Intraoperative electroencephalogram (EEG) monitoring during general anesthesia has become an increasingly valuable tool in contemporary practice, offering a potential direct window into cortical activity. This approach allows anesthesiologists to individualize anesthetic dosing according to closest brain-level effects and hypnotic depth, thereby minimizing excessive exposure and its consequences, including hypotension or reduced cerebral perfusion (1,2). However, in pediatric patients in whom developing structures and functional connections are age-dependent, the resulting EEG waveforms arising from pyramidal neurons in the neocortex, may represent a topic that demands further understanding and accurate interpretation (3).

Building upon some basic concepts, in the EEG interpretation, waveforms are primarily described by two parameters: (I) amplitude (microvolts, µV), which reflects the synchrony of cortical electrical activity; and (II) frequency (hertz, Hz), representing the number of cycles per second (bands: slow <1 Hz, delta 1-4 Hz, theta 5-8 Hz, alpha 9-12Hz, Beta 13-25 Hz, Gamma >26 or 26-80 Hz) (1,3). EEG Power (Decibels) “is the quantity of EEG activity at a given frequency”(3).

In clinical practice, intraoperative EEG monitoring in children commonly involves devices such as BIS (bispectral index), Narcotrend, SedLine, and M-Entropy. Although processed EEG (pEEG) indices and the patient state index (PSI) have proven useful for guiding anesthetic depth in adults, in pediatric patients, particularly in infants under 1 year of age, their reliability remains limited. This finding arises because most algorithms are calibrated using adult EEG data.

Therefore, correlation between anesthetic depth and these proprietary indices is not recommended in this age group, and other strategies are required for anesthetic titration. Among these approaches, EEG assessment through unprocessed (raw) waveforms and non-proprietary processed parameters, including (I) the density spectral array (DSA), (II) the spectral edge frequency (SEF95), and (III) the burst suppression ratio (BSR) (3, 5-7) have been suggested. However, in patients younger than 3 months old, EEG waveforms provide more accurate guidance,  as certain SEF95 cutoff values may not correlate well or should be interpreted cautiously with anesthetic depth (3,4). In these cases, the presence of an isoelectric pattern on the raw EEG serves as a clearer indicator of cortical inactivity.

Evolution of EEG Patterns: Pediatric vs. Adult

Normal EEG changes are age-related and evolve markedly across development (3,8). For instance, adult pattern activity is characterized by low-amplitude beta waves (>13 Hz, 10–20 µV), with predominance during active wakefulness. Closing the eyes produces an immediate shift toward higher-amplitude alpha activity (8–13 Hz, 20–40 µV), and a progressive transition to theta and delta activity during drowsiness and sleep.

On the other hand, preterm neonates (28- 34 weeks) commonly exhibit intermittent periods of discontinuous or isoelectric EEG in wakefulness and sleep, which gradually disappear as age advances. Moreover, with further brain development, newborns present with a predominance of slow or delta-frequency oscillations which gradually decrease in amplitude. By 3- 5 months, an early alpha rhythm begins to emerge in the trace.

In children above 3 years, most conventional frequency band ranges typically become present. There are predominant theta waves that decrease during later childhood. In contrast, initially alpha is subtle but then strengthens and accelerates, becoming dominant during the following years.

Lastly, the EEG obtains an adult-like pattern by approximately 10- 19 years of age. Between 10 and 14 years, alpha activity becomes well defined, with theta and slow waves decreasing and becoming more organized. By late adolescence (>14 years), alpha rhythm prevails with reduced amplitude, while theta activity becomes minimal, and posterior slow waves disappear. After 19 years, the EEG closely resembles the adult pattern, with only occasional residual theta activity (8).

EEG Pattern under General Anesthesia

Under anesthesia, distinct electrical patterns can be identified, highlighting how anesthetic dose, patient age, and underlying neural structures interact to produce loss of consciousness. For instance, exposure to GABAergic agents such as sevoflurane and propofol has been studied in greater detail.

Propofol induced unconsciousness in adults is characterized by a disruption of the thalamocortical and corticocortical neurotransmitter systems, and its effect is dose-dependent in the brain. High doses of this agent shifts EEG into burst suppression, but in proper concentrations there are predominant slow (0.1- 1 Hz) and coherent frontal alpha (8–13 Hz) oscillations (6,9). Moreover, Lee at al. (9) in a prospective study of 97 patients aged 0-21 years noted that propofol produced EEG oscillations consistent with adulthood pattern from 1 year, whereas under this age alpha oscillations lacked coherence across regions; these changes mirror critical stages of neural maturation involving the thalamocortical circuit, whose connectivity and functionality are still developing in younger patients. Likewise, the EEG power (0.1- 40Hz) peaks around 8 years of life and progressively declines in the following years.

Adult patients under sevoflurane anesthesia exhibit organized and well-defined EEG patterns. These findings include alpha coherence, which is associated with loss of consciousness, along with increased activity in the 1–4 Hz (delta) and 0.1–1 Hz (slow) frequency ranges. On the other hand, in children under sevoflurane anesthesia, Cornelisset et al (4). observed in 26 infants between 0-6 months old, that slow and delta oscillations remained prominent in maintenance. But 0–3 months olds showed almost no EEG power above 4 Hz, whereas 4–6 month old infants developed clear theta and alpha components across frontal, central and parietal regions. During emergence, the EEG in 4–6 months patients showed declining theta and alpha power.

In parallel, Akeju et al. (10) examined age-related changes in frontal EEG power spectra and coherence across 54 patients aged 0–28 years. They found a substantial increase in EEG power (0.1- 50 Hz)  from infancy (>1 year) up to about 6 years of age, followed by a progressive decline that stabilized around 21 years. In terms of alpha coherence, infants older than 1 year were similar among all groups, in contrast with younger than 1 year who lacked significant alpha band power or coherence. See Figure 2 in the study by Akeju et al. (10).

Applying EEG to Practice with Anesthetics

The advantages of EEG monitoring in adults are well established in clinical practice supported by growing evidence and published studies. However, in the pediatric population, certain limitations have been identified with some devices, which explain why its use in early infancy remains under ongoing investigation. Nevertheless, its benefits may support broader incorporation into clinical practice whenever feasible, considering that EEG may still offer significant clinical value.

In patients younger than 1 year, it is important to remember that EEG differs from older children. For example, frontal alpha coherence is typically absent under 3 months of age, and overall EEG power tends to be reduced. This distinction is crucial, particularly in preterm infants, in whom baseline normal EEG patterns during wakefulness may appear discontinuous or burst-suppression–like activity, and could resemble anesthetic-induced suppression in older children, potentially leading to misinterpretation (3,6). Therefore, obtaining a pre-induction tracing is strongly recommended to better contextualize and avoid inappropriate dose adjustments.

In term neonates or infants younger than 3 months, the detection of sustained isoelectricity on the raw EEG during steady sevoflurane administration should prompt careful titration of the anesthetic concentration, regardless of the measured age-adjusted MAC value. Failure to deliver appropriately titrated anesthetic dosing may be associated with an increased risk of hemodynamic instability and subsequent arterial hypotension (3).

Likewise, another advantage of EEG monitoring is during total intravenous anesthesia (TIVA), where end-tidal concentrations cannot be measured and dosing errors may potentially increase the risk of unintended light anesthesia, intraoperative awareness or cardiovascular compromise. In such cases, the implementation and interpretation of raw EEG waveforms provides valuable guidance. For example, the presence of continuous isoelectricity with TIVA on the raw EEG suggests cortical suppression and possible anesthetic overdose. Furthermore, the analysis of EEG frequency bands or spectral patterns can assist in estimating anesthetic depth adjusted to age. Therefore, across infancy and childhood, findings of increased activity in the beta and alpha bands on DSA suggest deep sedation whereas a shift toward a slower theta and delta dominance (with reduced beta) is generally consistent with an adequate hypnotic state. Coherent frontal alpha oscillations may further support sufficient hypnotic depth in this age range (3).

In addition, non-proprietary quantitative processed measures, such as the density spectral edge frequency (SEF95), can also aid in assessing hypnotic levels, complementing waveform interpretation. Xu et al. (7) proposed SEF95 reference ranges of 15–20 Hz for sedation, 10–15 Hz for surgical maintenance, and 6–14 Hz during laryngoscopy or surgical incision in patients older than 3 months.

In summary, despite developmental differences that limit the direct application of adult EEG criteria to infants, EEG monitoring remains a valuable tool for providing anesthetic titration. Its integration into pediatric anesthesia practice may support and enhance safety in the clinical practice.

Proposed Exercise Readers may find it helpful to review representative figures in the referenced articles, including Figure 1 in reference 8 (EEG maturation with age) and Figure 2 in open-access reference 10 (age-dependent spectrograms during sevoflurane anesthesia).

REFERENCES

1. Purdon, P. L., Sampson, A., Pavone, K. J., & Brown, E. N. (2015). Clinical electroencephalography for anesthesiologists: Part I: Background and basic signatures. Anesthesiology, 123(4), 937–960. https://doi.org/10.1097/ALN.0000000000000841
2. Davidson, A., & Skowno, J. (2019). Neuromonitoring in paediatric anaesthesia. Current Opinion in Anesthesiology, 32(3), 370–376. https://doi.org/10.1097/ACO.0000000000000732
3. Yuan, I., Xu, T., & Kurth, C. D. (2020). Using electroencephalography (EEG) to guide propofol and sevoflurane dosing in pediatric anesthesia. Anesthesiology Clinics, 38(4), 709–725. https://doi.org/10.1016/j.anclin.2020.06.007
4. Cornelissen, L., Kim, S. E., Purdon, P. L., Brown, E. N., Berde, C. B., & Caruso, T. J. (2019). Intraoperative electroencephalogram burst suppression in children associates with postoperative emergence delirium. Frontiers in Neurology, 10, 189. https://doi.org/10.3389/fneur.2019.00189
5. Han, Y., Miao, M., Li, P., Yang, Y., Zhang, H., Zhang, B., Sun, M., & Zhang, J. (2022). EEG-parameter-guided anesthesia for prevention of emergence delirium in children. Brain Sciences, 12(9), 1195. https://doi.org/10.3390/brainsci12091195
6. Yuan, I., Bong, C. L., & Chao, J. Y. (2024). Intraoperative pediatric electroencephalography monitoring: An updated review. Korean Journal of Anesthesiology, 77(3), 289–305. https://doi.org/10.4097/kja.23843
7. Xu, T., Kurth, C. D., Yuan, I., Vutskits, L., & Zhu, T. (2020). An approach to using pharmacokinetics and electroencephalography for propofol anesthesia for surgery in infants. Pediatric Anesthesia, 30(12), 1299–1307. https://doi.org/10.1111/pan.14021
8. Constant, I., & Sabourdin, N. (2012). The EEG signal: A window on the cortical brain activity. Pediatric Anesthesia, 22(6), 539–552. https://doi.org/10.1111/j.1460-9592.2012.03883.x
9. Lee, J. M., Akeju, O., Terzakis, K., Pavone, K. J., Deng, H., Houle, T. T., Firth, P. G., Shank, E. S., Brown, E. N., & Purdon, P. L. (2017). A prospective study of age-dependent changes in propofol-induced electroencephalogram oscillations in children. Anesthesiology, 127(2), 293–306. https://doi.org/10.1097/ALN.0000000000001717
10. Akeju, O., Pavone, K. J., Thum, J. A., Firth, P. G., Westover, M. B., Puglia, M., Shank, E. S., Brown, E. N., & Purdon, P. L. (2015). Age-dependency of sevoflurane-induced electroencephalogram dynamics in children. British Journal of Anaesthesia, 115(Suppl 1), i66–i76. https://doi.org/10.1093/bja/aev114