From index to insight: clinical perspectives on electroencephalographic spectrogram-guided anesthesia—a narrative review

Akira Mukai; Jen-Ting Yang; Shao-Chun Wu; Tzu-Chun Wang; Feng-Sheng Lin; Chun-Yu Wu

doi:10.4097/kja.251022

Korean J Anesthesiol > Epub ahead of print

Mukai, Yang, Wu, Wang, Lin, and Wu: From index to insight: clinical perspectives on electroencephalographic spectrogram-guided anesthesia—a narrative review

Narrative Review

Korean Journal of Anesthesiology 2026;kja.251022.

Published online: February 3, 2026

DOI: https://doi.org/10.4097/kja.251022

From index to insight: clinical perspectives on electroencephalographic spectrogram-guided anesthesia—a narrative review

Akira Mukai¹

, Jen-Ting Yang¹

, Shao-Chun Wu²

, Tzu-Chun Wang³

, Feng-Sheng Lin⁴

, Chun-Yu Wu^5,⁶

¹Department of Anesthesiology and Pain Medicine, University of Washington, Seattle, WA, USA

²Department of Anesthesiology, Kaohsiung Chang Gung Memorial Hospital, Kaohsiung, Taiwan

³Department of Anesthesia, MacKay Memorial Hospital, Taipei, Taiwan

⁴Department of Anesthesiology, National Taiwan University Hospital, Taipei, Taiwan

⁵Department of Anesthesiology, National Taiwan University Hospital Hsinchu Branch, Hsinchu, Taiwan

⁶College of Medicine, National Taiwan University, Taipei, Taiwan

Corresponding author: Chun-Yu Wu, M.D., Ph.D.

Department of Anesthesiology, National Taiwan University Hospital Hsinchu Branch and National Taiwan University College of Medicine, Taipei, Taiwan

Tel: +886-7-7317123 (ext. 2788)

Fax: +886-2-23415736

Email: wuchunyu@ntu.edu.tw

Received November 12, 2025 Revised January 1, 2026 Accepted January 4, 2026

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Processed electroencephalogram (EEG) indices, such as the Bispectral Index, have markedly influenced anesthesia practice as they translate brain activity into simple numerical indices. Nevertheless, as the manufacturing algorithms are not disclosed, the underlying neurophysiology remains obscured. Additionally, these indices are often affected by electromyographic contamination, pharmacological variability, and patient-specific EEG heterogeneity. In contrast, an EEG spectrogram, or density spectral array, preserves the frequency- and time-resolved structures of cortical oscillations. This information is presented in a form that is both physiologically meaningful and clinically interpretable. In this review, we trace the evolution of anesthesia from an index-based to a spectrogram-guided approach, and summarize the clinical rationale for adopting the latter. Key applications of this approach include the use of frontal alpha power as a biomarker of cortical stability and postoperative brain health, the identification of nociceptive arousal through alpha dropout and beta or delta arousal patterns, and individualized titration of multimodal or age-specific anesthetic management. Although current devices lack standardized quantitative alpha metrics and have limited sensitivity for low-frequency brain wave components, structured EEG education programs have proven to be effective in terms of fostering spectrogram literacy among anesthesiologists. By combining neurophysiological precision with bedside practicality, the EEG spectrogram represents a pivotal advance toward individualized, mechanism-based, and brain-protective anesthesia, transforming anesthetic monitoring from mere algorithmic abstraction to cortical insight.

Keywords: Alpha Rhythm; Brain; Delirium; Electroencephalography; General anesthesia; Intraoperative monitoring.

Introduction

In the 1990s, the introduction of electroencephalogram (EEG) monitors that can provide processed EEG indices marked a major conceptual advance as these numerical indices derived from cerebral electrical activity can be used to guide anesthetic titration [1]. These indices transform complex EEG signals into a single dimensionless value, which simplifies intraoperative decision-making. Consequently, brain monitoring has become widely adopted in anesthesia practice. Nevertheless, after three decades of clinical use, this reductionist approach has demonstrated inherent shortcomings: the “black-box” algorithms involved conceal their underlying physiological mechanisms, and the derived values are often influenced by multiple perioperative confounders.

Recent advances in computational visualization now allow the real-time display of the EEG spectrogram, which is interpretable at the bedside while preserving the frequency-rich information of cortical dynamics [2]. These tools bridge the gap between raw EEGs and processed EEG indices, combining neurophysiological details and clinical usability.

In this review, we sought to outline the clinical rationale for adopting EEG spectrogram monitoring over processed index-based approaches in contemporary anesthesia practice. Throughout this article, the term “EEG spectrogram” is used as a generic descriptor of time–frequency EEG representations, whereas “density spectral array (DSA)” is used to refer to the visualization tool provided by contemporary anesthesia monitors. Building on prior reviews of anesthetic-specific spectrogram patterns [2–6], this review focuses on the added clinical value and current limitations of DSA, with examples primarily derived from BIS-based recordings.

The development of processed EEG indices

In the field of anesthesiology, cerebral indicators that reflect anesthetic depth more directly than conventional physiological parameters, such as heart rate, blood pressure, or end-tidal anesthetic concentration, are desirable [7]. This reflects the understanding that the brain is the primary target organ of anesthesia, and that EEG provides the most direct measure of the hypnotic component of the anesthetized state. By the 1990s, advances in signal processing had enabled transformation of complex EEG activity into simplified numerical indices of anesthetic depth [1]. While real-time interpretation of raw EEGs requires specialized expertise, these processed EEG indices were designed to provide intuitive and readily interpretable information that is suitable for routine clinical use [3]. Their introduction substantially expanded the adoption of brain monitoring during anesthesia and facilitated anesthetic titration at the individual level.

The Bispectral Index (BIS) monitoring system (BIS^TM; Aspect Medical Systems, now Medtronic), which was introduced in 1996, was among the first systems to use processed EEG indices intraoperatively [8]. The BIS condenses the frontal EEG into an index ranging from 0 to 100, with values of 40–60 indicating surgical hypnosis [9]. Several alternative systems have subsequently emerged, including the Entropy^TM, Narcotrend^TM, SNAP II^TM, CONOX^TM, NeuroSENSE^TM, and the Patient State Index (PSI) derived from SedLine^®, all of which translate multichannel EEG features into numerical indices calibrated so as to reflect anesthetic depth [10–15]. Among these, the BIS remains the most widely used processed EEG index worldwide, with its latest hardware iteration, BIS Advance^TM, released in 2024.

Applications and efficacy of processed EEG monitoring

Processed EEG monitors were introduced to prevent unintended intraoperative awareness among patients by maintaining the hypnotic depth within a defined target range. Despite their widespread clinical adoption, their efficacy in preventing intraoperative complications remains controversial [16]. In 2004, Myles et al. [17] reported that BIS-guided anesthesia reduced the risk of intraoperative awareness by 82% in at-risk adults undergoing general anesthesia. However, subsequent large-scale trials failed to confirm this benefit. Avidan et al. [18] found no significant differences in the incidence of intraoperative awareness between BIS-guided anesthesia and anesthesia protocols based on end-tidal anesthetic gas concentrations. Similarly, Mashour et al. [19] observed no overall difference between BIS-based and anesthetic concentration-based monitoring strategies, although a post hoc analysis revealed a lower incidence of awareness with BIS-guided management than with routine care. These discrepant results raise questions regarding the generalizability of processed EEG index-guided protocols across diverse patient populations and anesthetic techniques. Nevertheless, current evidence suggests that processed EEG monitoring may be particularly useful during total intravenous anesthesia when end-tidal anesthetic concentrations are unavailable [20].

Evolving roles and limitations of processed EEG based monitoring

In recent years, research on processed EEG monitoring has focused on personalized anesthesia management, aiming to optimize anesthetic dosing to improve postoperative neurocognitive outcomes and reduce postoperative delirium (POD) [21–23]. Early evidence has suggested that this approach holds benefits, with Chan et al. [24] reporting reduced POD and cognitive dysfunction with the use of BIS-guided anesthesia in older adults. However, larger randomized trials have yielded mixed results. The ENGAGES trial (2019) found no reduction in POD with this approach, despite lower anesthetic exposure and reduced EEG suppression. These findings may be due to limited dose separation, prolonged EEG suppression in both the test and control groups, and the use of once-daily delirium assessments, which may have underestimated POD incidence [22,25]. Similar neutral findings were reported in the ENGAGES–Canada trial among high-risk cardiac surgery patients [26]. In contrast, a BALANCED trial sub-study showed that targeting lighter anesthesia (BIS 50 vs. 30) reduced POD as well as long-term cognitive impairment, although the main trial found no differences in major complications or mortality [23,27]. A recent meta-analysis suggested that using processed EEG-based anesthesia is associated with a lower incidence of POD, but found substantial heterogeneity [28]. Importantly, recent evidence has indicated that specific EEG features, particularly alpha power and alpha burst suppression, are more strongly linked to POD risk than are the processed indices themselves [29]. This underscores the value of raw EEGs or interpretable EEG parameters over sole reliance on processed EEG indices.

Limitations of processed EEG indices

Although processed EEG monitoring systems have been widely adopted in anesthetic practice, they have substantial and multifaceted limitations, which warrant careful consideration [30]. First, processed EEG indices are derived by manufacturer algorithms that are not disclosed, hampering interpretation of the discrepancies between the index values and the patient’s clinical state. Schuller et al. [31] demonstrated that awake volunteers who were administered neuromuscular blockers without anesthesia exhibited a decrease in BIS values, indicating that the BIS partly depends on electromyographic (EMG) activity for estimating wakefulness. Similar findings have been reported with the GE Entropy^TM [32] and Conox^TM monitoring systems [33]. Another study showed that the reversal of neuromuscular blockade with sugammadex or neostigmine increased BIS values [34]. This is because EMG signals, which are typically within the 30–50 Hz range, may be misinterpreted by the BIS algorithm as cortical activity, thereby inflating BIS values even in the context of deep sedation. Thus, in patients receiving neuromuscular blocking agents, processed EEG indices must be interpreted carefully.

Secondly, although some reports have suggested that processed EEG indices may help detect painful stimuli in sedated or critically ill patients [35], the current indices lack specific EEG features that reliably reflect nociception. To address this limitation, use of alternative nociception monitoring tools, such as the Analgesia Nociception Index and the Nociception Level Index, during surgery have been proposed [36].

Third, processed EEG indices lack pharmacological specificity for certain anesthetic agents. Ketamine, for instance, increases high-frequency gamma activity and reduces alpha power, mimicking an awake EEG pattern, and causing misleadingly high BIS values even when anesthesia is adequate [37]. In contrast, dexmedetomidine induces EEG patterns resembling natural non–rapid eye movement sleep, with prominent slow-delta and spindle oscillations, but yields lower BIS values than propofol at equivalent levels of sedation [38].

Fourth, the accuracy of processed EEG monitoring decreases in patients at the extremes of the age spectrum, those with neurological comorbidities, or those with atypical baseline EEG patterns. For example, older adults may paradoxically exhibit higher BIS values despite receiving higher age-adjusted end-tidal anesthetic concentrations [39]. The PSI has also been shown to increase unreliably with age [40].

Finally, the processed EEG indices often differ across devices, leading to inconsistent outputs even when analyzing identical EEG signals. Hight et al. [41] compared five commercial depth-of-anesthesia monitors (BIS, Entropy-SE, Narcotrend, qCON, and SedLine) and found that their clinical recommendations agreed in only one-third of cases. In another third, at least one monitor indicated excessive hypnotic depth, whereas spectral EEG analysis suggested a lighter plane of anesthesia. Such discrepancies undermine the reliability and cross-comparability of the processed EEG indices among devices.

EEG spectrograms offer several advantages that can help overcome the abovementioned limitations. They demonstrate greater resistance to high-frequency artifacts, provide agent-specific spectral signatures, and allow the recognition of EEG features associated with noxious stimulation and with age extremes. Moreover, as the spectrogram retains most of the information present in the raw EEG, it allows case-to-case comparisons and more detailed physiological interpretations. The following section discusses the clinical implications of EEG spectrogram use during general anesthesia.

The EEG spectrogram: the optimal compromise between informational richness and clinical utility

Each class of anesthetic agents produces distinct EEG oscillatory patterns that reflect drug-specific neural circuit-related mechanisms. Increasingly, evidence suggests that these oscillations can be used to infer anesthetic depth [2,42]. Although raw EEGs provide the richest information for anesthetic management, their real-time interpretation is impractical for routine clinical use. In contrast, processed EEG indices may oversimplify anesthetic pharmacodynamics and can obscure clinically relevant EEG features. The EEG spectrogram offers a pragmatic compromise that reduces cognitive workload while preserving much of the information contained in the raw EEG [2–4,43]. Using fast Fourier transform-based spectral analysis, the spectrogram displays EEG power across frequencies over time, allowing visualization of the evolution of anesthetic-induced oscillations [2,4]. Contemporary anesthesia monitors commonly implement this time–frequency EEG representation, which is displayed as a color-coded DSA. Importantly, brief structured training has been shown to improve anesthesiologists’ ability to interpret DSAs substantially, suggesting a modest learning curve [44,45]. Consequently, spectrogram-based EEG monitoring, clinically operationalized as a DSA, represents a promising next-generation approach that may supersede processed EEG indices in monitoring anesthetic depth intraoperatively.

Clinical information beyond processed EEG indices conveyed by the spectrogram

Importance of maintaining sufficient alpha power

Phase–amplitude coupling between thalamocortical alpha (ca. 10 Hz) and delta (0.1–4 Hz) oscillations is a hallmark of general anesthesia, which disrupts neuronal communication through coupled rhythmic dynamics [46]. Anesthesia also induces alpha anteriorization, i.e., dissipation of the posterior alpha rhythm and emergence of a dominant frontal alpha oscillation [47,48]. This coherent frontal alpha activity across the thalamocortical networks reflects a stable, rhythmically organized, unconscious state [48] and a robust alpha rhythm indicates that the anesthetics have effectively engaged the thalamocortical circuitry [2].

A major advantage of EEG DSA over processed EEG indices is its ability to visualize the alpha power intraoperatively. Intraoperative alpha power is a neurophysiological marker of brain vulnerability and has been associated with perioperative and postoperative neurological and cognitive outcomes. For instance, reduced intraoperative alpha power has been linked to poor preoperative neurocognitive function [49], whereas both a lower peak alpha frequency [50] and diminished alpha power [51] have been associated with an increased risk of POD. A decrease in anesthesia-induced frontal alpha power also predicts a higher propensity for burst suppression [52], reflecting impaired cortical stability under anesthesia. Moreover, lower anesthetic dose-adjusted frontoparietal alpha power has also been correlated with greater odds of developing moderate-to-severe POD as well as with predisposing cognitive deficits, such as slower processing speed and impaired executive function, particularly in older adults [53]. Beyond delirium, intraoperative reductions in frontal alpha power have been associated with increased 30-, 90-, 180-day, and 1-year mortality as well as with discharge to long-term care facilities [54]. Quantitatively, each decibel increase in alpha power has been linked to a 14% reduction in the odds of POD [55].

Collectively, these findings underscore the prognostic significance of intraoperative alpha dynamics as potential biomarkers of perioperative brain health and resilience. This finding highlights the importance of applying DSA to guide the depth of anesthesia.

Identification of noxious stimulation

Processed EEG indices, such as the BIS, are not reliable for detecting noxious stimulation because the responses on these indices are variable, including increased, decreased, or unchanged values [43,56]. In contrast, several EEG spectrographic features, such as alpha dropout, beta arousal, and delta arousal, have been proposed as indicators of noxious stimulation during general anesthesia [57]. Alpha dropout, i.e., a transient, unexpected loss of alpha power, has been reported as a potential marker of thalamocortical depolarization that is induced by noxious stimulation arising from body cavity manipulation [58]. Fig. 1 shows two examples of beta arousal on DSA, each demonstrating three characteristic features: an increase in beta power, alpha dropout, and a decrease in delta power.

While beta arousal is the “classic” pattern of noxious stimulation under general anesthesia, “delta arousal,” i.e., a simultaneous increase in delta power and a reduction in alpha power, is often paradoxically misinterpreted by processed EEG index monitoring systems as “deeper” anesthesia. Thus, this pattern requires careful consideration because it poses a higher risk of clinical mismanagement [57]. Delta arousal is thought to occur more frequently during noxious stimulation with inadequate analgesia, particularly during abdominal surgery where the visceral pain pathways are involved [59]. Although the midbrain reticular formation is believed to contribute, the mechanisms underlying delta arousal remain incompletely understood. One hypothesis is that it involves exaggerated GABAergic negative feedback loops within the brainstem and thalamic regions [57]. The simultaneous increase in delta power and reduction in alpha power may paradoxically result in a decrease in the BIS value, falsely suggesting that the hypnotic dose is too high [57,60]. DSA offers a more detailed and physiologically meaningful assessment for guiding anesthesia management than that provided by processed EEG indices. Fig. 2 illustrates two examples of delta arousal on DSA: one observed during trocar insertion in laparoscopic surgery (Fig. 2A) and the other observed in a pediatric patient undergoing radiofrequency ablation of an orbital tumor (Fig. 2B).

In addition to the aforementioned spectrogram patterns associated with noxious stimulation, fentanyl has been reported to increase theta power on EEG spectrograms [61]. This feature may further facilitate the real-time monitoring of intraoperative opioid effects.

Precise identification of anesthetic overdosage and artifacts

When evaluating over-anesthesia by using EEG monitoring, the most apparent EEG feature of over-anesthesia is burst suppression, which has been associated with an increased risk of POD [62]. All current commercial frontal EEG monitoring systems report the Suppression Ratio (SR), which quantifies the percentage of time the EEG is flat or suppressed (isoelectric) within a given period, and is generally recommended to be maintained as close to zero as possible during surgery. However, this numerical parameter may not accurately reflect true burst suppression because it can be affected by algorithmic assumptions and technical artifacts. For example, visual analysis of EEG recordings and the SedLine^®-generated SR have yielded significantly different estimates of EEG suppression time, with the SedLine^® system markedly underestimating the extent of suppression [63].

The SR reported by commercial frontal EEG monitoring systems simply quantifies the low-voltage (flatline) EEG activity percentage in the preceding 30 (e.g., Conox^®) or 63 seconds (e.g., BIS^®), and hence false negative results may be calculated when interference is present from intraoperative high-frequency artifacts, which are not uncommon during surgery. For instance, alternating-current power sources operating at 50 or 60 Hz can introduce electrical interference in electrocardiography signals [64]. Because these frequencies fall within the spectrum of brain activity, power- line noise at 50/60 Hz is also commonly detected by EEG sensors and may significantly distort the processed EEG indices [65,66]. In addition to power-line interference, electrocautery is another common source of high-frequency noise [67].

In contrast, DSA is relatively resistant to such interference, because the characteristic alpha–delta coupling remains discernible even in the presence of 50/60 Hz noise. Fig. 3 presents the findings in three representative cases demonstrating spuriously elevated BIS readings, a phenomenon that can mislead anesthetic titration, resulting in excessive dosing with an increased risk of over-anesthesia.

In addition, the spectral edge frequency (SEF), which represents the frequency below which 95% of the total EEG power is encompassed, also showed falsely elevated values arising from artifacts in all three cases. Therefore, while the SR and SEF offer convenient quantitative indicators, direct inspection of the DSA is essential for distinguishing true physiological patterns from artifacts. The abovementioned discrepancies emphasize that numerical indices alone should not form the basis of anesthetic titration, but should be corroborated by DSA patterns.

Application of multimodal general anesthesia

Multimodal general anesthesia, a concept introduced in 2018, integrates the use of multiple anesthetic agents with distinct central nervous system targets under EEG spectrogram-guidance to optimize hypnotic and antinociceptive effects, while minimizing drug-related adverse events [68]. This approach enables more precise anesthetic titration, promotes faster postoperative recovery, and reduces opioid requirements [69]. Among the common adjuvants, ketamine and dexmedetomidine play key roles [68]. Although ketamine is known to increase the values of processed EEG indices, such as the BIS [70], Linassi et al. [37] reported a biphasic, nonlinear BIS response relative to the ketamine effect-site concentration, which peaked at lower doses and declined thereafter, making the BIS unreliable for dose adjustment. In contrast, changes in DSA consistently remained correlated with ketamine concentration [37]. Fig. 4 shows two patients of the same age and sex who underwent the same surgery and received identical classes of anesthetic agents (propofol, dexmedetomidine, and ketamine) within a standardized multimodal anesthesia protocol. However, DSA-guided anesthetic titration and the resulting spectrogram patterns differed, reflecting individualized patient responses and highlighting the role of DSA in personalized anesthetic management.

When used with GABAergic agents, dexmedetomidine produces DSA patterns overlapping those of GABAergic hypnotics, but has GABA-sparing effects [71,72]. In a propensity score-matched study that compared BIS- and DSA-guided multimodal anesthesia, which included propofol, dexmedetomidine, remifentanil, and a scalp block, for craniotomy, the DSA-guided approach reduced propofol consumption by approximately 30% [73]. This difference likely reflected the greater susceptibility of BIS to false elevation of readings by high-frequency surgical artifacts, as well as the superior accuracy of DSA in capturing the true depth of anesthesia and guiding precise anesthetic titration.

A combination of ketamine and dexmedetomidine is frequently used for opioid-free anesthesia (OFA) techniques. However, the existing literature reports minimal clinical benefits for OFA as compared with conventional opioid-based approaches [74]. Most prior OFA studies have relied on processed EEG indices, using manufacturer-recommended index ranges or purely clinical signs for anesthetic guidance, with limited DSA application [74]. Notably, DSA-guided OFA has been associated with processed EEG index values that are higher than the manufacturer-recommended targets [75], suggesting that reliance on conventional index-based approaches may lead to excessive anesthetic dosing. Such overdosing can result in adverse outcomes, including prolonged post-anesthesia care unit stay and profound bradycardia [76].

Collectively, these findings support the rationale for employing DSA-guided multimodal general anesthesia over using processed EEG indices to realize the pharmacological and recovery benefits of the multimodal approach more fully.

Reflecting anesthetic responses among pediatric and geriatric patients

The utility of processed EEG indices is known to be limited in populations at the extremes of the age spectrum, i.e., pediatric and geriatric patients. In this context, although anesthesia titration based on DSA patterns remains challenging, DSA offers clear advantages over processed EEG indices.

Age-specific EEG patterns during sevoflurane anesthesia have been characterized in children. Slow-delta oscillations are present in individuals of all ages, theta and alpha oscillations emerge at the age of approximately 4 months; alpha power is increased between the ages of 4 and 10 months; frontal alpha predominance appears in at the age of approximately 6 months; frontal slow oscillations become coherent from birth to 6 months of age; and frontal alpha coherence develops by the age of 10 months and persists thereafter [77]. Sevoflurane anesthesia has been reported to induce alpha–delta coupling in children aged 10 months to 3 years, resembling an adult EEG pattern [78]. As concerns regarding the relationship between early anesthetic exposure and neurodevelopmental outcomes persist [79] anesthetic depth must be accurately assessed in pediatric patients.

Because EEG dynamics are highly age-dependent, DSA-guided anesthesia is preferable to a “one-size-fits-all” processed EEG index approach, which often yields inconsistent depth values in younger patients [80]. To date, studies evaluating EEG-guided anesthesia in pediatric populations have mainly compared BIS- or DSA-guided protocols and standard clinical monitoring without EEG. Nevertheless, prospective investigations directly comparing DSA-guided and processed EEG index-guided anesthesia in children remain lacking. However, a recent meta-analysis reported that BIS monitoring was associated with a shorter time to emergence, with a mean difference of 2.67 minutes, but did not influence the incidence of pediatric anesthesia emergence delirium (PAED) [81]. In contrast, studies utilizing DSA-guided anesthesia have demonstrated greater reductions in emergence time, ranging from 6 to 21 minutes, and a lower incidence of PAED than that observed with standard clinical management [82,83]. Given the marked variability in the developing brain in children, the use of DSA in conjunction with raw EEG is recommended over the reliance on processed EEG indices alone in pediatric patients [84].

In geriatric patients, age-related neurophysiological changes result in altered EEG dynamics under anesthesia. Advancing age is associated with attenuated alpha power, reduced alpha–delta coupling, and a shift toward slower oscillatory activity [85]. These alterations contribute to the frequent observation of higher processed EEG index values in older adults when using almost all currently available neuromonitoring devices, despite adequate or excessive anesthetic dosing [86]. Fig. 5 depicts the cases of two geriatric patients demonstrating a mismatch between the processed EEG index- and DSA-based assessments of anesthetic depth. As in pediatric populations, prospective trials directly comparing DSA-guided and processed index-guided anesthesia in geriatric patients are lacking. A recent study reported that, in older patients undergoing major abdominal surgery, DSA-guided anesthesia significantly decreased the intraoperative EEG suppression time as compared with standard care [87]. Furthermore, a single-center randomized controlled trial demonstrated that implementation of a DSA-guided management curriculum for anesthesia residents was associated with a shorter hospital length of stay among surgical patients aged ≥ 60 years than that of a BIS-guided control group, even though the mean administered age-adjusted minimum alveolar concentration was the same [88]. These preliminary results support the potential benefit of EEG spectrogram-guided anesthesia over processed EEG index-based protocols in geriatric patients.

Barriers to and limitations of using EEG spectrograms for anesthesia guidance

Processed EEG index-guided anesthesia is touted to be more easily attained than that guided by EEG spectrograms. However, a recent study reported that the rate of processed EEG implementation for anesthesia guidance was only 32.8% in a large cohort (n = 42 932) of cardiac surgery patients [89]. A recent survey revealed that raw EEG waveform analysis or EEG spectrograms were seldom used (9%) for depth of anesthesia monitoring in Norway [90]. However, EEG spectrograms are relatively easy to obtain. The ability of inexperienced anesthesia clinicians to recognize characteristic EEG patterns at various depths of hypnosis and detect discrepancies between the processed EEG indices and corresponding EEG waveforms was markedly enhanced by implementation of a brief, structured training program [45]. In addition, the Safe Brain Initiative, now an European Society of Anaesthesiology and Intensive Care research group, has developed a standardized EEG bootcamp curriculum to support anesthesiologists’ understanding of EEG monitoring in routine practice. Participants reported significant improvements in self-rated knowledge, competence, and attitudes toward EEG-guided anesthesia [44]. Collectively, these findings strongly support the concept that education on EEG spectrography is both essential and effective in contemporary anesthesiology training.

Despite these advantages, EEG spectrograms generated by current commercial EEG monitoring systems have substantial limitations. First, contemporary systems are relatively insensitive to subtle low-frequency changes, such as delta arousal [57]. Differentiating slow-delta (0.33–2.33 Hz) from fast-delta/theta activity (2.67–6.33 Hz) is physiologically important, as these sub-bands have distinct relationships with sleep depth, anesthetic dose, and clinical outcomes, including mortality in the intensive care unit [91]. Second, although perioperative alpha power has emerged as a clinically relevant biomarker, current monitoring systems do not provide quantitative alpha power metrics, limiting standardization and cross-study comparability, as compared to index-based targets (e.g., BIS 40–60). The incorporation of quantitative alpha metrics into future EEG monitoring systems could enhance anesthetic titration as well as research consistency. Third, rigorous attention to EEG acquisition and display settings is required, as undocumented changes in amplitude, sampling rate, or signal quality may occur when display configurations are modified, as reported with the SedLine® system [92]. Therefore, standardization of display and export parameters is essential. Finally, a significant technical limitation is that a DSA alone cannot fully resolve certain complex EEG patterns that are clinically relevant for assessing anesthetic depth. A prominent example is the differentiation of alpha–delta phase-coupling configurations. Although the presence of frontal alpha–delta activity is a hallmark of anesthesia, it does not necessarily preclude volitional responsiveness, which occurs particularly in “trough-max” coupling, where the alpha amplitude peaks at the trough of slow-delta oscillations [93]. In contrast, peak-max coupling reflects widespread cortical “OFF” states, consistent with deeper unconsciousness [94]. Because DSA alone cannot distinguish these patterns, analysis of a comodulogram or inspection of the raw EEG is required, both of which are challenging to perform at the bedside and are susceptible to environmental noise. Thus, although DSA offers richer neurophysiological information than that obtained from processed EEG indices, a foundational understanding of raw EEG principles remains essential for anesthesiologists aiming to deliver precise, physiologically based, and individualized anesthesia care.

Conclusion

The evolution of brain monitoring during anesthesia reflects a shift from convenience to neurophysiological precision. Processed EEG indices provided a valuable starting point for cerebral monitoring, but now risk constraining interpretation within opaque numerical boundaries. In contrast, EEG spectrograms provide transparency and context, allowing anesthesiologists to visualize anesthetic agent-specific signatures, cortical stability, and arousal-related transitions in real-time. Although the learning curve and current monitoring limitations are valid barriers, they are not insurmountable. The feasibility and value of effective training on the use of EEG spectrograms in anesthesia was established. Therefore, the remaining central challenges are cultural and technological. Anesthesiologists must embrace EEG spectrogram interpretation as a core competency in modern anesthesiology, as this marries technology with true neurophysiological insights.

Acknowledgments

We thank Dr. Jyu-shiou Ho for providing Figure 2A and Dr. Po-Yuan Shih for assistance in improving the image quality of the figures.

Funding

None.

Conflicts of Interest

No potential conflict of interest relevant to this article was reported.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Author Contributions

Akira Mukai (Conceptualization; Data curation; Investigation; Project administration; Validation; Visualization; Writing – original draft; Writing – review & editing)

Jen-Ting Yang (Conceptualization; Data curation; Investigation; Validation; Writing – original draft; Writing – review & editing)

Shao-Chun Wu (Conceptualization; Data curation; Investigation; Supervision; Writing – review & editing)

Tzu-Chun Wang (Investigation; Validation; Visualization; Writing – original draft)

Feng-Sheng Lin (Investigation; Resources; Validation; Writing – original draft)

Chun-Yu Wu (Conceptualization; Data curation; Investigation; Resources; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing)

Fig. 1.

Classic Density Spectral Array Pattern of Noxious Stimulation: Beta Arousal. (A) The density spectral array (DSA) obtained for a 64-year-old male undergoing lumbar spine fusion under sevoflurane, dexmedetomidine, and low-dose ketamine anesthesia. At 14:51, a noxious stimulus induced a marked increase in beta power, which was accompanied by a decrease in alpha and delta power. (B) The DSA obtained for a 71-year-old female who received total intravenous anesthesia for lumbar spine fusion surgery. During pedicle screw insertion at 09:10, the DSA showed a subtle beta arousal pattern characterized by a modest beta power increase and a concomitant reduction in delta power.

Fig. 2.

Paradoxical Density Spectral Array Pattern of Noxious Stimulation: Delta Arousal. (A) Delta arousal observed at 13:45 on density spectral array (DSA) during trocar insertion during laparoscopic surgery. (B) Delta arousal recorded on the DSA of a pediatric patient undergoing radiofrequency ablation of an orbital tumor, demonstrating similar DSA changes at 15:40.

Fig. 3.

Falsely Elevated Bispectral Index Values Without Evidence of Under-Anesthesia. (A) A 74-year-old patient underwent awake craniotomy (asleep–awake–asleep technique). During the initial asleep phase, high-frequency artifacts from scalp resection produced spuriously elevated Bispectral Index (BIS) values. However, the density spectral array (DSA) revealed stable anesthetic depth with preserved alpha–delta coupling. Reliance on the BIS value alone could have led to excessive anesthetic dosing and delayed awakening. (B) A 65-year-old patient who underwent craniotomy for intracranial lymphoma biopsy exhibited falsely high BIS values and inaccurate suppression metrics (zero Suppression Ratio and Suppression Time) due to 50/60-Hz power-line noise. The over-anesthetized state was evident on the DSA, which normalized once the power-line interference was corrected. (C) An example of falsely high BIS readings caused by electrocautery interference, even though adequate alpha–delta coupling was visible on the DSA.

Fig. 4.

Density Spectral Array Changes Following Ketamine Dose Adjustments. (A) The density spectral array (DSA) obtained for a 15-year-old girl who received multimodal general anesthesia (propofol, dexmedetomidine, ketamine, and erector spinae plane block) for scoliosis correction. Following reduction of ketamine infusion at ca. 11:40, alpha power increased and an alpha–delta-dominant pattern re-emerged. The Bispectral Index (BIS) values showed minimal variation despite the dose change (bottom yellow trace). (B) The DSA obtained for another 15-year-old girl who received multimodal general anesthesia (propofol, dexmedetomidine, ketamine, and erector spinae plane block). Increased ketamine dosing produced a frequency shift toward higher-frequency alpha activity together with increased beta bands, while BIS values remained relatively low, illustrating that the DSA has a greater dose-sensitivity than the BIS.

Fig. 5.

Geriatric Patients Demonstrating Over-Anesthesia as Evidenced by Density Spectral Array, Despite Misleadingly High BIS Values. (A) DSA obtained for a 79-year-old patient with dementia who underwent hip debridement under propofol and alfentanil anesthesia. The DSA demonstrated over-anesthesia, as evidenced by a non-zero suppression ratio, while the BIS value remained falsely elevated (bottom yellow trace). The BIS signal was temporarily lost during the position change from supine to prone when the electrodes were briefly detached. (B) DSA obtained for an 85-year-old patient who underwent lymphovenous bypass under propofol and alfentanil anesthesia. The BIS value was discrepant from the observed decline in alpha power, indicating a mismatch between the processed EEG index and spectrogram-based assessment of anesthetic depth.

References

1. Al-Kadi MI, Reaz MB, Ali MA. Evolution of electroencephalogram signal analysis techniques during anesthesia. Sensors (Basel) 2013; 13: 6605-35.

2. Purdon PL, Sampson A, Pavone KJ, Brown EN. Clinical electroencephalography for anesthesiologists: Part I: background and basic signatures. Anesthesiology 2015; 123: 937-60.

3. Lee KH, Egan TD, Johnson KB. Raw and processed electroencephalography in modern anesthesia practice: a brief primer on select clinical applications. Korean J Anesthesiol 2021; 74: 465-77.

4. Kim MC, Fricchione GL, Brown EN, Akeju O. Role of electroencephalogram oscillations and the spectrogram in monitoring anaesthesia. BJA Educ 2020; 20: 166-72.

5. Berger-Estilita J, Saxena S, Gisselbaek M. Advancing electroencephalography education in anesthesiology. Curr Opin Anaesthesiol 2025; 38: 576-83.

6. Hight DF, Kaiser HA, Sleigh JW, Avidan MS. Continuing professional development module: an updated introduction to electroencephalogram-based brain monitoring during intended general anesthesia. Can J Anaesth 2020; 67: 1858-78.

7. Rossini PM, Cole J, Paulus W, Ziemann U, Chen R. 1924-2024: first centennial of EEG. Clin Neurophysiol 2025; 170: 132-5.

8. Dahaba AA. Different conditions that could result in the bispectral index indicating an incorrect hypnotic state. Anesth Analg 2005; 101: 765-73.

9. Bard JW. The BIS monitor: a review and technology assessment. AANA J 2001; 69: 477-83.

10. Gorges M, West NC, Cooke EM, Pi S, Brant RF, Dumont GA, et al. Evaluating NeuroSENSE for assessing depth of hypnosis during desflurane anesthesia: an adaptive, randomized-controlled trial. Can J Anaesth 2020; 67: 324-35.

11. Aho AJ, Kamata K, Jantti V, Kulkas A, Hagihira S, Huhtala H, et al. Comparison of Bispectral Index and Entropy values with electroencephalogram during surgical anaesthesia with sevoflurane. Br J Anaesth 2015; 115: 258-66.

12. Jensen EW, Valencia JF, Lopez A, Anglada T, Agusti M, Ramos Y, et al. Monitoring hypnotic effect and nociception with two EEG-derived indices, qCON and qNOX, during general anaesthesia. Acta Anaesthesiol Scand 2014; 58: 933-41.

13. Springman SR, Andrei AC, Willmann K, Rusy DA, Warren ME, Han S, et al. A comparison of SNAP II and bispectral index monitoring in patients undergoing sedation. Anaesthesia 2010; 65: 815-9.

14. Kreuer S, Wilhelm W, Grundmann U, Larsen R, Bruhn J. Narcotrend index versus bispectral index as electroencephalogram measures of anesthetic drug effect during propofol anesthesia. Anesth Analg 2004; 98: 692-7.

15. Prichep LS, Gugino LD, John ER, Chabot RJ, Howard B, Merkin H, et al. The Patient State Index as an indicator of the level of hypnosis under general anaesthesia. Br J Anaesth 2004; 92: 393-9.

16. Han DW. Do you believe that processed EEG helps to prevent intraoperative awareness? Korean J Anesthesiol 2018; 71: 427-9.

17. Myles PS, Leslie K, McNeil J, Forbes A, Chan MT. Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004; 363: 1757-63.

18. Avidan MS, Zhang L, Burnside BA, Finkel KJ, Searleman AC, Selvidge JA, et al. Anesthesia awareness and the bispectral index. N Engl J Med 2008; 358: 1097-108.

19. Mashour GA, Shanks A, Tremper KK, Kheterpal S, Turner CR, Ramachandran SK, et al. Prevention of intraoperative awareness with explicit recall in an unselected surgical population: a randomized comparative effectiveness trial. Anesthesiology 2012; 117: 717-25.

20. Gao WW, He YH, Liu L, Yuan Q, Wang YF, Zhao B. BIS monitoring on intraoperative awareness: a meta-analysis. Curr Med Sci 2018; 38: 349-53.

21. Laferriere-Langlois P, Morisson L, Jeffries S, Duclos C, Espitalier F, Richebe P. Depth of anesthesia and nociception monitoring: current state and vision for 2050. Anesth Analg 2024; 138: 295-307.

22. Wildes TS, Mickle AM, Ben Abdallah A, Maybrier HR, Oberhaus J, Budelier TP, et al. Effect of electroencephalography-guided anesthetic administration on postoperative delirium among older adults undergoing major surgery: the engages randomized clinical trial. JAMA 2019; 321: 473-83.

23. Short TG, Campbell D, Frampton C, Chan MT, Myles PS, Corcoran TB, et al. Anaesthetic depth and complications after major surgery: an international, randomised controlled trial. Lancet 2019; 394: 1907-14.

24. Chan MT, Cheng BC, Lee TM, Gin T. BIS-guided anesthesia decreases postoperative delirium and cognitive decline. J Neurosurg Anesthesiol 2013; 25: 33-42.

25. Ackland GL, Pryor KO. Electroencephalography-guided anaesthetic administration does not impact postoperative delirium among older adults undergoing major surgery: an independent discussion of the ENGAGES trial. Br J Anaesth 2019; 123: 112-7.

26. Deschamps A, Ben Abdallah A, Jacobsohn E, Saha T, Djaiani G, El-Gabalawy R, et al. Electroencephalography-guided anesthesia and delirium in older adults after cardiac surgery: the ENGAGES-Canada randomized clinical trial. JAMA 2024; 332: 112-23.

27. Evered LA, Chan MT, Han R, Chu MH, Cheng BP, Scott DA, et al. Anaesthetic depth and delirium after major surgery: a randomised clinical trial. Br J Anaesth 2021; 127: 704-12.

28. Sumner M, Deng C, Evered L, Frampton C, Leslie K, Short T, et al. Processed electroencephalography-guided general anaesthesia to reduce postoperative delirium: a systematic review and meta-analysis. Br J Anaesth 2023; 130: e243-53.

29. Xu A, Arnaout B, Scott DA, McGuigan S. Intraoperative electroencephalogram-derived measures and their association with postoperative delirium: a systematic review and meta-analysis. Br J Anaesth 2025; 135: 1684-703.

30. Hajat Z, Ahmad N, Andrzejowski J. The role and limitations of EEG-based depth of anaesthesia monitoring in theatres and intensive care. Anaesthesia 2017; 72 Suppl 1: 38-47.

31. 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.

32. Schuller PJ, Pretorius JP, Newbery KB. Response of the GE Entropy™ monitor to neuromuscular block in awake volunteers. Br J Anaesth 2023; 131: 882-92.

33. Schuller PJ, Pretorius JP, Newbery KB. Response of the Conox quantitative electroencephalographic monitor to neuromuscular block in awake volunteers. Br J Anaesth 2025; 135: 660-7.

34. Dahaba AA, Bornemann H, Hopfgartner E, Ohran M, Kocher K, Liebmann M, et al. Effect of sugammadex or neostigmine neuromuscular block reversal on bispectral index monitoring of propofol/remifentanil anaesthesia. Br J Anaesth 2012; 108: 602-6.

35. Coleman RM, Tousignant-Laflamme Y, Ouellet P, Parenteau-Goudreault E, Cogan J, Bourgault P. The use of the bispectral index in the detection of pain in mechanically ventilated adults in the intensive care unit: a review of the literature. Pain Res Manag 2015; 20: e33-7.

36. Shahiri TS, Richebe P, Richard-Lalonde M, Gelinas C. Description of the validity of the Analgesia Nociception Index (ANI) and Nociception Level Index (NOL) for nociception assessment in anesthetized patients undergoing surgery: a systematized review. J Clin Monit Comput 2022; 36: 623-35.

37. Linassi F, Troyas C, Kreuzer M, Spano L, Burelli P, Schneider G, et al. Effect of ketamine on the bispectral index, spectral edge frequency, and surgical pleth index during propofol-remifentanil anesthesia: an observational prospective trial. Anesth Analg 2025; 140: 1276-85.

38. Kasuya Y, Govinda R, Rauch S, Mascha EJ, Sessler DI, Turan A. The correlation between bispectral index and observational sedation scale in volunteers sedated with dexmedetomidine and propofol. Anesth Analg 2009; 109: 1811-5.

39. Ni K, Cooter M, Gupta DK, Thomas J, Hopkins TJ, Miller TE, et al. Paradox of age: older patients receive higher age-adjusted minimum alveolar concentration fractions of volatile anaesthetics yet display higher bispectral index values. Br J Anaesth 2019; 123: 288-97.

40. Obert DP, Schneider F, Schneider G, von Dincklage F, Sepulveda P, Garcia PS, et al. Performance of the SEDLine monitor: age dependency and time delay. Anesth Analg 2023; 137: 887-95.

41. Hight D, Kreuzer M, Ugen G, Schuller P, Stuber F, Sleigh J, et al. Five commercial 'depth of anaesthesia' monitors provide discordant clinical recommendations in response to identical emergence-like EEG signals. Br J Anaesth 2023; 130: 536-45.

42. Akeju O, Brown EN. Neural oscillations demonstrate that general anesthesia and sedative states are neurophysiologically distinct from sleep. Curr Opin Neurobiol 2017; 44: 178-85.

43. Hagihira S. Changes in the electroencephalogram during anaesthesia and their physiological basis. Br J Anaesth 2015; 115 Suppl 1: i27-31.

44. von Dincklage F, Helfrich J, Koch S, Soehle M, Berger-Estilita J, Bublitz V, et al. Introducing the Safe Brain Initiative's EEG boot camp for anaesthesia for standardised training on how to use the electroencephalogram for perioperative care. BMC Anesthesiol 2025; 25: 449.

45. Bombardieri AM, Wildes TS, Stevens T, Wolfson M, Steinhorn R, Ben Abdallah A, et al. Practical training of anesthesia clinicians in electroencephalogram-based determination of hypnotic depth of general anesthesia. Anesth Analg 2020; 130: 777-86.

46. Soplata AE, McCarthy MM, Sherfey J, Lee S, Purdon PL, Brown EN, et al. Thalamocortical control of propofol phase-amplitude coupling. PLoS Comput Biol 2017; 13: e1005879.

47. Vijayan S, Ching S, Purdon PL, Brown EN, Kopell NJ. Thalamocortical mechanisms for the anteriorization of α rhythms during propofol-induced unconsciousness. J Neurosci 2013; 33: 11070-5.

48. Weiner VS, Zhou DW, Kahali P, Stephen EP, Peterfreund RA, Aglio LS, et al. Propofol disrupts alpha dynamics in functionally distinct thalamocortical networks during loss of consciousness. Proc Natl Acad Sci U S A 2023; 120: e2207831120.

49. Giattino CM, Gardner JE, Sbahi FM, Roberts KC, Cooter M, Moretti E, et al. Intraoperative frontal alpha-band power correlates with preoperative neurocognitive function in older adults. Front Syst Neurosci 2017; 11: 24.

50. Hight D, Ehrhardt A, Lersch F, Luedi MM, Stuber F, Kaiser HA. Lower alpha frequency of intraoperative frontal EEG is associated with postoperative delirium: a secondary propensity-matched analysis. J Clin Anesth 2024; 93: 111343.

51. Kinoshita H, Saito J, Kushikata T, Oyama T, Takekawa D, Hashiba E, et al. The perioperative frontal relative ratio of the alpha power of electroencephalography for predicting postoperative delirium after highly invasive surgery: a prospective observational study. Anesth Analg 2023; 137: 1279-88.

52. Shao YR, Kahali P, Houle TT, Deng H, Colvin C, Dickerson BC, et al. Low frontal alpha power is associated with the propensity for burst suppression: an electroencephalogram phenotype for a "Vulnerable Brain". Anesth Analg 2020; 131: 1529-39.

53. Reese M, Wright MC, Roberts KC, Browndyke JN, Bennett M, Acker L, et al. Associations between anaesthetic dose-adjusted intraoperative EEG alpha power, processing speed, and postoperative delirium: analysis of data from three prospective studies. Br J Anaesth 2025; 135: 109-20.

54. Mather RV, Nipp R, Balanza G, Stone TA, Gutierrez R, Raje P, et al. Intraoperative frontal electroencephalogram alpha power is associated with postoperative mortality and other adverse outcomes. Anesthesiology 2025; 142: 500-10.

55. Freedman IG, Boncompte G, Qu JZ, Khawaja ZQ, Turco I, Mueller A, et al. Anesthesia-induced electroencephalogram oscillations and perioperative outcomes in older adults undergoing cardiac surgery. J Clin Anesth 2025; 102: 111770.

56. Takamatsu I, Ozaki M, Kazama T. Entropy indices vs the bispectral index for estimating nociception during sevoflurane anaesthesia. Br J Anaesth 2006; 96: 620-6.

57. Garcia PS, Kreuzer M, Hight D, Sleigh JW. Effects of noxious stimulation on the electroencephalogram during general anaesthesia: a narrative review and approach to analgesic titration. Br J Anaesth 2021; 126: 445-57.

58. Hight DF, Gaskell AL, Kreuzer M, Voss LJ, Garcia PS, Sleigh JW. Transient electroencephalographic alpha power loss during maintenance of general anaesthesia. Br J Anaesth 2019; 122: 635-42.

59. Bischoff P, Kochs E, Haferkorn D, Schulte am Esch J. Intraoperative EEG changes in relation to the surgical procedure during isoflurane-nitrous oxide anesthesia: hysterectomy versus mastectomy. J Clin Anesth 1996; 8: 36-43.

60. Morimoto Y, Matsumoto A, Koizumi Y, Gohara T, Sakabe T, Hagihira S. Changes in the bispectral index during intraabdominal irrigation in patients anesthetized with nitrous oxide and sevoflurane. Anesth Analg 2005; 100: 1370-4.

61. Balanza GA, Bharadwaj KM, Mullen AC, Beck AM, Work EC, McGovern FJ, et al. An electroencephalogram biomarker of fentanyl drug effects. PNAS Nexus 2022; 1: pgac158.

62. Park SK, Han DW, Chang CH, Jung H, Kang H, Song Y. Association between intraoperative electroencephalogram burst suppression and postoperative delirium: a systematic review and meta-analysis. Anesthesiology 2025; 142: 107-20.

63. Muhlhofer WG, Zak R, Kamal T, Rizvi B, Sands LP, Yuan M, 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.

64. Wan SW, Nguyen HT. 50Hz interference and noise in ECG recordings--a review. Australas Phys Eng Sci Med 1994; 17: 108-15.

65. Roy V. Effective EEG artifact removal from EEG signal. In: Biosignal Processing. Edited by Asadpour V, Karakuş S: IntechOpen, 2022. Available from https://www.intechopen.com/chapters/80529.

66. Teplan M. Fundamentals of EEG measurement. Meas Sci Rev 2002; 2: 1-11.

67. Tavakoli Golpaygani A, Movahedi MM, Reza M. A study on performance and safety tests of electrosurgical equipment. J Biomed Phys Eng 2016; 6: 175-82.

68. Brown EN, Pavone KJ, Naranjo M. Multimodal general anesthesia: theory and practice. Anesth Analg 2018; 127: 1246-58.

69. Lersch F, Correia PC, Hight D, Kaiser HA, Berger-Estilita J. The nuts and bolts of multimodal anaesthesia in the 21st century: a primer for clinicians. Curr Opin Anaesthesiol 2023; 36: 666-75.

70. Hans P, Dewandre PY, Brichant JF, Bonhomme V. Comparative effects of ketamine on Bispectral Index and spectral entropy of the electroencephalogram under sevoflurane anaesthesia. Br J Anaesth 2005; 94: 336-40.

71. Vetter C, Meyer ER, Seidel K, Bervini D, Huber M, Krejci V. Co-administration of dexmedetomidine with total intravenous anaesthesia in carotid endarterectomy reduces requirements for propofol and improves haemodynamic stability: A single-centre, prospective, randomised controlled trial. Eur J Anaesthesiol 2025; 42: 255-64.

72. Chen PH, Tsuang FY, Lee CT, Yeh YC, Cheng HL, Lee TS, et al. Neuroprotective effects of intraoperative dexmedetomidine versus saline infusion combined with goal-directed haemodynamic therapy for patients undergoing cranial surgery: A randomised controlled trial. Eur J Anaesthesiol 2021; 38: 1262-71.

73. Lin FS, Shih PY, Sung CH, Chou WH, Wu CY. Electroencephalographic spectrogram-guided total intravenous anesthesia using dexmedetomidine and propofol prevents unnecessary anesthetic dosing during craniotomy: a propensity score-matched analysis. Korean J Anesthesiol 2024; 77: 122-32.

74. Tripodi VF, Sardo S, Ippolito M, Cortegiani A. Effectiveness and safety of opioid-free anesthesia compared to opioid-based anesthesia: a systematic review and network meta-analysis. J Anesth Analg Crit Care 2025; 5: 53.

75. Mogianos K, Persson AK. Anesthesia depth monitoring during opioid free anesthesia - a prospective observational study. BMC Anesthesiol 2025; 25: 37.

76. Beloeil H, Garot M, Lebuffe G, Gerbaud A, Bila J, Cuvillon P, et al. Balanced opioid-free anesthesia with dexmedetomidine versus balanced anesthesia with remifentanil for major or intermediate noncardiac surgery. Anesthesiology 2021; 134: 541-51.

77. Cornelissen L, Kim SE, Lee JM, Brown EN, Purdon PL, Berde CB. Electroencephalographic markers of brain development during sevoflurane anaesthesia in children up to 3 years old. Br J Anaesth 2018; 120: 1274-86.

78. Zakaria L, Desowska A, Berde CB, Cornelissen L. Electroencephalographic delta and alpha oscillations reveal phase-amplitude coupling in paediatric patients undergoing sevoflurane-based general anaesthesia. Br J Anaesth 2023; 130: 595-602.

79. Reighard C, Junaid S, Jackson WM, Arif A, Waddington H, Whitehouse AJ, et al. Anesthetic Exposure During Childhood and Neurodevelopmental Outcomes: A Systematic Review and Meta-analysis. JAMA Netw Open 2022; 5: e2217427.

80. Kim D, Kim J, Kim I, Gil NS, Shin YH, Jeong JS. Correlation between bispectral index and patient state index in children under sevoflurane anesthesia. Paediatr Anaesth 2022; 32: 740-6.

81. de Oliveira DA, Dos Santos Monteiro J, de Macedo NR, Tavares Costa Santos AH, Rosa Neves SM. Bispectral index-guided anesthesia in children: a systematic review and meta-analysis. Anaesth Crit Care Pain Med 2025; 44: 101607.

82. Miyasaka KW, Suzuki Y, Brown EN, Nagasaka Y. EEG-guided titration of sevoflurane and pediatric anesthesia emergence delirium: a randomized clinical trial. JAMA Pediatr 2025; 179: 704-12.

83. de Heer IJ, Raab HA, de Vries J, Karaoz-Bulut G, Weber F. The influence of electroencephalographic density spectral array guidance of sevoflurane administration on recovery from general anesthesia in children. a randomized controlled trial. Paediatr Anaesth 2025; 35: 287-93.

84. Yuan I, Bong CL, Chao JY. Intraoperative pediatric electroencephalography monitoring: an updated review. Korean J Anesthesiol 2024; 77: 289-305.

85. Purdon PL, Pavone KJ, Akeju O, Smith AC, Sampson AL, Lee J, et al. The ageing brain: age-dependent changes in the electroencephalogram during propofol and sevoflurane general anaesthesia. Br J Anaesth 2015; 115 Suppl 1(Suppl 1): i46-57.

86. Obert DP, Schweizer C, Zinn S, Kratzer S, Hight D, Sleigh J, et al. The influence of age on EEG-based anaesthesia indices. J Clin Anesth 2021; 73: 110325.

87. He Z, Zhang H, Xing Y, Liu J, Gao Y, Gu E, et al. Effect of raw electroencephalogram-guided anesthesia administration on postoperative outcomes in elderly patients undergoing abdominal major surgery: a randomized controlled trial. BMC Anesthesiol 2023; 23: 337.

88. Berger M, Eleswarpu SS, Cooter Wright M, Ray AM, Wingfield SA, Heflin MT, et al. Developing a real-time electroencephalogram-guided anesthesia-management curriculum for educating residents: a single-center randomized controlled trial. Anesth Analg 2022; 134: 159-70.

89. Lombard FW, Roy S, Shah AS, Feng X, Shotwell MS, Kertai MD. Processed electroencephalographic use during anesthesia and outcomes: analysis of the Society of Thoracic Surgeons adult cardiac surgery database. Ann Thorac Surg 2022; 114: 1688-94.

90. Aasheim A, Rosseland LA, Leonardsen AL, Romundstad L. Depth of anesthesia monitoring in Norway-A web-based survey. Acta Anaesthesiol Scand 2024; 68: 781-7.

91. Rodrigues A, Subira C, Bizios A, Younes M, Gerardy B, Fernandez R, et al. Sedation-related electroencephalographic patterns in acute hypoxemic respiratory failure. Anesthesiology 2025; 143: 1266-78.

92. von Dincklage F, Jurth C, Schneider G, S García P, Kreuzer M. Technical considerations when using the EEG export of the SEDLine Root device. J Clin Monit Comput 2021; 35: 1047-54.

93. Gaskell AL, Hight DF, Winders J, Tran G, Defresne A, Bonhomme V, et al. Frontal alpha-delta EEG does not preclude volitional response during anaesthesia: prospective cohort study of the isolated forearm technique. Br J Anaesth 2017; 119: 664-73.

94. Brown EN, Purdon PL, Akeju O, An J. Using EEG markers to make inferences about anaesthetic-induced altered states of arousal. Br J Anaesth 2018; 121: 325-7.