Examining the Link Between Inflammation and Post-Operative Delirium in a Hip Fracture Repair Cohort of Older Adults

Background Advancing age and neurodegenerative disease are risk factors for postoperative delirium; both are linked to dysregulation of the immune system and imbalances in inflammatory mediators in the brain. One such mediator, Tumor Necrosis Factor-alpha Receptor 1 (TNFR1), is a marker of chronic inflammation. Delirium frequently affects older surgical patients and is associated with cognitive decline. While the pathophysiology of delirium is poorly understood, inflammation likely contributes to its development. Objective We hypothesized that older adults undergoing hip fracture repair with higher chronic inflammatory profiles (measured by Tumor Necrosis Factor-alpha Receptor 1 (TNFR1) levels in CSF and plasma) would be at increased risk of delirium after surgery and cognitive decline. Methods We completed a secondary analysis of data from the Randomized clinical trial STRIDE (A-Strategy-to Reduce-the-Incidence-of-Postoperative-Delirium-in-Elderly-Patients) and analyzed preoperative TNFR1 levels. Delirium and delirium severity (preoperatively and days 1-5 postoperatively) were assessed, and cognition (Mini-Mental-State-Examination [MMSE] and Clinical-Dementia-Rating-sum-of-boxes [CDR-SB]) preoperatively and at 12 months to measure cognitive changes. We used logistic regression to model delirium and linear regression to model delirium severity as functions of TNFR1 levels, and random-intercept linear spline models to examine relationships between TNFR1 and cognitive changes. Results Postoperative delirium developed in 37% of patients. In unadjusted models, a 1-SD difference in plasma TNFR1 levels was positively associated with incident delirium (OR 1.40, 95%CI [1.02, 1.93]), while a 1-SD difference in preoperative CSF TNFR1 levels was not statistically associated with delirium (OR 1.22, 95% CI [0.90, 1.66]). In adjusted analyses, preoperative plasma TNFR1 was not significantly associated with delirium or delirium severity (β =0.001, p=0.411). No consistent relationship was observed between preoperative CSF or plasma TNFR1 levels and cognitive changes. Conclusion We observed an association between preoperative inflammation, as measured by plasma TNFR1 levels, and postoperative delirium; this was attenuated in adjusted analyses. Future studies exploring mechanisms associated with delirium are needed as this may uncover targets for management.