Authors: Dragovic SZ et al.
Anesthesia & Analgesia 142(2): 249–260, February 2026, 10.1213/ANE.0000000000007530
This retrospective EEG analysis challenges the “one-size-fits-all” assumption in processed EEG monitoring—that equivalent index targets represent equivalent anesthetic states regardless of drug. The authors compared steady-state frontal EEG under propofol versus volatile anesthetics (sevoflurane/desflurane) using conventional spectral methods and 2 newer approaches: fitting oscillations & one-over-f and variational mode decomposition.
Study Design
• 108 surgical patients
• General anesthesia with either propofol or fluranes
• Frontal EEG analyzed at clinically titrated hypnotic/analgesic levels
• Analyses performed at 2 spectral edge frequency (SEF) ranges: 8–15 Hz and 15–21 Hz
Key Findings
-
Fluranes were similar to each other
Sevoflurane and desflurane demonstrated comparable spectral signatures across analytic approaches. -
Propofol showed consistently higher central oscillatory frequency
Using variational mode decomposition:
• Propofol central frequency ~1.5 Hz higher than fluranes
Propofol: ~10.8 Hz
Flurane: ~9.26 Hz
AUC 0.88
Using fitting oscillations & one-over-f:
• Propofol ~2.0 Hz higher center frequency than fluranes
Propofol: ~10.6 Hz
Flurane: ~8.56 Hz
AUC 0.82
-
Differences in the aperiodic exponent
Propofol showed a lower exponent (~0.26–0.31 Hz−1 difference depending on subgroup), suggesting differences in background “1/f” spectral slope. In some SEF ranges this difference remained significant; in others it did not. -
Robust across SEF subgroups
Differences persisted across both lower and higher SEF ranges, indicating that spectral distinctions were not simply artifacts of depth variation.
Interpretation
These results reinforce that propofol and volatile anesthetics generate distinct EEG spectral architectures—even when titrated to similar processed EEG targets (e.g., BIS-like values).
Traditional EEG interpretation often focuses on alpha oscillations and power distribution. The newer analytic methods separate oscillatory components from the aperiodic 1/f background, allowing more nuanced characterization of anesthetic state.
The implications are substantial:
• Propofol produces higher alpha central frequency patterns.
• Volatile agents demonstrate lower-frequency alpha signatures.
• Differences in the aperiodic exponent suggest differing cortical network dynamics.
Thus, titrating different drugs to the same processed EEG index may not represent identical neurophysiologic states. This may introduce bias in comparative outcome studies and in anesthetic depth interpretation.
Key Points
• Propofol and fluranes generate measurably different EEG spectral patterns.
• Propofol central alpha frequency is approximately 1.5–2 Hz higher.
• Novel analytic techniques (variational mode decomposition; oscillations & one-over-f) sensitively detect these differences.
• Agent-specific EEG signatures challenge universal processed EEG targets.
What You Should Know
For clinicians who use depth monitoring, this study highlights that identical BIS or SEF values may not reflect identical cortical states across agents.
Long term, this work supports development of:
• Agent-specific processed EEG indices
• Drug-adaptive depth monitoring
• More physiologically grounded anesthesia titration
For academic anesthesia and intraoperative monitoring research, separating oscillatory and aperiodic components may unlock more precise mechanistic understanding of anesthetic effects on cortical networks.
This paper pushes the field toward precision neuro-monitoring rather than generic depth targets.
Thank you to Anesthesia & Analgesia for allowing us to summarize and share this article.