Towards Neurophysiological Biomarkers of Brain Vulnerability: Frontal Alpha Waves Predict the Propensity for Intraoperative Burst Suppression

Abstract

BACKGROUNDS: Neurocognitive complications of surgery and anesthesia are common. Recent studies show that electroencephalography (EEG) burst suppression during general anesthesia predicts postoperative delirium (POD) and impairment of functional independence. A prominent feature of EEG under propofol and sevoflurane anesthesia is alpha oscillation. Anatomical evidence exists that generators of alpha waves overlap significantly with brain regions vulnerable in aging and neurodegeneration. We hypothesize that alpha power under anesthesia reflects the degree of brain vulnerability and therefore correlates with the propensity for intraoperative burst suppression and postoperative neurocognitive impairment. METHODS: We analyzed EEG data from a previously reported cohort in which 155 patients received propofol (n=60) or sevoflurane (n=95) as the primary anesthetic. We estimated the EEG spectrum from a 2-minute period of stable anesthetic maintenance. Burst suppression is analyzed within a window from 10 minutes after induction to the end of the procedure, and it is defined operationally by the presence of at least three consecutive suppression events within a 1-minute period. We characterized the relationship between burst suppression and various factors (including alpha power, age, etc) using logistic regression analyses. RESULTS: Consistent with a desirable predictor of postoperative neurocognitive outcomes, alpha oscillations exhibit both age-dependent and age-independent variabilities. We tested three physiologically based models: 1) a Base Model with alpha power as the only predictor; 2) an Extended Mechanistic Model with additional predictor variables, including age, the primary anesthetic drug and its maintenance dose, and the rate of intraoperative propofol boluses; 3) a Clinical Confounding Factor Model, which included gender, intraoperative hypotension, and the dose of perioperative midazolam. In all three models, alpha power is a strong predictor of burst suppression, with decreased alpha power associated with increased probability of burst suppression. Based on the Extended Mechanistic Model (best model assessed by Receiver Operating Characteristic and Akaike Information Criterion analyses), after adjusting for age and other relevant factors, for each decibel decrease in alpha power, the odds of experiencing burst suppression increase by 25% (1.25 fold, 95% CI (1.09, 1.44)). If we compare an individual at the 25th percentile vs. an individual at the 75th percentile of alpha power, the odds ratio of experiencing burst suppression is 3.31 (95% CI (1.57, 7.00)). CONCLUSIONS: Anesthesia-induced frontal alpha oscillations capture both age-dependent and age-independent variabilities in burst suppression. Alpha power, independent of age, is a strong predictor of burst suppression, and may help identify patients with lower anesthetic requirements and higher risk of postoperative neurocognitive impairment. We hypothesize that alpha oscillations could serve as a neurophysiological biomarker for brain vulnerability under general anesthesia.