Brain Monitoring

EEG and AEP monitoring during surgery

Dr. Gerhard Schneider, MD
Klinik für Anaesthesiologie,
Klinikum Rechts der Isar Technische Universität
Munich, Germany

Article also available in PDF: 47 KB

Presented at the 9th ESA Annual Meeting, Gothenburg, Sweden: April 7-10, 2001.

For many years, neurological signs have been used to quantify the effect of anesthetics on the brain. When balanced general anesthesia with a combination of drugs became popular, Guedel’s traditional classification of stages of general anesthesia [1] was not useful, as it was based on patients’ neurological signs.

Today, monitoring the target organ of general anesthesia can be accomplished using spontaneous and evoked electrical responses of the brain. This abstract focuses on the use of EEG and auditory evoked potentials (AEP) during general anesthesia and surgery.

Spontaneous EEG

The conventional EEG is recorded from scalp electrodes, and shows cortical electrical activity. This includes cortical manifestation of the sub cortical regions (projection pathways, thalamus, reticular formation, mesencephalon).

Recording electrodes
Standard placement of scalp electrodes follows the international 10-20 system. This allows for the anatomic localization of the signal. In order to obtain good signal quality, recording electrode impedance should be kept below 5kOhms, with only little differences between the electrodes. If the electrode impedance increases, background noise and artifacts may obscure the EEG signal. As a consequence, the signal to noise ratio would then decrease.

Characteristics of the EEG signal
The EEG signal can be described by three basic parameters:

• Amplitude (20-250 µV)
• Frequency (0.5-70 Hz)
• Time (continuous in raw EEG, epoch in processed EEG).

Processed EEG
Since the interpretation of the standard raw EEG is time consuming and requires much experience, different EEG processing methods have been developed to facilitate interpretation. Typically, raw EEG can be described as a sum of superimposed sine waves. The Fast Fourier analysis is a method to decompose the signal into sine waves, and this
analysis results in the power spectrum of the EEG. According to predefined frequencies, the spectrum can be divided into spectral bands: delta (0.5-4 Hz), theta (4-8 Hz), alpha (8-13 Hz) and beta (over 13 Hz). Further analysis can be applied to calculate single numerical parameters, such as spectral edge frequency (SEF) and median frequency (MF) [2]. Lately, scientific papers have appeared suggesting non-linear analysis as a target for future EEG research [3-5].

As the EEG reflects the functional status of the brain, it can be used to monitor its functional integrity. However, changes in EEG are not necessarily specific to underlying mechanisms. For example, slowing of EEG may reflect either changes in anesthetic concentrations or cerebral ischemia.

Auditory Evoked Potentials

Audio evoked potentials (AEP) reflect the function of the auditory pathway [6]. AEP can be evoked by repeated clicks of short duration (100-500 ms) given into an ear piece. Trigger-synchronized averaging of a defined number of EEG-segments is used to extract the AEP by reduction of the underlying EEG signal (background noise). The reduction in background noise is proportional to the square root of the number of averaged segments; the more averaged segments, the better the quality of the AEP. However, increasing number of segments, results in prolonged duration of measurement and delays display of results. Then, the system becomes less responsive to rapid changes.

The extracted AEP signal consists of a number of waves. Conventional analysis of the AEP response measures latencies and amplitudes of particular peaks. Three main groups of peaks can be distinguished and they can be correlated to the anatomical structures [7]:

• Brainstem AEP (BAEP) with latencies shorter than 10 milliseconds. Anatomical structures: cochlea, acoustic nerve (BAEP wave I, II), brainstem (BAEP wave III-V)
• Middle latency AEP (MLAEP) with latencies of 10-50 milliseconds. Anatomical structures: medial geniculate and primary auditory cortex (temporal lobe).
• Late cortical waves with latencies over 50 milliseconds. Anatomical structures: frontal cortex, association fields.

Sophisticated research level methods of AEP analysis include wavelet-transformation [8], and the use of an autoregressive model with exogenous input (ARX model) [9].

Some factors which may influence AEP
Variations in the auditory stimulus influence the AEP waveform. The increased volume of the auditory stimulus increases amplitudes and decreases latencies. If binaural stimulation is used, it can increase AEP amplitudes and decreases latencies.
Some physiological variables may also influence AEP. These include hypothermia (increasing latencies, decreasing amplitudes) and hyperthermia (producing the opposite effect). In addition, psycho-physiological factors like habituation, vigilance, and attention (with influence on late cortical waves) may have an impact. As an example, sleep and arousal leads to dramatic changes on MLAEP (wave peak Pa), up to complete loss of wave peaks Pb and P1 [6].

Pathophysiological factors may also influence AEP signals. That list includes conductive and sensory neural hearing disorders, demyelinating diseases (multiple sclerosis), ischemia, coma, and tumors.

Clinical Applications

Application of AEP
In anesthesia, clinical applications of AEP include use of BAEP (hardly influenced by anesthetics) in acoustic neurinoma (vestibulocochlear nerve) and posterior fossa surgery. Anesthetic drugs may influence MLAEP (Wave Peaks Pa, Nb). Scientific papers have appeared suggesting use of that information to quantify anesthesia [10].

Application of EEG in anesthesia and surgery
Although anesthetic drugs often cause drug-specific EEG changes, a general pattern of change in anesthesia can be observed. Most anesthetic drugs (not ketamine) lead to a decrease of frequencies and an increase of amplitudes.

Specific applications of the EEG include mapping of brain electrical activity for seizure surgery [11]. Epileptic discharges can be detected in the spontaneous EEG, which may not only be useful in

epileptic patients, but also with the use of some inhalation agents (enflurane, sevoflurane) [12-15]. In epileptic patients, not only seizure detectioncontrol of therapeutic measures (i.e. medications) can be accomplished by EEG [16].

Detection of brain ischemia
Before irreversible damage, global ischemia can be detected by a slowing of the EEG, followed by burst suppression and electrical silence. The electric failure precedes membrane failure [17].

The decrease of cerebral perfusion (40-50 ml min-1 100 g-1 brain tissue) results in progressive EEG changes (increase of polymorph slow waves, i.e. theta and delta) (16-20 ml min-1 100 g-1 brain tissue), loss of evoked brainstem potentials (12-15 ml min-1 100 g-1 brain tissue) before irreversible neuronal death occurs (<6 ml min-1 100 g-1 brain tissue).

The influence of mean arterial pressure, body temperature, arterial O2, and Hct can be assessed directly, as EEG provides functional monitoring. Focal changes can be judged in clinical context as e.g. provoked by a subdural hematoma, resulting in depression of amplitudes.

During surgery of the carotid artery, ischemia can be detected by observing differences between brain hemispheres. Characteristic changes include increase of polymorph slow waves (theta and delta) or changes of processed parameters, like spectral edge frequency (SEF). The changes usually occur within 60 seconds following ischemia [18-19].

In addition to detection of critical cerebral perfusion, the effect of therapy can directly be assessed. The effects of measures for cerebral protection (e.g. induction of total electrical suppression) can also be monitored [20].

Summary

The advantages of EEG and AEP monitoring outweigh their limitations, in spite of some complexity and multifactorial influences that need to be understood. Hence, monitoring of these signals has not been routine practice. However, their measurement would enable observation of functional integrity and changes, where clinical symptoms could not be reliably observed. The introduction of computerized signal analysis makes it easier for the non-expert to interpret the EEG and AEP signals in daily clinical work.

References

1. Guedel, A.E.: Third Stage Ether Anesthesia: A Sub-Classification Regarding the Significance of the Position and Movement of the Eyeball. American Journal of Surgery, Quarterly Supplement of Anesthesia and Analgesia, 1920. 34(4): p. 53-7.
2. Drummond, J.C., et al.: A comparison of median frequency, spectral edge frequency, a frequency band power ratio, total power, and dominance shift in the determination of depth of anesthesia. Acta Anaesthesiologica Scandinavica, 1991. 35(8): p. 693-9.
3. Viertiö-Oja, H.E., et al.: New Method to Determine Depth of Anesthesia From EEG Measurements. Abstracts of The Annual Meeting of the Society for Technology in Anesthesia www.anestech.org/publications/Annual_2000/Viertio-Oja.html, 2000.
4. Bruhn, J., H. Röpcke, and A. Hoeft: Approximate entropy as an electroencephalographic measure of anesthetic drug effect during desflurane anesthesia. Anesthesiology, 2000. 92(3): p. 715-26.
5. Abke, J., et al.: Detection of Inadequate Anesthesia by EEG Power and Bispectral Analysis. Anesthesiology, 1996. 85(3A): A477.
6. Thornton, C. and R.M. Sharpe: Evoked responses in anaesthesia. British Journal of Anaesthesia, 1998. 81(5): p. 771-81.
7. Picton, T.W., et al.: Human auditory evoked potentials. I. Evaluation of components. Electroencephalography & Clinical Neurophysiology, 1974. 36(2): p. 179-90.
8. Stockmanns, G., et al.: Wavelet-Analyse akustisch evozierter Potentiale wahrend wiederholter Propofol-Sedierung. Biomedizinische Technik, 1997. 42(S): p. 373-4.
9. Jensen, E.W., P. Lindholm, and S.W. Henneberg: Autoregressive modeling with exogenous input of middle-latency auditory-evoked potentials to measure rapid changes in depth of anesthesia. Methods of Information in Medicine, 1996. 35(3): p. 256-60.
10. Schneider, G. and P.S. Sebel: Monitoring depth of anaesthesia. European Journal of Anaesthesiology, 1997. 15(S): p. 21-8.
11. MacDonald, D.B. and N. Pillay: Intraoperative electrocorticography in temporal lobe epilepsy surgery. Canadian Journal of Neurological Sciences, 2000. 27(S 1): p. S85-91.
12. Woodforth, I.J., et al.: Electroencephalographic evidence of seizure activity under deep sevoflurane anesthesia in a nonepileptic patient. Anesthesiology,1997. 87(6): p. 1579-82.
13. Yli-Hankala, A., et al.: Epileptiform electroencephalogram during mask induction of anesthesia with sevoflurane. Anesthesiology, 1999. 91(6): p. 1596-603.
14. Kaisti, K.K., et al.: Epileptiform discharges during 2 MAC sevoflurane anesthesia in two healthy volunteers. Anesthesiology, 1999. 91(6): p. 1952-5.
15. Hilty, C.A. and J.C. Drummond: Seizure-like activity on emergence from sevoflurane anesthesia. Anesthesiology, 2000. 93(5): p. 1357-9.
16. Van Ness, P.C.: Pentobarbital and EEG burst suppression in treatment of status epilepticus refractory to benzodiazepines and phenytoin. Epilepsia, 1990. 31(1): p. 61-7.
17. Drummond, J.C. and Patel P.J.: Cerebral physiology and the effects of anesthetics and techniques. In: Miller R.D. (ed) Anesthesia. Churchill Livingstone, New York, 2000: p. 695-733.
18. Rampil, I.J., et al.: Prognostic value of computerized EEG analysis during carotid endarterectomy. Anesthesia & Analgesia, 1983. 62(2): p. 186-92.
19. Minicucci, F., et al.: Computer-assisted EEG monitoring during carotid endarterectomy. Journal of Clinical Neurophysiology, 2000. 17(1): p. 101-7.
20. Doyle, P.W. and B.F. Matta: Burst suppression or isoelectric encephalogram for cerebral protection: evidence from metabolic suppression studies. British Journal of Anaesthesia, 1999. 83(4): p. 580-4.

Last updated: 1 June 2001Created
Legal notice	© GE Healthcare 2008 ISSN 1795-6269 (Web) ISSN 1795-6277 (CD)	Webmaster