The electroencephalogram (EEG) is a specialised piece of diagnostic and monitoring equipment. It uses electrodes to detect the electrical activity of the brain and to derive useful information from this. Understandably, the complexity of the brain’s electrical activity means that such information is relatively broad in its nature, hence the requirement for expert interpretation to gain key information. However, a basic understanding of the EEG is useful to general clinicians, such as to aid understanding of the different parts of depth of anaesthesia monitoring. This video is a useful basic introduction to EEG overall: https://www.youtube.com/watch?v=XMizSSOejg0 The ICE-TAP website is also a brilliant resource: http://icetap.org/
The EEG is one form of measuring biological potentials. Essentially, it is looking at the flow of current within tissues of the body. This results in a depolarization pattern that we know from basic cellular physiology. Of importance to the EEG is that this is clearly happening within multiple different tissues at differing times and with clearly differing vectors of polarity. At first look, this can create the impression of a chaotic mess, with a wildly oscillating electrical activity. However, closer examination using the magic of Fourier transformation reveals that this is actually made up of a pattern of repeating waves of differing frequencies which overlie one another. Breaking down the global signal into these components is one way that the signal can be better understood and useful information extracted.
Components of the waves that will be important include the frequency and amplitude. These can often vary inversely with each other i.e. higher frequency, lower amplitude. Additional aspects of the EEG that can be relevant for interpretation include the topography (the anatomical spread), cross correlation (how regions relate), and reactivity.
Measurement As with other biological signals, the measurement of the EEG is through electrodes. This is through a standardised placement called the 10-20 system (best seen in the initial video). This essentially uses landmarks of the skill to identify reproducible points across the scalp to place electrodes in order to achieve coverage of the desired underlying brain region. Importantly, a differential amplifier is used to aid signal detection. This is where the difference between the signal from two separate electrodes is used to produce the output signal i.e. removing the common signal and leaving what is different. The electrodes that are used for this are selected in a standardised manner to create the different channels that we see on the monitor. The formation of these channels together is called a montage. The signal input is also amplified and filtered to produce an interpretable image.
These signals are small, so having the appropriate voltage scale is needed to ensure that the output is interpretable i.e. the activity is not too big or too small). This is usually 5-10 microvolts for most patients.
Waveforms As noted above, fourier transformation of the signal input can reveal that there is an underlying pattern of multiple different waves. It is these wave patterns that are used in interpreting the EEG and they are described according to the frequency of their waveform. There are 5 core components with decreasing frequency.
Gamma (>30 Hz)
Alpha (8-12 hz)
Theta (4-8 Hz)
Delta (0-4 Hz)
Gamma (>30 Hz) These are the highest frequency waves. However, the skull is a notable insulator against waves of this frequency meaning they are hard to detect from standard EEG leads.
Beta (12-30 Hz) These waveforms are most obvious in the frontal regions and seen bilaterally and roughly symmetrically. They are the most common waveform in people in the awake and alert state and considered a ‘normal’ waveform.
Alpha (8-12 Hz) These waveforms are also considered as ‘normal’, but seen when the patient is relaxed and with eyes closed. They are more prominent in the occipital region and a higher amplitude in the dominant hemisphere. They tend to disappear during activation e.g. opening eyes or thinking.
Theta (4-8 Hz) These waveforms are not considered as normal in the awake adult. However, they are normally seen in adults during sleep, and are actually normal in children under the age of 13. They may also be seen in pathological conditions in adults, generally cases of diffuse disturbance such as encephalopathy.
Delta (0-4Hz) These slow waves are only really seen in stage 3 and 4 of sleeping adults. However, they are the dominant waveform in children under 1 year of age. Again, they can be seen in pathological conditions in adults, globally with diffuse pathology and more focally with localised subcortical lesions.
Morphology These specific waveforms are often not individual occurrences but overlap with each other within an ECG. The way they might overlap can present in a few ways:
Monomorphic - predominantly a single waveform
Polymorphic - a complex mix of several ovelying waveforms
Sinusoidal - a waveform that resembles a sine wave
Transient - a background picture with intermittent overlaps
As a good general rule for a ‘normal’ EEG:
Beta activity anteriorly
Alpha activity with reactivity (e.g. visual) posteriorly.
There are some patterns that are recognised as being associated with sleep, but also (more usefully for us) general anaesthesia. These are looked at in more detail elsewhere. The relevance is that they can hopefully be used to demonstrate a neurological state where awareness is not present - the goal of general anaesthesia.
Some of these features can be summarised as:
Greater uniformity of waveform patterns across the montage
Progression to lower frequency waveforms
In wakefulness, there are distinct regional differences noticeable across the different regions of the cortex. These begin to disappear with deepening sleep or anaesthesia.
There is also a shift towards lower frequency waveforms, moving away from the higher frequency forms seen in wakefulness. In deeper sleep and anaesthesia, there is more noticeable predominance of delta waves, seen as a slower oscillation.
In addition, there can be seen episodic ‘spindles’, which are bursts of EEG activity in the alpha range (or possibly beta range). The combination of these with the delta wave background, termed the ‘delta-spindle’ pattern, has been described as representing thalamic hyperpolarisation. This therefore theoretically represents the inhibition of thalamic transmission of sensory signals, and the subsequent unconsciousness that is a feature of sleep and anaesthesia. In even deeper anaesthesia, there is progression to almost purely delta waves.
Burst Suppression This is where there are periods or generalised suppression (almost isoelectric in appearance) punctuated by polymorphic bursts of electrical activity. This is an abnormal EEG state. In cases of anoxic brain injury it is highly correlated with a poor neurological outcome. It is not yet clear the relationship that is present between burst suppression and anaesthesia, although there is negative correlation. It isn’t clear whether harm is arising from excess anaesthesia, or whether the effect is purely epiphenomenal i.e. it is more vulnerable patients that will experience burst suppression.
Periodic discharges These are repeating waves that have a relatively similar morphology and temporal relationship to each other which occupy the majority of the recording. They come in different forms and are categorised by their topography and their inter-discharge interval. These different forms are associated with different pathologies e.g. left hemisphere PDs can be associated with herpes encephalitis or focal stroke. Generalised PDs can be associated with encephalopathies of varying aetiologies (e.g. toxic, ischaemic).
Suppression This refers to general reduction of wave amplitude (< 10 microV). It can be seen (reversibly) with anaesthesia, as well as a number of pathologies e.g. ischaemia, oedema. When associated with pathology in is generally a marker of significant cerebral disease and a concerning outlook. When this suppression is severe (<2 microV) it is worth being mindful of some reversible pathologies as the cause e.g. profound hypothermia, intoxication with CNS depressant.
Epileptiform The EEG will also clearly be able to demonstrate epileptiform activity that may be subclinical. As can be imagined, the appearance is of generalised high frequency discharges. This can be focal in nature, and identified as such on the EEG.
Drugs As noted regarding the anaesthetic pattern, sedative drugs generally result in a steady reduction in wave frequency. However, initial action may result in increased beta activity. Some classes of agents may show a very different pattern e.g. ketamine, N2O, clonidine, which may interfere with some depth of anaesthesia monitoring.
Encephalopathy This is a broad term referring to a pattern of generalised cerebral dysfunction, correlating with the clinical effects. The causes are clearly very broad, including brain trauma, systemic or local infection, drug toxicity and hepatic or renally driven. This initially looks like a slowing of the alpha rhythm, with progressive worsening associated with features of slow wave activity (theta and delta), triphasic waves and even burst suppression.