Introduction:

Prior Electroencephalogram (EEG) articles have focused on the pharmacologic effects of ketamine, propofol, and nitrous oxide on intraoperative EEG. This article shifts the focus to nociceptive EEG changes occurring under general anesthesia. These EEG changes include beta arousal, delta arousal, and alpha dropout.¹ Beta arousal refers to increased power in the beta frequency (12-25 Hz) from underlying cortical activity that can be seen in patients with low dose hypnosis (typically 0.5-1 MAC) and no/inadequate opioid administration. Beta arousal is typically paired with somatic movement in an unparalyzed patient.  On the other hand, in balanced techniques or those with higher anesthetic doses, delta arousal [increased delta power (0.4-4 Hz)] and alpha dropout [decreased alpha power (8-12 Hz)] may be appreciated following a painful stimulus. Physiologically, delta arousal may localize to the midbrain reticular formation, while alpha dropout likely represents a return of some degree of thalamocortical signaling.¹ The following two cases demonstrate all three phenomena.

Case 1 (Basic Case):
71 yo M with CAD s/p CABG and complex abdominal surgical history presented for abdominal wall reconstruction.  Despite a continuous intraoperative epidural infusion, classic alpha dropout paired with delta arousal can be seen starting at 1239 and ending at 1248 (Fig. 1).  This was following visceral manipulation/incision and accompanied by an increase in blood pressure and heart rate. 0.2 mg hydromorphone was administered.

Fig. 1. Density spectral array (DSA) representing a classic pain signature between 1239-1248 (red arrows). Dropout of the peak alpha frequency band (just under 10 Hz) and delta arousal can be clearly appreciated.

Alpha dropout with delta arousal can be seen again at 1427.   Table one defines a list of key words and states their significance. The spectral edge frequency (SEF) is defined as the frequency under which a certain percentage of the EEG power is below. In our monitor, the percentage is 95%. In Figure 2, the SEF is 6 Hz, suggesting dominance of delta and theta power, also visible in the raw EEG.  In this case, the surgical stimulation is unclear.  There was no vital sign correlate, change in hypnotic dose or opioid administration.

Fig. 2. Second occurrence of alpha dropout and delta arousal at 1427 (red arrow). Higher frequency power appears at 1430, but this is short-lived.

Csse 2 (Advanced Case):

46 yo patient with hypertension and recurrent presacral mass presented to the operative room for trans coccygeal resection. The hypnotic used for the case was sevoflurane. Following a painful stimulus (bone drilling), there was a small increase in blood pressure and heart rate around 1210 (Fig. 3).

Fig. 3.  Successful intubation with hypertensive response around 1135.  Pain stimulus around 1210 led to elevation in blood pressure and heart rate (red circle).  The inspired sevoflurane dose was increased to 3.4% in response to the change in vital signs.

At 1210, one can see significant fading of the alpha peak at approximately 10 Hz and increased prominence of the delta band on the density spectral array (DSA), consistent with a noxious stimulus (Fig. 4). This event was not treated by opioid, but rather by increasing sevoflurane dose short term (Fig. 3). Interestingly, it appears that the increased sevoflurane dose simultaneously compressed the alpha, theta, delta power bands (subtle downward bend of peak alpha frequency band at 1210).² At 1220, one can see the EEG delta band revert to the baseline.  The alpha power also improved but subjectively did not appear quite as intense thereafter.

The first dose of post-induction opioid (0.2 mg hydromorphone) was between 1230-1245, which did not grossly improve the alpha power (Fig. 5), nor did subsequent small opioid bolus doses.  Around 1226, the SEF increases to about 20 Hz and there is a clear extension of beta power (beta arousal). With a relatively low sevoflurane dose (now < 1 MAC), this finding likely represents a pain signature and undertreatment with opioid (Fig 5). It is noteworthy that ketamine was not administered during this case. The total dose of hydromorphone was 1 mg for the 2.5-hour case.

Fig. 4. The density spectral array (DSA) demonstrates alpha dropout and delta arousal from about 1210 to 1220.  The initial change can be seen in the boxed sections of the spectrogram. Return of some alpha power can be appreciated in the circled areas.

Fig. 5. In this continued tracing from Fig. 4, the density spectral array (DSA) demonstrates modest alpha power. It is worth noting a weakening of the delta band and step up in the spectral edge frequency) SEF to around 20 Hz (white irregular line).  The increase in SEF (red arrows) is likely caused by the combination of delta band fading and increased beta power (beta arousal). This may be attributed to a relatively low dose hypnotic dose (sevoflurane dose < 1 MAC) with inadequate analgesia.

Discussion:

A goal during the maintenance of general anesthesia is to minimize the response to noxious stimuli.  Hemodynamic variables, such as heart rate and blood pressure, have been the traditional targets of analgesic and hypnotic titration. However, the EEG represents a potentially valuable and more direct measure of the adequacy of the changing analgesic state. There are times where obvious alpha dropout and delta arousal occur without vital sign correlates. Indeed, hemodynamic changes may result from numerous pathophysiologic processes including the effect of the hypnotic, making them unreliable as an indicator of pain.

Often times anesthesiologists have to decide whether it is appropriate to modify hypnotic dose or administer an analgesic in response to hemodynamic changes, such as elevated blood pressure. Although both treatments may have a similar beneficial effect on blood pressure, the effect on EEG differs. In Fig. 4, inappropriately increasing hypnotic dose causes a compression of EEG bands but a wash out of alpha power, since pain was not addressed. Once an opioid is administered, some alpha power returns and the compressed EEG bands widens, as a consequence of (appropriately) reducing hypnotic dose. Some other situations may be more unclear. In Fig. 2, a nociceptive signal is seen on EEG in the alpha and delta bands, but there is no correlation to a nociceptive stimulus or change in hemodynamics. The authors would hypothesize that an increase in nociceptive activation, whether visually appreciated or not, may be responsible for this change. Although a fading opioid level may contribute, abrupt changes in EEG would not be expected in that situation. How aggressively to prevent this loss of alpha power with opioid administration is a topic of active investigation and debate.

As discussed, EEG changes signifying nociceptive stimulation include beta arousal, delta arousal, and alpha dropout. These EEG changes were seen in the two case scenarios. Alpha dropout has been associated with surgical incision, longer operations (> 3 hrs) and those involving body cavity surgery.³ The ability to identify EEG changes associated with painful stimulus can guide analgesic titration with opioids and regional anesthesia. However, as the impact of identifying and treating the EEG changes associated with pain (i.e. maximizing alpha power) is not yet understood, there is a need to investigate the effect on postoperative outcomes, including the incidence and severity of acute and chronic postoperative pain. Furthermore, caution in interpreting delta arousal and the influence on EEG indices is warranted. For example, interpreting delta arousal as an indication of excessive hypnosis and erroneously responding with decreasing hypnotic dose could be problematic, leading to a state of inadequate hypnosis and analgesia.¹

When applying described EEG patterns to individualized patients, personalization becomes important.¹ Decreased alpha power may be seen in patients with sleep disorders, prior stroke, neurodegeneration, cognitive impairment, and older age. Pre-incisional alpha power provides valuable insight into a patient’s baseline alpha power. Giattino et al⁴ evaluated 50 adults aged 60 and over revealing a correlation between preoperative cognitive function and decreased frontal alpha power under general anesthesia. There was no correlation between preoperative cognition and other EEG frequencies, including delta, theta, or beta.  Identifying patients with decreased frontal alpha power under general anesthesia may allow prevention and monitoring strategies for delirium and post-operative cognitive decline (POCD).

The preceding cases provide examples for recognition of individual nociceptive EEG changes. We look forward to better understanding the significance of preventing and treating nociceptive EEG changes with future studies.

Table 1. Key Terms and Meanings

Term	Definition	Significance
Density Spectral Array (DSA)	Two-dimensional display of the spectrogram, showing EEG power through color scheme of frequency over time.	Allows useful visualization of the EEG power of frequency over time.2
Spectral Edge Frequency (SEF)	Frequency value which a defined percent (95% in our monitors) of the power is below. White irregular horizontal line in the DSA.	Informative to follow clinical changes in lower or higher power2
Alpha Dropout	Decreased alpha power (8-12 Hz)	Nociceptive EEG change under general anesthesia
Delta Arousal	Increased delta power (0.4-4 Hz)	Nociceptive EEG change under general anesthesia
Beta Arousal	Increased beta power (13-25)	Nociceptive EEG change under general anesthesia, may be seen with lower dose hypnosis and inadequate analgesia

References:

1. García, Paul S et al. “Effects of noxious stimulation on the electroencephalogram during general anaesthesia: a narrative review and approach to analgesic titration.” British Journal of Anaesthesia vol. 126,2 (2021): 445-457.
2. Patrick L. Purdon, Aaron Sampson, Kara J. Pavone, Emery N. Brown; Clinical Electroencephalography for Anesthesiologists: Part I: Background and Basic Signatures. Anesthesiology 2015; 123:937–960
3. Darren F. Hight, Amy L. Gaskell, Matthias Kreuzer, Logan J. Voss, Paul S. García, Jamie W. Sleigh,
   Transient electroencephalographic alpha power loss during maintenance of general anaesthesia, British Journal of Anaesthesia, Vol 122, 5 (2019): 635-642.
4. Giattino, Charles M et al. “Intraoperative Frontal Alpha-Band Power Correlates with Preoperative Neurocognitive Function in Older Adults.” Frontiers in systems neuroscience vol. 11 24. 8 May. 2017, doi:10.3389/fnsys.2017.00024