The electroencephalogram (EEG), a graphic record of the electrical activity of the brain, can be used for patient monitoring in the operating theatre and in the intensive care unit.
Indications for the use of EEG monitoring in the operating theatre are the assessment of hypnotic drug effects, the early detection of the effects of hypoxia, and the depiction of the EEG effects of induced hypothermia. In the intensive care unit, EEG monitoring can be used for the control of sedation, for therapy control, e.g. in status epilepticus, for the assessment of a patient`s current clinical status and trends thereof, and as an exploratory diagnostic tool with respect to epileptiform activity and focal brain disorders.
Generalized and focal EEG changes
Attenuation of brain function is typically accompanied by a slowing of the EEG. The frequency composition of the EEG can therefore be used to support the assessment of the severity of alterations or impairments of brain function. Furthermore, EEG potentials with a specific shape, e.g. epileptiform potentials, are of interest in EEG monitoring.
On principle, generalized and focal EEG changes have to be distinguished. Typical examples for generalized EEG changes are hypnotic drug effects and alterations of the EEG caused by hypothermia, hypoglycaemia or global hypoxia. A focal EEG change is restricted to a circumscribed area, it can be caused e.g. by a tumour, a bleeding or by localized hypoxia.
Conventional multi-channel EEG recordings take a considerable practical effort. But generalized EEG changes can be assessed on the basis of a single EEG channel with high reliability. This is an important precondition for the use of the EEG as a monitoring method.
The conventional visual assessment of the EEG requires expert knowledge and experience. Therefore, it is a considerable simplification for the user in the operating theatre and in the intensive care unit, if the EEG is analyzed automatically by means of computer-based methods and if support for the interpretation is given.
Spectral analysis is a mathematical method for signal evaluation. The signal is subdivided into segments, which are analysed with regard to their frequency components.
The result is a power spectrum, which can be displayed graphically and can be used as a basis for further evaluations. Single spectral parameters can be extracted, as e. g. the 50 % and the 90 % (or 95 %) quantile of the spectrum, which are also called median and spectral edge frequency. Other commonly used spectral parameters are the total power and the power in the different frequency bands. The following frequency bands are distinguished: delta (0.5 – 3.5 Hz, theta (3.5 – 7.5 Hz) alpha (7.5 – 12.5 Hz), beta (> 12.5 Hz). Such monoparameters are often able to show general trends within the EEG, but the complex information of an EEG cannot be described adequately by monoparameters.
For the purpose of EEG monitoring in the operating theatre and in the intensive care unit, also other mathematical and statistical parameters are used.
EEG stages of anaesthesia and sedation
Hypnotic drug effects are accompanied in a dose related manner by a progressive slowing of the EEG. Criteria to classify these EEG changes visually into the stages A (awake) to F (very deep hypnosis) were proposed by Kugler (1981).
A scale from A to F which was modified for the purpose of EEG monitoring served as the basis for the development of the automatic EEG classification incorporated in the Narcotrend (Figure). It was realized by means of a multivariate approach using special pattern recognition algorithms. For the assessment of the hypnotic component of anaesthesia and sedation with this device, one EEG channel is used as a standard, for this three electrodes have to be attached to the patient`s forehead. Depending on the kind of the surgery performed or if injuries to the head exist, other electrode types (e.g. needle electrodes) or electrode positions can be used.
Two channel recordings are also possible.
The Narcotrend analyzes the EEG in realtime, the assessments are updated every second.
Visually classified EEG epochs are a stable and clear basis for the development of classification algorithms and support easy validation. Using a visual EEG classification as a basis means methodological transparency for the user, because a clear relationship exists between the characteristics of the raw EEG and the automatic classification. This can be regarded as an important advantage of the Narcotrend compared to other concepts for automatic EEG classification which do not comprise a defined relationship between the raw EEG and mathematical-statistical parameters, as e.g. indexes.
References:Fish BJ. Fish and Spehlman`s EEG Primer: Basic Principles of Digital and Analog EEG. Third Edition.
Amsterdam New York: Elsevier 1999
Cerebral Monitoring in the OR and ICU.
Dordrecht: Springer, 2005
Niedermeyer E, Lopes da Silva FH (Eds) Electroencephalography. Basic Principles, Clinical Applications, and Related Fields. Fifth Edition.
London: Williams and Wilkins 2005.
Tatum WO, Husain AM, Benbadis SR, Kaplan PW. Handbook of EEG Interpretation.
New York: Demos Medical Publishing 2008