Frailty as a mediator of postoperative delirium in older adults undergoing orthopedic surgery: A causal mediation analysis

Background: Postoperative delirium has a high prevalence in hospitalized older adults. Frail older adults have an increased risk for developing it. Preoperative pain-a common symptom in older adults with orthopedic conditions-shows a connection with frailty through an inflammatory process. However, an association among preoperative pain, frailty, and postoperative delirium remains unexplored among older adults undergoing orthopedic surgery. Objectives: To examine the magnitude, direction, and mediation effect of frailty on the association between preoperative pain and postoperative delirium among older adults undergoing orthopedic surgery. Design: A secondary analysis of a prospective cohort study. Setting and participants: A cohort of 200 older adults (60 years and older) who underwent major orthopedic surgery at a university hospital in Thailand was recruited for the study. Methods: Participants responded to the Demographic and Health Information Form, the Reported Edmonton Frailty Scale-Thai version, and the Thai version of the 4 A's test. Analyses were conducted using a causal mediation analysis with 95 % confidence intervals. Results: We found that 12.5 % of participants developed postoperative delirium. For the direct paths associated with postoperative delirium, statistically significant effects were observed for frailty, preoperative pain, and comorbidity. Considering frailty as a mediator of postoperative delirium, a significant positive indirect effect was identified from preoperative pain. Controlling for age and comorbidities, frailty mediated the association between preoperative pain and postoperative delirium in a statistically significant manner; the average direct effect was 0.014 (95 % confidence interval: 0.008-0.020), the average causal mediating effect was 0.002 (95 % confidence interval: 0.000-0.010), the total effects was 0.017 (95 % confidence interval: 0.010-0.020), and the probability mediation accounted for 14 % (95 % confidence interval: 0.031-0.300). Conclusion: Integrating the concept of age-related decline and frailty assessment may offer opportunities to provide disease-specific care and strengthen precision perioperative care, which ultimately enhance quality of life in older adults undergoing orthopedic surgery.