots in the AUC plot indicate significance and grey dots an AUC > 0.7. The “x” indicate the boundaries of the 95% CI. Patients with slow induction had less relative power in the delta range and higher relative power in the alpha and beta range. (C) The course of beta ratio in patients with fast induction showed a strong decrease before LOpR and it was significantly different from patients with slow induction before LOpR as indicated by the AUC and 95% CI presented in the lower graph.

Figure 2

DSA and PSD of patients with slow induction at loss of responsiveness to verbal command vs. painful stimulus. Differences in the spectral EEG features in patients with slow induction between the loss of responsiveness to verbal (LOvR) or painful (LOpR) stimulation. (A) Density spectral array (DSA) of the normalized power spectral density (nPSD) for patients undergoing slow induction centred to the LOvR. The development of the characteristic alpha and delta dominant power indicative of an anaesthesia EEG pattern develops prior to LOvR. (B) DSA of the nPSD for patients undergoing slow induction centred to the LOpR. When centred to LOpR the increase in EEG alpha-band power around 10 Hz is more pronounced as if centred to LOvR as in (A). (C) PSD plot showing the normalized spectral power for patients at LOvR (green) and LOpR (blue) in combination with the area under the curve and 95% confidence interval (CI) for each frequency for the 10 s after LOR. Black dots indicate significance and grey dots an AUC > 0.7. At LOpR patients had higher relative power in the 10–15 Hz range. The “x” indicate the boundaries of the 95% CI. (D) Ratio of the spectral power between LOvR and LOpR. The plot highlights that the power in the 10–15 Hz frequency band is significantly higher at LOpR indicated by the black and gray dots in the AUC plot.

Figure 3

Relative band powers of patients with slow induction at loss of responsiveness to verbal command and painful stimulus. Course of the relative band powers for the 10 min before loss of responsiveness to verbal command (LOvR) in green and loss of responsiveness to painful stimulus (LOpR) in blue. The orange shading indicates the 95% confidence interval of the particular frequency band considering the first 10 s after LOvR or LOpR. The lower plots display the area under the curve. Black dots indicate significance. The “x” indicate the boundaries of the 95% CI. (A) The relative delta band power was significantly higher until around 300 s to 400 s before LOvR. The relative delta power decreased after LOvR. (B) The relative delta band power was significantly higher until around 200 s before LOpR, but did not decrease further after LOpR. (C) The relative theta band power was significantly lower for a short period before LOvR. (D) The relative theta band power was significantly lower for a short period before LOpR. (E) The relative alpha band power was significantly lower until around 3 min before LOvR and significantly higher after around 3 to 4 min after LOvR. (F) The relative alpha band power was significantly lower until around 2–3 min before LOpR and continued to increase after LOpR. (G) The relative beta power was significantly lower till around 300 s before LOvR. For the remainder of the observation period, it was not significantly different compared to the LOvR. (H) The relative beta power was significantly lower till around 200 s before LOpR and again starting after around 250 s 450 s LOpR.

Figure 4

Band power ratios of patients with slow induction at loss of responsiveness to verbal command and painful stimulus. Course of the alpha delta and beta alpha ratio for the 10 min before loss of responsiveness to verbal command (LOvR) in green and loss of responsiveness to painful stimulus (LOpR) in blue. The orange shading indicates the 95% confidence interval of the particular band power ratio considering the first 10 s after LOvR or LOpR. The lower plots display the area under the curve. Black dots indicate significance. The “x” indicate the boundaries of the 95% CI. (A) The alpha to delta ratio was significantly lower till 3 min before LOvR and significantly higher around 5 min after LOvR. (B) The alpha to delta ratio was significantly lower till 2–3 min before LOpR, but did not increase significantly afterwards. (C) The beta to alpha ratio was significantly higher 250–150 s before LOvR and significantly lower around 3 min after LOvR. (D) The beta to alpha ratio was significantly higher until around 150 s before LOpR and significantly lower around 3 min after LOvR.

Figure 5

Visualization of the trajectories of patients with slow induction at loss of responsiveness to verbal command and painful stimulus using the PCA algorithm. Trajectories of different EEG band powers and band power ratios in the principle component (PCA) space constructed by the first two PCAs during loss of responsiveness to verbal command (LOvR) and loss of responsiveness to painful stimulus (LOpR). The blue to red colours indicate the time period relative to LOvR/LOpR starting with dark blue for the 10 min to 5 min episode before LOvR/LOpR until the 5 min to 10 min episode in dark red. (A) LOvR alpha: The colours change in a directional way, but with substantial overlap. (B) LOvR beta to alpha ratio: The colours change in a directional way with some overlap in the blue (pre-LOvR) colors. (C) LOvR alpha to delta ratio: The colours change in a directional way. (D) LOpR alpha: The colours change in a directional way, but with substantial overlap, especially in the red colours (post LOpR). (E) LOpR beta: The colours change in a directional way, but with substantial overlap, of the light blue and light red colour indicating the transient increase of beta power around LOpR. (F) LOpR beta to alpha ratio: The colours change in a directional way with almost completely overlapping red colours.

Figure 6

Performance of different EEG parameters to detect loss of responsiveness to verbal command and painful stimulus. This plot displays the performance of various (processed) EEG parameters to detect loss of responsiveness to verbal command (LOvR) and painful stimulus (LOpR). A change of colours around the LOvR/LOpR indicates a significant in-/decrease of the parameter when compared to the parameter at LOvR/LOpR. (A) LOvR: The relative alpha power and the alpha to delta ratio show a steady change throughout the transition. The beta to alpha ratio also shows the change in parameter, but less pronounced. (B) LOpR: The relative alpha power and the beta to alpha ratio show a steady change throughout the transition.