Canadian Journal of Anesthesia/Journal canadien d’anesthésie · February 16, 2018
Author: Michael S. Avidan
We have less information regarding the functioning of the brain than of the other vital organs during general anesthesia. This is sobering considering that the brain is the most vital of vital organs and is also the target organ of general anesthesia. Anesthesiologists have developed sophisticated monitors and understanding of human physiology, especially as it pertains to the cardiovascular and respiratory systems. Perturbations in these systems, however, occur as “collateral damage” with the administration of general anesthesia. In contrast, the profound alterations occurring in brain function with general anesthesia are therapeutically indicated. With desirable general anesthesia, there are neurological changes in attention, thinking, nociception, memory formation, and consciousness – with a state of oblivion produced that is more akin to coma than sleep.1 It would be useful if we were able to monitor these various neurological changes accurately.
In an attempt to address this major deficit in anesthesiology practice, several “brain function monitors” have been developed. Generally, these monitors obtain electroencephalographic (EEG) information from one or two frontal EEG channels, process the information with proprietary algorithms, and display a number intended to reflect the hypnotic depth of anesthesia. In contrast to many other physiological monitors,2 these brain monitors have yet to be adopted as standard of care.
In this issue of the Journal, Epstein et al. present the findings of a large, impressive observational study examining whether there are periods during which patients are likely to be appropriately anesthetized based on sevoflurane administered at > 0.75 minimum alveolar concentration despite a processed EEG index that displays a value suggesting insufficient anesthetic depth (i.e., a bispectral index or state entropy value > 70).3 They label this apparent discrepancy as a form of “discordant elevation” of the index (Table, pattern d). The primary hypothesis of the Epstein et al. study was that this particular manifestation of discordance was more commonly displayed with the entropy monitor (GE Healthcare, Helsinki, Finland) than with the bispectral index monitor (Medtronic, Minneapolis, MN, USA). The endpoint of the study was discordant elevations lasting more than two minutes. The investigators speculated that such discordance would result in the unnecessarily excessive administration of anesthetic agents with negative clinical consequences to patients, including slow recovery from general anesthesia and respiratory depression. In agreement with the investigators’ hypothesis, this pattern of discordance occurred in 3.6% of cases in which an entropy monitor was used but in only 0.24% of cases in which a bispectral index monitor was used.3 No data were presented, however, that supported the conclusion that use of the entropy monitor was associated with an increased incidence of clinically relevant negative consequences (e.g., hypotension, slow recovery from anesthesia, postoperative nausea and vomiting).
PONV, postoperative nausea and vomiting; PTSD, post-traumatic stress syndrome. This table shows six broad patterns of discordance (a to f) for processed electroencephalographic (EEG) indices and actual anesthetic depth. The most typical side effect of excessive anesthetic administration is hypotension, which is typically treated by anesthesiologists. Slow recovery from general anesthesia can occur, but it is not usually a major problem with modern anesthetic agents. PONV is a serious and proven complication of excessive volatile anesthetic administration. Postoperative delirium is a theoretical complication of deep general anesthesia. The most serious consequences of discordance might occur when patients are “light,” yet the processed EEG index shows a value consistent with (“appropriate” or “deep”) general anesthesia. Specifically, this could result in harmful patient movement, distressing intraoperative awareness, and even PTSD25
Based on the results of this study, one might conclude that the entropy monitor is a less reliable indicator of anesthetic depth than the bispectral index monitor. Such a conclusion, however, would be overly simplistic and could potentially be wrong. A much more important question than frequency of discordance is whether the discordance has negative consequences for patients. For example, most anesthesiologists would probably not regard it as an important failure if slightly more anesthetic agent than necessary was administered for a couple of minutes in one in every 25 patients. Rather, what they would doubtless regard as a catastrophic failure would be an index that did not reliably warn them in a timely manner that a pharmacologically paralyzed patient was potentially awake. This alert should always occur in situations where patients are awake and unable to move, regardless of the patient’s age and frailty, the anesthetic cocktail administered, or if there was painful stimulation. Indeed, this concern is now more germane than ever because, as Epstein et al. show, the use of neuromuscular blocking agents in modern anesthesiology practice is pervasive.3 Furthermore, indiscriminate and profound drug-induced paralysis during surgery is likely to increase with the recent introduction of sugammadex, a true pharmacological reversal agent for paralyzing agents. In a bold study in which volunteer anesthesiologists received a neuromuscular blocking drug (succinylcholine or rocuronium) without a hypnotic anesthetic agent, Schuller et al. showed that the bispectral index monitor almost always displayed values < 70 and sometimes even dipped into the 40s.4 Such disastrous discordance (pattern f in the Table) appropriately undermines clinicians’ confidence in the bispectral index monitor. In the situation where anesthesiologists most need our “coal-mine canary” to raise a shrill alarm, she instead “flatters to deceive” with a soporific song.
Another clinically important discordance might be failure of an index to reflect excessively deep anesthesia (Table, patterns a and b). There is evidence suggesting that EEG suppression, possibly indicating deep anesthesia, might be associated with adverse outcomes, such as postoperative delirium and even death.5–8 In view of this theoretical concern, there are several ongoing major clinical trials seeking to answer the question of whether deep anesthesia is injurious.9,10 Current indices could be deficient in their ability to reflect deep anesthesia accurately. (i) They might be imprecise in their detection of EEG suppression,11 and (ii) they often show values only suggestive of deep anesthesia when the percentage time of EEG suppression passes a particular arbitrary threshold.12
Unfortunately, the currently available processed EEG monitors are demonstrably discordant in a variety of circumstances, which has likely curtailed their adoption among discerning anesthesiologists.13 Their algorithms have generally been developed in cohorts of healthy, young adults in the absence of neuromuscular blocking agents. We need them to perform accurately in vulnerable patients – i.e., those who are older with many co-morbidities, those who are prone to severe side effects with general anesthesia, and those who require pharmacological paralysis for invasive surgery. The important deficiencies of current processed EEG monitors can be summarized as follows:
- There is typically a delay of one to two minutes between a change in a patient’s state and the index.14,15
- The depth of anesthesia is unlikely to be equivalent when using two anesthetic regimens (e.g., sevoflurane inhalation versus propofol intravenous) just because the index shows the same value.
- Given that EEG features and changes with general anesthesia are very different depending on age, it is improbable that a particular index value in a 25-yr-old patient has the same neurophysiological meaning as it does in a 90-yr-old patient.16
- Electrical activity outside the brain (e.g., due to eye movement, heart muscle, skeletal muscle, and electrical device) often contaminates the EEG and deceives processed indices.17
The last deficiency in this list is often responsible for the finding reported by Epstein et al. – that processed EEG indices frequently indicate insufficient anesthetic depth even though the anesthetic depth is likely to be appropriate.3
The major discordances documented with processed EEG indices could promote disillusionment with, and abandonment of, monitoring the target organ of the general anesthesia. This would be an unfortunate outcome. As Epstein et al. recommend, for now we should encourage anesthesiologists and other anesthesia clinicians to view and interpret the raw EEG during general anesthesia.3 To this end, there are useful online educational resources18,19 and informative journal articles.20,21 As we learn more about general anesthesia, from the molecular to the systems neuroscience level,22–24 we will be able to design a new generation of brain monitors with ever-stronger scientific foundations that will accord more accurately with different neurological states and depths of anesthesia.
Brown EN, Lydic R, Schiff ND. General anesthesia, sleep, and coma. N Engl J Med 2010; 363: 2638-50.
Dobson G, Chong M, Chow L, et al. Guidelines to the practice of anesthesia – revised edition 2017. Can J Anesth 2017; 64: 65-91.
Epstein RH, Maga JM, Mahla ME, Schwenk ES, Bloom MJ. Prevalence of discordant elevations of state entropy and bispectral index in patients at amnestic sevoflurane concentrations: a historical cohort study. Can J Anesth 2018; 65
Schuller PJ, Newell S, Strickland PA, Barry JJ. Response of bispectral index to neuromuscular block in awake volunteers. Br J Anaesth 2015; 115(Suppl 1): i95-103.
Fritz BA, Kalarickal PL, Maybrier HR, et al. Intraoperative Electroencephalogram Suppression Predicts Postoperative Delirium. Anesth Analg 2016; 122: 234-42.
Willingham M, Ben Abdallah A, Gradwohl S, et al. Association between intraoperative electroencephalographic suppression and postoperative mortality. Br J Anaesth 2014; 113: 1001-8.
Leslie K, Short TG. Anesthetic depth and long-term survival: an update. Can J Anesth 2016; 63: 233-40.
Zorrilla-Vaca A, Healy RJ, Wu CL, Grant MC. Relation between bispectral index measurements of anesthetic depth and postoperative mortality: a meta-analysis of observational studies. Can J Anesth 2017; 64: 597-607.
Short TG, Leslie K, Chan MT, Campbell D, Frampton C, Myles P. Rationale and design of the balanced anesthesia study: a prospective randomized clinical trial of two levels of anesthetic depth on patient outcome after major surgery. Anesth Analg 2015; 121: 357-65.
Wildes TS, Winter AC, Maybrier HR, et al. Protocol for the Electroencephalography Guidance of Anesthesia to Alleviate Geriatric Syndromes (ENGAGES) study: a pragmatic, randomised clinical trial. BMJ Open 2016; 6: e011505.
Muhlhofer WG, Zak R, Kamal T, et al. Burst-suppression ratio underestimates absolute duration of electroencephalogram suppression compared with visual analysis of intraoperative electroencephalogram. Br J Anaesth 2017; 118: 755-61.
Bruhn J, Bouillon TW, Shafer SL. Bispectral index (BIS) and burst suppression: revealing a part of the BIS algorithm. J Clin Monit Comput 2000; 16: 593-6.
Palanca BJ, Mashour GA, Avidan MS. Processed electroencephalogram in depth of anesthesia monitoring. Curr Opin Anaesthesiol 2009; 22: 553-9.
Zanner R, Pilge S, Kochs EF, Kreuzer M, Schneider G. Time delay of electroencephalogram index calculation: analysis of cerebral state, bispectral, and Narcotrend indices using perioperatively recorded electroencephalographic signals. Br J Anaesth 2009; 103: 394-9.
Kreuzer M, Zanner R, Pilge S, Paprotny S, Kochs EF, Schneider G. Time delay of monitors of the hypnotic component of anesthesia: analysis of state entropy and index of consciousness. Anesth Analg 2012; 115: 315-9.
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): i46-57.
Vilge S, Kreuzer M, Karatchiviev V, Kochs EF, Malcharek M, Schneider G. Differences between state entropy and bispectral index during analysis of identical electroencephalogram signals: a comparison with two randomised anaesthetic techniques. Eur J Anaesthesiol 2015; 32: 354-65.
International Consortium for EEG Training of Anesthesia Practitioners. ICE-TAP.
Partners HealthCare Office of Continuing Professional Development & Implementation Science. Clinical Electroencephalography for the Anesthesiologist.
Purdon PL, Sampson A, Pavone KJ, Brown EN. Clinical electroencephalography for anesthesiologists: part i: background and basic signatures. Anesthesiology 2015; 123: 937-60.
Bennett C, Voss LJ, Barnard JP, Sleigh JW. Practical use of the raw electroencephalogram waveform during general anesthesia: the art and science. Anesth Analg 2009; 109: 539-50.
Franks NP, Zecharia AY. Sleep and general anesthesia. Can J Anesth 2011; 58: 139-48.
Brown EN, Purdon PL, Van Dort CJ. General anesthesia and altered states of arousal: a systems neuroscience analysis. Annu Rev Neurosci 2011; 34: 601-28.
Flores FJ, Hartnack KE, Fath AB, et al. Thalamocortical synchronization during induction and emergence from propofol-induced unconsciousness. Proc Natl Acad Sci U S A 2017; 114: E6660-8.
Whitlock EL, Rodebaugh TL, Hassett AL. Psychological sequelae of surgery in a prospective cohort of patients from three intraoperative awareness prevention trials. Anesth Analg 2015; 120: 87-95.