Electroencephalography

Measurement of cerebral cortical electrical activity is one of the oldest methods of neurophysiological monitoring. The EEG signal represents summated post-synaptic potentials (20–200 μV) of pyramidal neurones, typically measured simultaneously between several pairs of scalp electrodes (Figure 31.1).

Figure 31.1 The internationally standardized 10-20 system of EEG electrode placement. Four bony landmarks – the nasion, inion and preauricular points – act as a reference and the cranium is then apportioned into 10% or 20% segments between these landmarks, as originally described by Jasper. F, frontal; Fp, frontal polar; C, central; O, occipital; P, parietal; T, temporal; A, ear lobe; Pg, nasopharyngeal. Right-sided placements are indicated by even numbers, left-sided placements by odd numbers and midline placements by Z. Auricular (A₁ and A₂) positions complete the standard 21-electrode positions. The figure also identifies the anticipated locations of the Rolandic and Sylvian fissures, and commonly used additional Pg and cerebellar (Cb) electrodes. The location and nomenclature of these electrodes is standardized by the American Clinical Neurophysiology Society, formerly the American Electroencephalographic Society.

(Adapted with permission from Malmivuo J, Plonsey R. Bioelectromagnetism – Principles and Applications of Bioelectric and Biomagnetic Fields. New York: Oxford University Press; 1995)

The EEG is described in terms of location, amplitude and frequency. Frequency is conventionally grouped into four bands: δ, θ, α and β (Table 31.2). A normal awake adult has a posteriorly located, symmetrical EEG frequency of around 9 Hz (i.e. α rhythm). Opioids and most anaesthetic agents produce a dose-dependent EEG slowing (↓ α and ↑ δ and θ) culminating in periods of very low EEG amplitude and burst suppression (periods of high-voltage electrical activity alternating with periods of silence). N₂O induces high-frequency frontal activity and decreased amplitude. Ketamine increases the EEG amplitude at low doses and slows the EEG at higher doses.

Table 31.2 EEG waveforms

Waveform	Frequency (Hz)		Amplitude (μV)	Comments
Delta (δ)	1.5–3.5		>50	Normal during sleep and deep anaesthesia, indication of neuronal dysfunction
Theta (θ)	3.6–7.5		20–50	Normal in children and elderly, normal adults during sleep, produced by hypothermia
Alpha (α)	7.6–12.5		20–50	Awake, relaxed, eyes open, mainly over occiput
Beta (β)	12.6–25		<20	Awake, alert, eyes open, mainly in parietal cortex, produced by barbiturates, benzodiazepines, phenytoin, alcohol
Gamma (γ)	25.1–50		<20

The EEG has long been regarded as the ‘gold standard’ for the detection of cerebral ischaemia. At constant temperature and depth of anaesthesia (DOA), progressive ischaemia produces a reduction in total power and slowing – decreased α and β power and increased δ and θ power. These changes only become apparent when the cerebral blood flow (CBF) halves (i.e. <50 ml 100 g^–1 min^–1). An EEG amplitude attenuation of <50% or increased δ power is regarded as being indicative of mild ischaemia, whereas >50% attenuation or a doubling in δ power is regarded as being indicative of severe ischaemia. An isoelectric or ‘silent’ EEG is seen when CBF < 7–15 ml 100 g^–1 min^–1. EEG changes are not specific for pathology.

Under conditions of constant DOA and autoregulated cerebral perfusion, the effects of hypothermia on the EEG can be used to guide temperature management in cases requiring deep hypothermic circulatory arrest (DHCA). The wide range of temperatures at which patients reach burst suppression (15.7–33.0 °C) and EEG silence (12.5–27.2 °C) suggests that the use of arbitrary DHCA temperatures (e.g. 20 °C) may be inappropriate for many patients (Figure 31.2). Temperature-induced EEG changes are not predicted by age, isoflurane concentration, PaCO₂, or other physiological factors.

Figure 31.2 Intraoperative nasopharyngeal temperatures of 109 patients undergoing DHCA with EEG monitoring. The distribution of temperatures at which burst suppression (A) and EEG silence (B) were achieved.

From Stecker et al., Ann Thorac Surg 2001; 71(1): 14–21.

Individualized temperature management may allow safe DHCA at higher temperatures, reducing the physiological burden of extreme hypothermia and CPB duration.

Depth of Anaesthesia

Although ubiquitous, routine clinical monitoring (e.g. MAP, HR, respiratory pattern, diaphoresis and pupil size) is a crude and imprecise measure of DOA. The use of more reliable monitors of DOA has been driven, at least in part, by the recognition that:

• 
  

  Pharmacological suppression of EEG activity does not mitigate against cerebral ischaemic injury

  
  
• 
  

  Anaesthetic agents may themselves be neurotoxic

  
  
• 
  

  There is a correlation between DOA and mortality

  
  
• 
  

  There may be a link between DOA and postoperative cognitive dysfunction

Using a 16-channel EEG montage to monitor DOA is fraught with problems:

• 
  

  Interpreting the EEG requires training and is time-consuming

  
  
• 
  

  There is no ‘gold standard’ for wakefulness, sedation, anaesthesia and deep anaesthesia

  
  
• 
  

  The EEG is altered by ischaemia, temperature and metabolism – as well as drugs

Currently available DOA monitors process the raw EEG obtained from electrodes placed on the forehead, to produce a time-averaged, proprietary number: typically, a dimensionless integer (e.g. 0 to 100). For the reasons stated above, both ketamine and N₂O may paradoxically increase this index, suggesting an apparent decrease in DOA.