Plasma Amyloid-β 1-40 and Postoperative Delirium After Cardiac Surgery

Importance: Postoperative delirium (POD) is a frequent and severe complication after cardiac surgery that is associated with increased risk of dementia and adverse outcomes. Early identification of high-risk patients remains challenging. Objective: To examine whether preoperative plasma Alzheimer disease biomarkers are associated with baseline cognition and can independently estimate POD in patients undergoing elective cardiac surgery. Design, Setting, and Participants: This prospective, observational cohort study (FIND Delirium Risk Factors [FINDERI]) included patients 50 years or older undergoing elective cardiac surgery at a comprehensive cardiac surgery center in Germany between February 1, 2021, and October 31, 2022. Statistical analysis was performed from January to December 2025. Exposures: Preoperative plasma concentrations of amyloid-β 1-40 (Aβ1-40), amyloid-β 42 (Aβ1-42), phosphorylated tau 181 (pTau181), and phosphorylated tau 217 (pTau217) were measured using validated immunoassays between November 1, 2022, and July 31, 2024. Main Outcomes and Measures: The primary outcome was POD, assessed daily for 5 days postoperatively using the Confusion Assessment Method algorithm. Baseline cognitive function was assessed using the Montreal Cognitive Assessment (MoCA). Risk estimation accuracy was evaluated using receiver operating characteristic analysis and multivariate logistic regression, adjusting for relevant clinical covariates. Results: Of 504 enrolled patients, 491 completed POD assessment (mean [SD] age, 68.4 [8.3] years; 385 [78.4%] male), of whom 106 (21.6%) developed POD. In multiple linear regression, pTau217 (β = -0.51; 95% CI, -0.89 to -0.12; P = .01) and age (β = -0.12; 95% CI, -0.16 to -0.08; P < .001) remained significant factors associated with lower MoCA scores. Patients who developed POD had higher preoperative Aβ1-40, Aβ1-42, pTau217, and Aβ1-40/Aβ1-42 × pTau217, and lower Aβ1-42/Aβ1-40. Aβ1-40 was the independent biomarker most strongly associated with POD (area under the receiver operating curve [AUC], 0.65; 95% CI, 0.59-0.71; P < .001). Combining biomarkers with clinical factors improved risk estimation (AUC, 0.74; 95% CI, 0.69-0.79; P < .001) compared with clinical factors alone (AUC, 0.73; 95% CI, 0.67-0.78; P < .001). A combined model including MoCA achieved an AUC of 0.77 (95% CI, 0.72-0.82; P < .001), driven primarily by Aβ1-40. Conclusions and Relevance: In this prospective cohort study, preoperative Aβ1-40 showed the strongest association with POD risk among the measured biomarkers. These findings suggest that preoperative plasma Aβ1-40 may contribute to POD risk stratification, but further research is needed to validate these findings and determine clinical utility.