Protective factors influence resilience to delirium after elective orthopaedic surgery

Objectives
Delirium is a non-specific syndrome which presents as acute and transient alteration in cognition, consciousness, attention, or orientation. The incidence of delirium after total joint surgery in adults is 17%, and has been associated with poor outcomes including increased risk of mortality, new diagnosis of dementia and institutionalization. Whilst an increasing number of, mostly observational, studies have explored risk factors for postoperative delirium, none to date have considered resilience to delirium as an outcome, nor have they defined delirium resilience.
This study aimed to test the hypothesis that there are ‘protective factors’ that may provide higher levels of resilience to delirium after elective arthroplasty surgery.
Methods
This observational cohort study recruited participants aged 65 years and over, without dementia, undergoing elective hip or knee arthroplasty. The study was performed in accordance with local ethical committee procedures and all participants gave informed written consent (REC reference: 10/NIR01/5; protocol number: 09069PP-OPMS). Baseline characteristics were recorded, and neuropsychological tests completed. Blood and cerebrospinal fluid (CSF) samples were collected perioperatively. Post-operatively, participants were assessed daily for delirium using the Confusion Assessment Method (CAM), supported by the Mini Mental State Examination (MMSE), and nursing staff enquiries
‘Resilient’ participants were defined as having a preoperative MMSE score above 24, not losing points across any MMSE sections post-operatively, and never becoming delirious as assessed using the CAM. Factors associated with resilience were identified using logistic regression. Variables included in the logistic regression model were included based on statistical significance or clinical significance. Included variables are displayed in Table 1.
Results
292 participants were included in this analysis, and 78 participants were classed as ‘resilient’ to delirium. Results of this logistic regression analysis are shown in Table 1. Lower age (OR= 0.899, CI: 0.829, 0.974, p=0.009), higher score in the National Adults Reading Test (NART) (OR= 1.065, CI: 1.023, 1.110, p=0.002), lower preoperative pain levels (OR= 0.978, CI: 0.961, 0.995, p=0.011), and lower concentration of CSF T-tau (OR= 0.996, CI: 0.992, 1.000, p=0.031) were significantly associated with resilience to delirium.
Conclusions
This study found younger age, higher NART score, lower pain score on movement and lower concentration of CSF T-tau to be independently associated with delirium resilience following elective arthroplasty in an older population without prior dementia diagnosis. 78 of the included participants were resilient to delirium, and 214 were non-resilient, highlighting the large proportion of patients experiencing some cognitive change after surgery, even if the change was minimal. To our knowledge, the ‘delirium resilient’ phenotype is yet to be explored, therefore these findings are the first to indicate variables which may provide resilience after elective surgery. Further analyses of delirium cohorts with resilience as an outcome are required to validate these ‘protective factors’ and explore their potential application to clinical settings. Investigation of the delirium resilience phenotype may expand field knowledge of the condition, while enabling patients and care providers to map risk of delirium.