Relationship between the preoperative AST/ALT ratio and postoperative delirium and 3-year mortality in patients undergoing total knee or hip arthroplasty: a prospective cohort study.

objectiveTo investigate the association between the preoperative serum aspartate aminotransferase to alanine aminotransferase (AST/ALT) ratio and postoperative delirium (POD) and 3-year mortality in POD patients.

methodsA total of 538 clinical participants were enrolled from the Perioperative Neurocognitive Disorder and Biomarkers Lifestyle (PNDABLE) study. In this study, patients were first categorized into a POD group and a non-POD group, and the demographic characteristics of the two groups were compared. Preoperative serum AST and ALT levels were measured to calculate the AST/ALT ratio. Cerebrospinal fluid (CSF) Alzheimer's disease (AD)-related biomarkers were analyzed. Logistic regression and sensitivity analyses were performed to identify protective and risk factors for POD. Mediation effect models were applied to evaluate the potential mediating roles of CSF biomarkers. Receiver operating characteristic curves and decision curve analysis were used to validate predictive performance. Restricted cubic spline (RCS) regression was employed to explore the dose-response relationship between AST/ALT levels and POD risk. Finally, patients with POD were followed up for 3 years, and Kaplan-Meier (K-M) survival analysis was used to compare the mortality rates of the AST/ALT ratio in patients with POD.

resultsThe incidence of POD was 17.53%. Logistic regression revealed that an elevated preoperative AST/ALT ratio was an independent risk factor for POD [odds ratio (OR) = 4.305, 95% confidence interval (CI) 2.404-7.706, P < 0.001]. Reduced CSF amyloid-beta 42 (Aβ42) levels (OR = 0.992, 95% CI 0.990-0.994, P < 0.001) were inversely associated with POD risk, whereas elevated total tau (OR = 1.016, 95% CI 1.012-1.019, P < 0.001) and phosphorylated tau (P-tau) (OR = 1.097, 95% CI 1.069-1.127, P < 0.001) were identified as risk factors. Sensitivity and post hoc analyses supported these findings. Mediation analysis demonstrated that the effect of AST/ALT on POD was partially mediated by CSF Aβ42 (12.9%) and the Aβ42/P-tau ratio (14.8%). The predictive model combining AST/ALT with CSF biomarkers achieved optimal performance (area under the curve = 0.92). A follow-up of 88 patients showed that K-M survival analysis revealed that although mortality rates were higher in patients with POD, there were no significant differences in 3-year survival rates between the high AST/ALT ratio group and the low AST/ALT ratio group.

conclusionPreoperative elevation of the serum AST/ALT ratio is a potential risk factor for POD, and its effect may be partially mediated by AD-related CSF biomarkers. However, an increase in the AST/ALT ratio did not have a significant effect on the three-year survival rate of POD patients.