In the August 2007 issue of Anesthesiology, Velly et al.  1reported the difference in electroencephalograms from cortical and subcortical electrodes during anesthesia and concluded that they reflect different actions of anesthetics on cortical and subcortical structures. However, the authors failed to fully explore what they really were recording from scalp and depth electrodes. Our group has done several similar recordings,2and based on these, we want to make some corrections in the interpretation of Velly et al. 

The electroencephalogram consists of several patterns and components that may occur simultaneously and with varying topology. During anesthesia, we can record slow wave oscillations, which resemble the slow waves of sleep and are widespread. Some other patterns are seen in limited areas, such as the spindles.2,3In figure 1, we show examples of both during burst suppression in propofol anesthesia. The patient data are from our published material.2The uppermost trace is bipolar scalp derivation P4–F4. The next traces show depth 4–depth 1, where depth 1 is the contact at the tip of the electrode and depth 4 is the one closest to the vertex, in a four-contact electrode in the subthalamic nucleus, similar to the depth electrode of Velly et al.  1The next two traces are the right frontal electrode F4 and the right parietal electrode P4 referred to depth 1. The lowest trace shows that the spindle is recorded between two surface electrodes. P4 and A1, although simultaneously it is not visible in P4-F4. In all traces positive at the first, active electrode is down.

Fig. 1. Electroencephalogram during burst suppression in propofol anesthesia. The  uppermost trace is the electroencephalogram from two scalp electrodes, the right parietal P4 and the right frontal F4. The  second trace is the electroencephalogram recorded between two contacts in the depth electrode in the subthalamic nucleus. The  two lower traces are the electroencephalogram recorded between scalp electrodes P4 and F4, and the electrode contact deepest in the subthalamic nucleus. Notice that practically all waveforms that are recorded between the two contacts of the depth electrode are also recordable from surface electrodes with varying amplitude and polarity. The depth electrode recording and scalp recording therefore are generated by the same structures, most likely the cortex, not two independent generators, one at the thalamus and one at the cortex. The patient data are from Sonkajärvi  et al. 2
View largeDownload slide

Fig. 1. Electroencephalogram during burst suppression in propofol anesthesia. The  uppermost trace is the electroencephalogram from two scalp electrodes, the right parietal P4 and the right frontal F4. The  second trace is the electroencephalogram recorded between two contacts in the depth electrode in the subthalamic nucleus. The  two lower traces are the electroencephalogram recorded between scalp electrodes P4 and F4, and the electrode contact deepest in the subthalamic nucleus. Notice that practically all waveforms that are recorded between the two contacts of the depth electrode are also recordable from surface electrodes with varying amplitude and polarity. The depth electrode recording and scalp recording therefore are generated by the same structures, most likely the cortex, not two independent generators, one at the thalamus and one at the cortex. The patient data are from Sonkajärvi  et al. 2 

Fig. 1. Electroencephalogram during burst suppression in propofol anesthesia. The  uppermost trace is the electroencephalogram from two scalp electrodes, the right parietal P4 and the right frontal F4. The  second trace is the electroencephalogram recorded between two contacts in the depth electrode in the subthalamic nucleus. The  two lower traces are the electroencephalogram recorded between scalp electrodes P4 and F4, and the electrode contact deepest in the subthalamic nucleus. Notice that practically all waveforms that are recorded between the two contacts of the depth electrode are also recordable from surface electrodes with varying amplitude and polarity. The depth electrode recording and scalp recording therefore are generated by the same structures, most likely the cortex, not two independent generators, one at the thalamus and one at the cortex. The patient data are from Sonkajärvi  et al. 2
View largeDownload slide

Fig. 1. Electroencephalogram during burst suppression in propofol anesthesia. The  uppermost trace is the electroencephalogram from two scalp electrodes, the right parietal P4 and the right frontal F4. The  second trace is the electroencephalogram recorded between two contacts in the depth electrode in the subthalamic nucleus. The  two lower traces are the electroencephalogram recorded between scalp electrodes P4 and F4, and the electrode contact deepest in the subthalamic nucleus. Notice that practically all waveforms that are recorded between the two contacts of the depth electrode are also recordable from surface electrodes with varying amplitude and polarity. The depth electrode recording and scalp recording therefore are generated by the same structures, most likely the cortex, not two independent generators, one at the thalamus and one at the cortex. The patient data are from Sonkajärvi  et al. 2 

Close modal

Notice that the suppression, slow wave burst with approximately 10 Hz activity on it, and the spindle occur synchronously in derivations between scalp electrodes and between two contacts in the depth electrode. Their relative amplitude, however, changes depending on the relative amplitude of these waves at electrodes F4 and P4. The uppermost trace is the difference of the third and fourth traces. The slow wave and spindle are on the average of the same amplitude in scalp–scalp derivation, uppermost trace, and between the two contacts of the depth electrode. Note that from a recording between two electrodes in a volume conductor we cannot conclude where is the source of the electrical activity. Conclusions about the generator must be based on multielectrode recordings, as in the case of location of epileptic foci, and similarity with patterns with known generator such as the cortical slow waves.

The voltage fluctuation generated by the cortex spreads by volume conduction through cerebrospinal fluid, bone, and skin-to-scalp electrodes. It also spreads through brain tissue, which is a volume conductor. When the current passes the contacts of the depth electrode, a voltage is recorded that equals the product of the current and the impedance of the brain tissue between the two contacts of the electrode. This is also how voltage, scalp electroencephalogram, is recorded between two scalp electrodes. The signal recorded between the two contacts of the depth electrode therefore is the electroencephalogram recorded with a depth electrode in the subthalamic nucleus. It probably has little contribution from the nearby structures, because the differences between transcortical and scalp traces and the depth electrode trace can be explained by the sensitivity distribution of the electrodes.

In conclusion, the signals recorded by Velly et al. ,1both the signal from scalp electrodes and that from depth electrodes, are probably mainly electroencephalogram generated by the cerebral cortex, which is also evident from the illustrations of their article. The reason why they get different spectra and different correlation dimensions from the two signals is the different topography of the different cortical electroencephalographic patterns and therefore different contributions to the depth electrode–recorded electroencephalogram and scalp electrode–recorded electroencephalogram. The derivation they use for scalp electroencephalogram minimizes the slow waves, which are the most important indicators of the effect of anesthetics.4Rather than the electrosubcorticogram, the signal from the subthalamic electrode is the cortical electroencephalogram, recorded from or near the subthalamic nucleus. It is mainly the same cortically generated signal, the electroencephalogram, but recorded from the other side of the cortex.

The article by Velly et al.  therefore may not present the differences in electrical activity of deep structures and the cerebral cortex, as the authors claim. Both signals are mainly electrical activity of the cerebral cortex. This shows the importance of understanding the physiology and electrical fields of the electroencephalogram during anesthesia.

*Tampere University Hospital, Tampere, Finland. ville.jantti@uta.fi

1.
Velly LJ, Rey MF, Bruder NJ, Gouvitsos FA, Witjas T, Regis JM, Peragut JC, Gouin FM: Differential dynamic of action on cortical and subcortical structures of anesthetic agents during induction of anesthesia. Anesthesiology 2007; 107: 202–12
2.
Sonkajärvi E, Puumala P, Erola T, Baer GA, Karvonen E, Suominen K, Jäntti V: Burst suppression during propofol anaesthesia recorded from scalp and subthalamic electrodes: Report of three cases. Acta Anaesthesiol Scand 2008; 52:274–9
3.
Huotari AM, Koskinen M, Suominen K, Alahuhta S, Remes R, Hartikainen KM, Jäntti V: Evoked EEG patterns during burst suppression with propofol. Br J Anaesth 2004; 92:18–24
4.
Jäntti V, Sloan T: EEG and anesthetic effects, Intraoperative Monitoring of Neural Function. Edited by Nuwer MR. Elsevier BV, 2008, chap. 4, pp. 77–93. Handbook of Clinical Neurophysiology, vol. 8