Impact of circadian rhythm change on postoperative delirium in elderly patients undergoing lower extremity orthopedic surgery: a propensity score-matched prospective cohort study

Background Postoperative delirium (POD) is a common and clinically significant complication associated with numerous adverse outcomes, including prolonged hospitalization, functional decline, and increased mortality. Older adults undergoing lower extremity orthopedic surgery constitute a particularly high-risk population for POD. Previous studies have reported an incidence of POD as high as 34.2% among elderly patients undergoing hip surgery receiving general anesthesia. Although some research has examined the influence of circadian rhythm on cognitive function, the comparative impact of perioperative circadian rhythm change on POD incidence in elderly patients undergoing lower extremity orthopedic procedures remains unexplored. This study therefore undertakes a prospective cohort design to investigate the association between perioperative circadian rhythm change and the risk of POD in this vulnerable patient group. Methods Eligible participants were patients aged ≥60 years, of any gender, with American Society of Anesthesiologists (ASA) physical status classification I–III, who underwent elective lower extremity orthopedic surgery under general anesthesia at our institution. Patients were categorized into two groups based on perioperative stability of circadian rhythm: the circadian rhythm change group (Group C) and the no circadian rhythm change group (Group NC). To minimize baseline imbalances between groups, 1:1 propensity score matching (PSM) was performed using a nearest-neighbor algorithm with a caliper width of 0.05. Following matching, multivariable logistic regression was used to assess the association between circadian rhythm change and the incidence of postoperative delirium (POD), while linear regression was employed to evaluate its effect on length of hospital stay. Subgroup analyses were further conducted to explore potential effect modification and to address residual confounding within the matched cohort. Results A total of 277 patients met the inclusion criteria and were included in the analysis between August 2024 and January 2025. Following propensity score matching, POD occurred in 37 patients (34.58%) in the Group C and 20 patients (18.69%) in the Group NC. After identifying independent variables potentially associated with POD and adjusting for confounders using binary logistic regression, we found that perioperative circadian rhythm change was independently associated with a higher risk of POD (OR = 2.26; p = 0.02). Additionally, multiple linear regression analysis revealed that circadian rhythm change was significantly associated with a longer hospital length of stay. Subgroup analyses suggested a stronger association between circadian rhythm change and POD in patients