Predicting and Mitigating Risk of Postoperative Cognitive Dysfunction: Is a Major Breakthrough on the Horizon?

Postoperative cognitive dysfunction (POCD) – also referred to as perioperative neurocognitive disorders – and delirium are pathologic states that may develop after surgery and often result in distress to the patient, family members, medical practitioners, and the health care system. Patients who suffer from POCD may experience short- or long-term physical or emotional disturbances. These may require supportive care such as home health aides or, in extreme cases, institutionalization in a nursing facility. Given the potential for significant financial and health care-related consequences, there has been great interest regarding research into POCD. Such research includes inquest of contributing factors, strategies to mitigate risk, and (importantly) the ability to predict increased risks for patients undergoing a surgical procedure under anesthesia.

Past studies have shown that both frailty and preexisting cognitive impairment are contributing risk factors toward the development of POCD. Nonetheless, patients who are neither frail nor explicitly cognitively impaired may still develop POCD, and the ability to stratify risk for these patients is a potential game-changer for surgical and perioperative planning. For example, a patient booked for an elective hernia repair may reconsider surgical options if they are revealed to be at high risk of POCD. Conversely, a patient with a pending joint replacement might be reassured about the low risk of POCD perioperatively, allowing them the ability to resume daily activities without assistance.

One promising area of research includes identifying biomarkers obtained in a patient during a preoperative workup. A recent investigation by Leung et al. assessed clinically silent neurodegenerative changes and their potential to cause postoperative delirium (Anesthesiology 2023;136:432-43). These changes were represented by plasma neurofilament light chain biomarkers, which were found to be significantly higher in patients who developed postoperative cognitive decline. Biomarkers have been more widely used to predict dementia and cognitive impairment unrelated to anesthetic care. Kivisakk et al. demonstrated that plasma biomarkers NF-L and PTP-1B were significantly higher in patients with dementia when compared to cognitively normal patients (Brain Commun 2022;4:fcac155). These plasma components indicate inflammation and vascular injury in the brain, which may be the underlying etiology of cognitive impairment.

An important consideration moving forward may be to determine if long-term cognitive decline as a consequence of aging in conditions such as Alzheimer’s disease has the same underlying pathophysiology as POCD. This may be an important question to answer given the proliferation and interest in using commercially available laboratory exams to predict the development of Alzheimer’s-related dementia in the elderly population. One such exam is the PrecivityAD blood test, available from C2N Diagnostics, which works to identify brain amyloid plaque status (determined by quantitative amyloid PET scans) with remarkable accuracy. This allows clinicians to potentially predict the likelihood of developing dementia, especially useful in patients with no underlying neurologic dysfunction at the time of the exam. This blood test measures amyloid beta 40 and 42, as well as apolipoprotein E (ApoE), and calculates an Amyloid Probability Score (APS). The Amyloid Probability Score predicts the likelihood of future development of consequential dementia. These tests continue to improve in accuracy as well, with a recent study demonstrating that adding yet another biomarker, plasma p-tau217/np-tau217, further improves their predictive algorithm (Alzheimers Dement 2024;20:1214-24).

The prediction of risk for POCD may ultimately play a vital role in perioperative planning for anesthesiologists. Variations in anesthetic technique and choice of medications may contribute to postoperative delirium. For example, recent investigation by Zhu et al. showed that during spinal anesthetic for hip surgery, dexmedetomidine was shown to have a lower incidence of delirium when compared to propofol for intraoperative sedation (Drug Des Devel Ther 2023;17:3845-54).

There has also been great interest in the use of various perioperative monitoring modalities to attempt to predict perioperative risk of cognitive dysfunction. For example, a team at the University of Michigan attempted to use electroencephalography (EEG) and cerebral oximetry to predict decline in cognitive function following noncardiac surgery (Anesthesiology 2023;139:568-79). However, their team uncovered no correlation between posterior alpha power, frontal-parietal connectivity, and cerebral oximetry with patients’ cognitive function, measured at three months postoperatively. Despite the lack of conclusive evidence supporting their hypothesis, this investigative team raises the crucial point that there is no standard for perioperative brain monitoring.

Further study in the perioperative use of the EEG suggests that it may be useful as a predictor of postoperative delirium severity. Acker et al. recently reported an inverse correlation relationship between postoperative delirium severity and preoperative alpha-attenuation magnitude – a measure of inattention (BR J Anaesth 2024;132:154-63). It is unclear if further study in this field will yield correlation between EEG or cerebral oximetry measurements in the perioperative period and postoperative neurologic dysfunction, but it is likely an area that requires further study.

Surgery-induced neuroinflammation could be a possible pathogenic mechanism for POCD. Inflammation, in association with neurotransmitter imbalance, altered biologic rhythms and metabolism, as well as impaired neuronal network connectivity, likely contributes to POCD (Medicine 2023;102:e32991). It is important to note that theories that associate inflammation with POCD assert that surgical inflammation is the catalyst that induces central processes affecting cognition. Xioa et al. report that this proposed pathway ultimately results in release of reactive oxygen species (ROS), nitric oxide, IL-6, IL-B, and other inflammatory mediators. The role of anesthesia in this pathway is indeterminate. Nonetheless, inflammation leading to POCD likely depends upon disruption of the blood brain barrier, activation of glial cells, dysfunction of the glymphatic system, and alterations in neuronal and synaptic function. Conditions of a patient under anesthesia may contribute to alterations in these patient characteristics, potentiating the pathway of neuroinflammation.

Though the exact mechanism that causes POCD remains elusive, it is an exciting area for current and future research. At Rutgers New Jersey Medical School, for example, a study was conducted in collaboration with C2N Diagnostics to determine how the aforementioned PrecivityAD test may be employed as part of a strategy to predict POCD. This pilot study involved correlating concentrations of biomarkers (amyloid beta 40/42 and Amyloid Probability Score) with neurological assessments. Though this pilot study yielded no conclusive results (likely due to the limited size of study cohort and lack of resources to conduct proper neurocognitive testing), it has shown promise as a platform for future investigations involving the use of a biomarker blood test in attempting to diagnose POCD. Given the great interest in predicting and mitigating POCD perioperatively, investigations will continue to seek the identification of a POCD pathway, a methodology to predict POCD risk, and strategies to mitigate risk associated with development of this potentially devastating condition.