Intraoperative electroencephalography neuromonitoring is the most feasible and straightforward approach for tracking brain activity during general anesthesia. Berger  has first described systematic EEG signatures during general anesthesia with phenobarbital in 1933, identifying intraoperative EEG pattern comparable with different sleep stages. However, a routine intraoperative EEG neuromonitoring during anesthesia was not feasible until the 1990s, as smaller EEG neuromonitoring devices were developed. There has been consensus that intraoperative EEG, recorded solely on the forehead of the patient and processed by a built-in EEG analysis algorithm, can aid anesthetic dosing and expedite recovery times after general anesthesia [2,3]. The processed EEG indicates depth of anesthesia by providing a single number (index: 100–90 indicates wakefulness, 40–60 refers to optimal depth of anesthesia, 20–0 refers to excessive anesthesia, including burst suppression periods and isoelectric EEG lines) [4–6]. These EEG-based neuromonitors are easy to handle, and the single index number allows for an effortless and immediate interpretation. However, these single number indices have been deemed unreliable in clinical practice, as they do not take into account the choice anesthetic agent, age of the patient, and other factors that can influence EEG readings, such as altered brain metabolism, hypocapnia or hypercapnia, etc. As different anesthetics interact with different molecular targets in the brain [7▪▪], anesthesiologists tend to distrust a generalized, ‘one-size fits all’ index. Furthermore, intraoperative EEG signatures are dependent on age, varying significantly among young children [8–10], healthy adults, and elderly patients [5,11,12].
Despite these limitations, EEG-guided general anesthesia – based on these derived indices – was able to reduce incidence of postoperative delirium and the occurrence of long-term cognitive deficits in elderly patients [13–15]. Based on these data, the European Society of Anaesthesiology recommends in their recently published ‘Guideline on postoperative delirium’ [16▪▪] to guide general anesthesia in elderly patients with EEG neuromonitors, so as to reduce the incidence of postoperative delirium by avoiding periods of excessive anesthesia (Level A).
Within this review, we summarize recent investigations on intraoperative EEG neuromonitoring related to the optimization of EEG-based anesthesia guidance in elderly patients and those at risk for cognitive complications. In addition, we present EEG findings related to postoperative delirium in elderly patients, providing insights to the pathophysiological background of neuronal excitability during delirium periods.
PROCESSED EEG IN ELDERLY
As the population in the Western part of the world ages, the number of elderly patients requiring general anesthesia is increasing rapidly. In elderly patients, postoperative delirium is the most frequent brain dysfunction occurring after general anesthesia, which can transition into a long-term cognitive decline and an increased mortality [17,18]. Therefore, to avoid postoperative complications and its consequences, it is critical to improve the safety of anesthesia for these patients.
The benefits of the use of intraoperative EEG-based neuromonitoring in elderly patients to reduce the incidence of postoperative cerebral dysfunctions, which has been shown in two large, randomized studies. Radtke et al. assessed the occurrence of postoperative delirium in 1155 elderly patients, randomized into groups with and without bispectral index (BIS)-guided anesthesia. Postoperative delirium was detected in 16.7% of the patients in the BIS-guided group, compared with 21.4% in the BIS-blinded group . In addition, a very similar study from Chan et al.  could show that postoperative cognitive decline at 3 months could be reduced in the BIS-guided group (10.2%) compared with the BIS-blinded group (14.7%). A detailed EEG data analysis indicated that prolonged periods with BIS values below 20 – indicative of a high percentage of burst suppression activity in the raw EEG – are the main trigger for the occurrence of postoperative delirium . The correlation between burst suppression activity and the development of postoperative delirium was also observed in a study from Fritz et al. , in which intraoperative duration of burst suppression periods was determined by visual raw EEG data analysis. Swank and Watson  first described burst suppression activity in the EEG in 1949. Its main feature consists of quasiperiodical bursts of bilateral high-voltage slow waves (mainly <15 Hz) separated by low-voltage (<5 μV) or absent activity, lasting from a few seconds to minutes, and indicating a very deep level of sedation (Fig. 1). Burst suppression activity is a characteristic EEG feature of a pathological, profoundly inactivated brain, which may also occur during hypoxia, septic encephalopathy, severe hypothermia, and other pathological brain states. Cerebral metabolism during burst suppression state is maximally reduced . Based on these data, it is possible that recovery from a severely suppressed brain metabolism is hindered in elderly patients, ultimately leading to the presentation of postoperative delirium. It is noteworthy that these findings are in contrast to emergence delirium in children, in which neuronal hyperexcitability – as seen by occurrence of epileptiform discharges during anesthesia induction or increased frontal connectivity during emergence of general anesthesia – has been related to agitated emergence delirium [22,23]. These striking differences in EEG activity related to delirium in elderly patients compared with children extends into its clinical presentation, in which delirious children present with crying, agitation, disorientation, and altered response to their surroundings, whereas elderly patients present primarily with a hypoactive form of delirium .
Intraoperative frontal EEG primarily records the cortical activity of the gyrus frontalis superior (=Fp1/Fp2) and the gyrus frontalis inferior (=F7/F8). EEG recorded potentials relate to the activity of cortical pyramidal neurons, the rhythm corresponds to neuronal interactions between cortical and subcortical brain regions. Neuronal activity is influenced by cerebral metabolism, cerebral blood flow, oxygen saturation, hypocapnia/hypercapnia, glucose metabolism, and the cortical, neuronal activity itself. In addition, external artifacts, such as electric currents, muscle activity, and eyelid movement, can disturb intraoperative EEG recordings. These influencing factors must be considered during interpretation of the intraoperative EEG. Moreover, intraoperative EEG recordings display characteristic signatures related to the anesthetic agent chosen [25▪,26–28], as well as to the age of the patient and their cognitive status [5,8,9,11,12,23].
As the processed EEG indices suppose the same level of unconsciousness independently of the anesthetic agent used, age, and preexisting cognitive status of the patient, they are widely considered to be inherently inaccurate, causing distrust, and reducing the implementation of these systems in clinical routine.
ANESTHETIC AGENT-RELATED CHANGES
Anesthetics alter neuronal activity primarily by either increasing inhibition via Gamma-aminobutyric acid (GABA) GABAA receptor activation or decreasing excitation via N-methyl-D-aspartat receptor blockage, inducing neuronal hyperpolarization [7▪▪]. Due to these different mechanisms, intraoperative EEG signatures differ depending on anesthetic agent used. Below are key characteristics of the intraoperative EEG signatures of the most widely used anesthetic agents – propofol and volatile anesthetics, such as sevoflurane, desflurane.
Propofol, the most widely used anesthetic agent, binds postsynaptically at the ß-subunit of the GABAA receptor. Its binding increases not only the channel opening frequency and binding affinity for GABA, but also directly activates the receptor, even in the absence of endogenic GABA [29–31], hyperpolarizing postsynaptic neurons, thus leading to neuronal inhibition. Propofol-induced EEG signatures present in a characteristic manner related to the level of sedation [25▪,32]. EEG signatures are characterized by spectral analysis of the EEG band power, which is defined as a function of EEG wave amplitudes and frequencies. The distinct EEG frequency bands are beta 13–35 Hz, alpha 8–12 Hz, theta 4–7 Hz, delta 1–3 Hz, and infraslow less than 1 Hz band. Furthermore, EEG coherence, which can be interpreted as a measure of synchrony between two signals of the same frequency band (e.g., alpha band) in different regions of the brain, can be analyzed. During light sedation – just after administration of propofol and shortly after extubation – a period of increased beta oscillation (13–35 Hz) and reduced alpha-band power (8–12 Hz) is visible. This period is called ‘paradoxical excitation’ – as an increased neuronal oscillation is seen, albeit clinically accompanied by sedation and drowsiness of the patient . After reaching ‘loss of consciousness’ through general anesthesia with propofol, infraslow (<1 Hz) and alpha-band power (8–12 Hz) increases (Fig. 2). The alpha-band power presents with a high coherent activity within the frontal cortex, in contrast to the incoherent infraslow oscillations [25▪]. In a mathematical model, this characteristic, highly coherent frontal alpha-band oscillation has been related to a thalamocortical feedback mechanism, which leads to a cortical synchrony, and induces a stable unconscious state [34,35].
Sevoflurane is an anesthetic agent with rapid induction and emergence profile. It binds at multiple targets, inducing potentiation of GABAA, glycine and two-pore potassium channels, and inhibition of voltage-gated potassium channels, N-methyl-D-aspartate, muscarinic, nicotinic acetylcholine, and serotonin channels [7▪▪]. Similar to propofol EEG signatures, sevoflurane induces a frontal, coherent alpha oscillation and infraslow oscillations at frequencies less than 1 Hz, but in contrast to propofol, sevoflurane is also associated with an increase of coherent theta-band activity  (Fig. 3). The theta-band activity may be related to the modulation of T-type calcium channels in the hippocampus and in thalamic relay neurons . The similarities in the alpha band are most likely related to the activation of GABAergic neural circuits, which is also seen in general anesthesia with isoflurane and desflurane , suggesting that these coherent alpha-band oscillations may be used as a general ‘loss of consciousness’ EEG signature. The coherent frontal alpha oscillation pattern limits the dimensionality of the thalamocortical network, reducing the ability of the thalamus to project and coordinate exogenous inputs to the neocortex .
AGE-RELATED EEG CHANGES
However, the identification of a coherent frontal alpha power may be difficult in elderly patients. Purdon et al.  examined 155 patients, aged 18–90 years of age, while receiving either propofol or sevoflurane anesthesia. With increasing age, they found a marked reduction in EEG signal power for all frequency bands, whereas this effect was most pronounced in the alpha-frequency band. In addition, when compared with younger patients, a loss of alpha-band coherence was observed in the elderly. They proposed that the age-related EEG power reduction might be caused by a decline in synaptic density and changes in dendritic dynamics, so that these changes might reflect functional alterations in the GABA-dependent frontal thalamocortical circuits.
Furthermore, Purdon et al.  could show that, compared with younger patients, elderly patients are more likely to experience burst suppression periods during general anesthesia. This finding is important, as duration of burst suppression during anesthesia correlates with the development of postoperative delirium in elderly patients [14,19].
Currently, all available EEG neuromonitors offer a ‘Density spectral array’ (Figs. 2 and 3), which enables anesthesiologist to guide general anesthesia in patients by observing the raw EEG activity and the frontal alpha band induced by propofol and/or volatile anesthetics. To increase the visibility of the frontal coherent alpha band in elderly patients during general anesthesia, it may be helpful to adjust the power sensitivity on EEG neuromonitors (e.g., from 15 to 30 μV) (Fig. 4). We recommend this option during general anesthesia in elderly patients, as it offers a more reliable way to identify a stable, unconscious state and avoid an excessive depth of anesthesia. Burst suppression activity may be easily seen in the raw EEG (Fig. 1), which is available in all EEG neuromonitors, clearly indicating states of anesthesia that are too deep.
EEG SPECTRAL ANALYSIS AND POSTOPERATIVE DELIRIUM
Alpha-band activity in awake patients is known to be reduced in cognitively impaired patients suffering from Alzheimer disease, as well as in those with mild cognitive impairment . As alpha-band activity is also related to attention and level of sedation and unconsciousness [25▪,38], alpha-band activity appears to be an important marker of cognitive function. Alteration in alpha-band power has also been observed in patients during postoperative delirium, in which patients presented a significantly reduced relative alpha-band power in parietal and occipital regions during eye-open EEG recordings . Most recently, it has been shown that intraoperative frontal alpha-band activity correlates with reduced cognitive function in patients more than 65 years . Intraoperative, frontal and parietal alpha-band power was significantly reduced in patients with lower overall cognitive abilities during general anesthesia with propofol or isoflurane. As intraoperative, frontal alpha-band activity is a signature of thalamic GABAA activation inducing a thalamocortical feedback mechanism, this suggests a critical role of cortical inhibition by GABAA activation in cognitive function in general.
EEG CONNECTIVITY AND POSTOPERATIVE DELIRIUM
It has been discussed for several years that neural synchronization is crucial for cognition, providing a means to analyze coactivation and functional connectivity. EEG connectivity refers to the synchrony and temporal correlation between two remote neurophysiological signals . Directed connectivity could provide additional insights into the direction of information flow, as global organization of functional brain networks can be characterized using graph theory [42▪].
Consciousness vanishes when anesthetics induce functional disconnectivity during general anesthesia, causing loss of corticocortical information flow . In elderly patients presenting with hypoactive postoperative delirium, a reduction in the alpha-band connectivity was found, similar to the disconnectivity seen in the EEG during regain of consciousness after general anesthesia [42▪]. In addition, delirious patients present with a less integrated network organization, which might reflect the inability of the delirious patient to regain organized cortical function promptly.
In conclusion, we recommend the consistent use of frontal EEG neuromonitoring during general anesthesia in elderly patients to prevent excessive depth of anesthesia and reduce the development of postoperative delirium. This approach is helpful, even when anesthesiologists guide anesthesia solely by the EEG-derived index. However, we advocate using the spectral density analysis, which is available in all commercially available EEG neuromonitors, to guide propofol and/or ether derived volatile anesthesia by focusing on the frontal alpha-band power. This easily visible alpha power band is a clear indicator of intraoperative unconsciousness. In elderly patients, it might be necessary to increase the sensitivity of EEG power presentation to better visualize the frontal alpha power band.
The authors thank Rudolf Mörgeli and Victoria Windmann for support in writing the article.
Financial support and sponsorship
Support was provided solely from institutional and/or departmental sources.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Berger H. Über das Elektrenkephalogramm des Menschen. Arch für Psychiatr Nervenkrankh 1933; 98:231–254.
2. Gan TJ, Glass PS, Windsor A, et al. Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanil, and nitrous oxide anesthesia
. BIS Utility Study Group. Anesthesiology 1997; 87:808–815.
3. Monk TG. Processed EEG and patient outcome. Best Pract Res Clin Anaesthesiol 2006; 20:221–228.
4. Glass PS, Bloom M, Kearse L, et al. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 1997; 86:836–847.
5. Schultz B, Schultz A, Grouven U, et al. Changes with age in EEG during anesthesia
. Anaesthesist 1995; 44:467–472.
6. Revuelta M, Paniagua P, Campos JM, et al. Validation of the index of consciousness during sevoflurane and remifentanil anaesthesia: a comparison with the bispectral index and the cerebral state index. Br J Anaesth 2008; 101:653–658.
7▪▪. Alkire MT, Hudetz AG, Tononi G. Consciousness and anesthesia
. Science 2008; 322:876–880.
The review gives a great overview on the different mechanisms and sites of action of different anesthetic agents and highlights the idea, that anesthetic-induced unconsciousness is produced by functional disconnectivity within corticothalamic regions reducing information flow.
8. Davidson AJ. Measuring anesthesia
in children using the EEG. Paediatr Anaesth 2006; 16:374–387.
9. Akeju O, Pavone KJ, Thum JA, et al. Age-dependency of sevoflurane-induced electroencephalogram
dynamics in children. Br J Anaesth 2015; 115 (Suppl 1):i66–i76.
10. Koch S, Stegherr AM, Morgeli R, et al. Electroencephalogram
dynamics in children during different levels of anaesthetic depth. Clin Neurophysiol 2017; 128:2014–2021.
11. Schultz A, Grouven U, Zander I, et al. Age-related effects in the EEG during propofol anaesthesia. Acta Anaesthesiol Scand 2004; 48:27–34.
12. Purdon PL, Pavone KJ, Akeju O, et al. The ageing brain: age-dependent changes in the electroencephalogram
during propofol and sevoflurane general anaesthesia. Br J Anaesth 2015; 115 (Suppl 1):i46–i57.
13. Chan MT, Cheng BC, Lee TM, et al. BIS-guided anesthesia
decreases postoperative delirium
and cognitive decline. J Neurosurg Anesthesiol 2013; 25:33–42.
14. Radtke FM, Franck M, Lendner J, et al. Monitoring depth of anaesthesia in a randomized trial decreases the rate of postoperative delirium
but not postoperative cognitive dysfunction. Br J Anaesth 2013; 110 (Suppl 1):i98–i105.
15. Whitlock EL, Torres BA, Lin N, et al. Postoperative delirium
in a substudy of cardiothoracic surgical patients in the BAG-RECALL clinical trial. Anesth Analg 2014; 118:809–817.
16▪▪. Aldecoa C, Bettelli G, Bilotta F, et al. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium
. Eur J Anaesthesiol 2017; 34:192–214.
A guideline (European Society of Anaesthesiology/Guideline committee postoperative delirium) to present recommendations for the prevention and treatment of postoperative delirium, pointing out preoperative identification and handling of patients at risk, adequate intraoperative care and postoperative detection and management of delirious patients.
17. Witlox J, Eurelings LS, de Jonghe JF, et al. Delirium in elderly patients
and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA 2010; 304:443–451.
18. Saczynski JS, Marcantonio ER, Quach L, et al. Cognitive trajectories after postoperative delirium
. N Engl J Med 2012; 367:30–39.
19. Fritz BA, Kalarickal PL, Maybrier HR, et al. Intraoperative electroencephalogram
suppression predicts postoperative delirium
. Anesth Analg 2016; 122:234–242.
20. Swank RL, Watson CW. Effects of barbiturates and ether on spontaneous electrical activity of dog brain. J Neurophysiol 1949; 12:137–160.
21. Brown EN, Lydic R, Schiff ND. General anesthesia
, sleep, and coma. N Engl J Med 2010; 363:2638–2650.
22. Martin JC, Liley DT, Harvey AS, et al. Alterations in the functional connectivity of frontal lobe networks preceding emergence delirium in children. Anesthesiology 2014; 121:740–752.
23. Koch S, Rupp L, Prager C, et al. Emergence delirium in children is related to epileptiform discharges during anaesthesia induction: an observational study. Eur J Anaesthesiol 2018; 35:929–936.
24. Robinson TN, Raeburn CD, Tran ZV, et al. Motor subtypes of postoperative delirium
in older adults. Arch Surg 2011; 146:295–300.
25▪. Purdon PL, Pierce ET, Mukamel EA, et al. Electroencephalogram
signatures of loss and recovery of consciousness from propofol. Proc Natl Acad Sci U S A 2013; 110:E1142–E1151.
The study is a very good presentation on the intraoperative EEG spectrogram presentation, in which loss of consciousness presents coherent frontal alpha oscillations.
26. Akeju O, Westover MB, Pavone KJ, et al. Effects of sevoflurane and propofol on frontal electroencephalogram
power and coherence. Anesthesiology 2014; 121:990–998.
27. Akeju O, Pavone KJ, Westover MB, et al. A comparison of propofol- and dexmedetomidine-induced electroencephalogram
dynamics using spectral and coherence analysis. Anesthesiology 2014; 121:978–989.
28. Akeju O, Song AH, Hamilos AE, et al. Electroencephalogram
signatures of ketamine anesthesia
-induced unconsciousness. Clin Neurophysiol 2016; 127:2414–2422.
29. Hales TG, Lambert JJ. The actions of propofol on inhibitory amino acid receptors of bovine adrenomedullary chromaffin cells and rodent central neurones. Br J Pharmacol 1991; 104:619–628.
30. Hara M, Kai Y, Ikemoto Y. Propofol activates GABAA receptor-chloride ionophore complex in dissociated hippocampal pyramidal neurons of the rat. Anesthesiology 1993; 79:781–788.
31. Jurd R, Arras M, Lambert S, et al. General anesthetic actions in vivo strongly attenuated by a point mutation in the GABA(A) receptor beta3 subunit. FASEB J 2003; 17:250–252.
32. Gugino LD, Chabot RJ, Prichep LS, et al. Quantitative EEG changes associated with loss and return of consciousness in healthy adult volunteers anaesthetized with propofol or sevoflurane. Br J Anaesth 2001; 87:421–428.
33. McCarthy MM, Brown EN, Kopell N. Potential network mechanisms mediating electroencephalographic beta rhythm changes during propofol-induced paradoxical excitation. J Neurosci 2008; 28:13488–13504.
34. Ching S, Cimenser A, Purdon PL, et al. Thalamocortical model for a propofol-induced alpha-rhythm associated with loss of consciousness. Proc Natl Acad Sci U S A 2010; 107:22665–22670.
35. Soplata AE, McCarthy MM, Sherfey J, et al. Thalamocortical control of propofol phase-amplitude coupling. PLoS Comput Biol 2017; 13:e1005879.
36. McDowell TS, Pancrazio JJ, Barrett PQ, Lynch CJ 3rd. Volatile anesthetic sensitivity of T-type calcium currents in various cell types. Anesth Analg 1999; 88:168–173.
37. van der Hiele K, Vein AA, van der Welle A, et al. EEG and MRI correlates of mild cognitive impairment and Alzheimer's disease. Neurobiol Aging 2007; 28:1322–1329.
38. Pfurtscheller G, Kalcher J, Neuper C, et al. On-line EEG classification during externally-paced hand movements using a neural network-based classifier. Electroencephalogr Clin Neurophysiol 1996; 99:416–425.
39. van der Kooi AW, Zaal IJ, Klijn FA, et al. Delirium detection using EEG: what and how to measure. Chest 2015; 147:94–101.
40. Giattino CM, Gardner JE, Sbahi FM, et al. Intraoperative frontal alpha-band power correlates with preoperative neurocognitive function in older adults. Front Syst Neurosci 2017; 11:24.
41. O’Neill GC, Tewarie P, Vidaurre D, et al. Dynamics of large-scale electrophysiological networks: a technical review. Neuroimage 2018; 180 (Pt B):559–576.
42▪. Numan T, Slooter AJC, van der Kooi AW, et al. Functional connectivity and network analysis during hypoactive delirium and recovery from anesthesia
. Clin Neurophysiol 2017; 128:914–924.
The study first to present, that loss of functional and directed connectivity are similarly disturbed during reduced level of consciousness during recovery from anesthesia and postoperative, hypoactive delirium in elderly patients.
43. Lee H, Noh GJ, Joo P, et al. Diversity of functional connectivity patterns is reduced in propofol-induced unconsciousness. Hum Brain Mapp 2017; 38:4980–4995.