Intraoperative Frontal Alpha-Band Power Correlates with Preoperative Neurocognitive Function in Older Adults

Abstract

Each year over 16 million older Americans undergo general anesthesia for surgery, and up to 40% develop postoperative delirium and/or cognitive dysfunction (POCD). Delirium and POCD are each associated with decreased quality of life, early retirement, increased 1-year mortality, and long-term cognitive decline. Multiple investigators have thus suggested that anesthesia and surgery place severe stress on the aging brain, and that patients with less ability to withstand this stress will be at increased risk for developing postoperative delirium and POCD. Delirium and POCD risk are increased in patients with lower preoperative cognitive function, yet preoperative cognitive function is not routinely assessed, and no intraoperative physiological predictors have been found that correlate with lower preoperative cognitive function. Since general anesthesia causes alpha-band (8-12 Hz) electroencephalogram (EEG) power to decrease occipitally and increase frontally (known as "anteriorization"), and anesthetic-induced frontal alpha power is reduced in older adults, we hypothesized that lower intraoperative frontal alpha power might correlate with lower preoperative cognitive function. Here, we provide evidence that such a correlation exists, suggesting that lower intraoperative frontal alpha power could be used as a physiological marker to identify older adults with lower preoperative cognitive function. Lower intraoperative frontal alpha power could thus be used to target these at-risk patients for possible therapeutic interventions to help prevent postoperative delirium and POCD, or for increased postoperative monitoring and follow-up. More generally, these results suggest that understanding interindividual differences in how the brain responds to anesthetic drugs can be used as a probe of neurocognitive function (and dysfunction), and might be a useful measure of neurocognitive function in older adults.

Keywords: EEG; aging; alpha oscillations; anteriorization; cognitive function; general anesthesia; isoflurane; propofol.

Figure 1

Spearman correlations between preoperative cognitive index score and intraoperative frontal electroencephalogram (EEG) power in different frequency bands. (A) Electrode locations for 32-channel EEG. Electrodes that are beyond the horizon of the head from this top view appear outside of the circular head schematic. Red denotes the frontal electrode region of interest (ROI; the data for which are plotted in B–E); other ROIs include central (magenta), parietal (blue), and occipital (cyan). (B) Alpha (8–12 Hz) showed a significant correlation with preoperative cognitive index score, while (C–E) delta (1–4 Hz), theta (4–8 Hz), and beta (12–30 Hz) did not.

Figure 2

Scalp topography of alpha power and correlation with cognitive index scores. (A) Spearman correlation coefficients for comparisons between preoperative cognitive index score and intraoperative alpha power at each electrode site. (B) Alpha power under general anesthesia.

Figure 3

Spearman correlation between preoperative cognitive index score and intraoperative alpha power measured by the bispectral index (BIS) monitor in a separate sample of 35 patients. The BIS alpha power shown here is lower than the 32-channel data alpha power (Figure 1B), most likely due to the proximity of the reference electrode for the BIS montage (also on the forehead) (For the 32-channel data, we referenced to the algebraic average of the mastoid electrodes).