Inflammation, intrinsic vulnerabilities and postoperative delirium in older people
Publication Date
July 23, 2025
Creator
Abstract
Postoperative delirium (POD), an acute and fluctuating confusional state, has the highest incidence (up to 50%) in older (i.e. age ≥ 65 years) people recovering from moderate/major surgery. This is significant in light of the life altering sequelae of POD, e.g. subsequent cognitive impairment, new long-term institutionalisation and increased mortality. While a large body of work has gone into understanding the aetiology of POD, there remains a lot of uncertainty.
Currently, recognised risk factors for developing POD include old age, cognitive impairment, dementia, functional impairment, hearing and/or visual difficulties, comorbidities and severity of illness. There is also a plethora of hypotheses regarding the aetiology of POD. While the neuroinflammatory hypothesis appears to have gained a large following, oxidative stress hypothesis, neurotransmitter hypothesis and neuronal aging hypothesis also have their merits.
These theories point to intrinsic vulnerabilities in different areas. In this work, the argument is made for POD being a symptom of the summation of intrinsic vulnerabilities in diverse areas. However, the individual contributions of each domain need to be measurable, preferably readily and at a reasonable cost.
The aim of this work was to identify and test reliable and readily measurable biomarkers of inflammatory response to surgery and intrinsic brain vulnerability in the context of POD in cognitively intact older people. Concerning intrinsic brain vulnerability, we made the case for preclinical neurodegeneration as a potential cause.
On inflammatory response, our systematic review identified an association between preoperative interleukin 6 (IL-6) and postoperative delirium. Our observational study went further by identifying IL-6 as a fair predictor of POD with an area under the curve (AUC) of 0.753. However, the change in cytokines across surgery was not predictive of POD, nor was the preoperative insulin-like growth factor 3 (IGF-3). Exploration of interactions between key inflammatory mediators and relevant cytokines, soluble receptor or binding protein was also not predictive of POD.
Concerning intrinsic brain vulnerability, a search through the literature on dementia identified two associated features that may be readily measured, i.e. retinal nerve fibre layer (RNFL) thickness measured using optical coherence tomography and resting state posterior dominant rhythm (PDR) on electroencephalography (EEG). RNFL is thinner in people with dementia compared with healthy controls and PDR has lower power in people with cognitive impairment. The systematic review in this work suggests that thinning of the RNFL is associated with mild cognitive impairment (MCI), particularly amnestic mild cognitive impairment. Given our premise of preclinical neurodegeneration contributing to the risk of POD, two studies were conducted exploring RNFL thickness and PDR in cognitively intact older surgical patients. Our study on optical coherence tomography in older hip fracture patients suggests that the RNFL is thinner in people who develop POD compared to those who do not develop POD. Our study on resting state posterior EEG power demonstrated feasibility of acquiring raw EEG data using a processed EEG monitor.
However, there was no significant difference in posterior dominant rhythm between people who had POD and people who did not experience POD.
Identification of measurable and accessible biomarkers of the individual vulnerabilities is critical to unravelling POD in older people.
Item Type
ethesis
Thesis Type
PhD
Supervisors
Subjects (LC)
Associated Schools / Departments
School of Medicine
eprints ID
79752
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
Noah-14342735-thesis Final.pdf
Type
Full-text
Description
Examined. Final version of Thesis on Inflammation, Intrinsic Vulnerabilities and postoperative delirium in older people
Size
4.28 MB
Format
Adobe PDF
Checksum (MD5)
c6ccef57887cd5dc7afffa96c7cd6b6b