Together with Dr. Joseph P. Mathew and Dr. Jeffrey Browndyke at Duke, we have participated in several of the first functional neuroimaging studies of postoperative cognitive dysfunction, utilizing resting- and task-based functional magnetic resonance imaging (fMRI) in older (age >60) adults. Our work suggests that dysregulation of the brain’s default mode network (DMN) is associated with the severity of cognitive dysfunction 6 weeks after cardiac surgery in older adults. We have found reduced degree centrality of the posterior cingulate cortex, a major DMN “hub,” suggesting reduced intra-network coherence in patients with postoperative cognitive dysfunction. This DMN “hub” paradoxically demonstrates reduced anticorrelation during cognitive task performance in patients with postoperative cognitive dysfunction, suggesting that poor task performance in these patients may be due, in part, to dynamic brain inter-network dysregulation. For more information on this work, please see https://pubmed.ncbi.nlm.nih.gov/27858963/ and https://pubmed.ncbi.nlm.nih.gov/29164774/ . Additional and ongoing laboratory studies in these cardiac and non-cardiac surgical patients are examining the structural and functional neuroimaging variable associations with perioperative biomarker changes and more granular changes in cognition and mood after surgery.

A large body of work in recent years has described specific electroencephalography (EEG) patterns detectable on the scalp surface in response to distinct anesthetic drugs (such as propofol and the volatile anesthetics, among others). Key work by Patrick Purdon, Ken Solt, Emery Brown and colleagues demonstrated that EEG alpha (i.e. 8-12 Hz) signal amplitude (or power) declines with age in older adults. Our group was then the first to show that older adults have significant variability in alpha oscillation power in response to propofol and the inhaled volatile anesthetics, and that the magnitude of frontal EEG alpha oscillation power correlates with baseline/preoperative cognitive function in older adults. We found this result in two different cohorts using both the raw EEG signal from a proprietary BIS monitor, and the EEG data from the frontal electrodes of a 32 channel EEG electrode array. For more on this work, please see https://pubmed.ncbi.nlm.nih.gov/28533746/.

In parallel with this line of work, we have also examined anesthetic dosage and brain monitoring practices as a function of increasing age. It has long been known that the minimum alveolar concentration (MAC) of the volatile inhaled anesthetics declines by age; our recent systematic literature search and meta-regression analysis demonstrated that MAC declines by 6.47% per decade after 1 year of age, with an R2 value of 0.98 (https://www.sciencedirect.com/science/article/pii/S000709121930755X?via%3Dihub). We hypothesized that clinicians at a major academic medical center would decrease the amount of volatile inhaled anesthetics delivered to older adults by less than this 6.47% decrease per decade, which would result in a relative increase in volatile inhaled anesthetic dosage in older adults. In a study of over 17,000 patients, we found that this hypothesis was correct- older adults received relatively larger anesthetic doses than young and middle aged adults. Yet, we found that a processed EEG measure of so-called “anesthetic depth” (the BIS index) actually increased in older adults, despite the fact that higher anesthetic dosage in older adults would be predicted to result in lower BIS numbers. For more information on this “paradox of age”, please see https://pubmed.ncbi.nlm.nih.gov/31279479/. This work suggests the need for better EEG monitors for use in clinical practice on older adults undergoing general anesthesia, and that more careful anesthetic titration in older adults could reduce dose-dependent side effects of the inhaled volatile anesthetics and propofol.