Development and validation of a LASSO-derived prognostic model predicting postoperative cognitive dysfunction risk in off-pump coronary artery bypass grafting patients

Read full text Summarise with VIRAT
Authors: Zehua Zhang, Xiaoyu Xu, W C Peng, Yifan Wang, Yang Lin
Cited by: 0
References: 51

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Background: Middle-aged and elderly patients face an elevated risk of postoperative cognitive dysfunction (POCD) after coronary artery bypass grafting (CABG), a complication that worsens prognosis. While off-pump CABG (OPCABG) reduces some injury mechanisms, no prediction model has been developed for predicting POCD in this surgical population. This study aimed to develop a predictive model for POCD following OPCABG to facilitate the earlier and more precise identification of high-risk patients. Methods: This prospective single-center study enrolled 412 patients who underwent elective OPCABG at Beijing Anzhen Hospital between March 2024 and March 2025. Patients were randomly allocated in a 7:3 ratio to a training set (n=288) and a testing set (n=124). Variable selection was performed using least absolute shrinkage and selection operator (LASSO) regression and multivariable logistic regression on the training set. A nomogram was subsequently developed using the chosen variables. The predictive performance of the model was assessed in both sets using the area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, calibration curves, and decision curve analysis (DCA). Results: POCD was observed in 174 out of 412 patients (42.2%). Predictive factors included age, body mass index (BMI), blood glucose, total cholesterol (TC), smoking history, and history of prior silent cerebrovascular disease (SCVD). The model demonstrated good diagnostic performance, with area under the curve (AUC) values of 0.722 [95% confidence interval (CI): 0.663-0.780] in the training set and 0.704 (95% CI: 0.609-0.800) in the testing set. Calibration curves showed close agreement between the predictions of the nomogram and the observed outcomes. DCA for both datasets suggested favorable clinical utility of the model. Conclusions: This study presents a LASSO-derived predictive model for the estimation of the risk of POCD in patients undergoing OPCABG.

Fields of Study & Subjects

Cardiac and Coronary Surgery Techniques Intensive Care Unit Cognitive Disorders Cardiac Anesthesia and Surgical Outcomes

Publication details

Journal / source
Journal of Thoracic Disease
Volume
18
Issue
3
Pages
220–220
Publication date
1 March 2026
Language
EN