Machine Learning-Enhanced TCAB Score for Predicting Postoperative Ischemic Stroke After CABG.

Area of Science:

• Neurology
• Cardiovascular Surgery
• Medical Imaging

Background:

• Postoperative acute ischemic stroke (AIS) is a significant risk following coronary artery bypass grafting (CABG).
• Predictive tools for AIS post-CABG are crucial for patient management and improving outcomes.

Purpose of the Study:

• To develop and validate a novel Total Cerebral Atherosclerosis Burden (TCAB) score.
• To assess the predictive capability of the TCAB score for AIS after CABG.

Main Methods:

• A prospective cohort of 909 patients undergoing CABG was enrolled.
• The TCAB score was calculated by quantifying stenosis severity across intracranial and extracranial arteries.
• Multivariable logistic regression and gradient boosting machine models were employed to evaluate TCAB's association with ischemic stroke and major adverse cardiovascular and cerebrovascular events.

Main Results:

• Patients with in-hospital AIS had a significantly higher mean TCAB score (8 vs. 2, P < 0.001).
• A TCAB score >3 predicted in-hospital AIS with an AUC of 0.756.
• Higher TCAB scores were independently associated with increased risk of in-hospital AIS, 1-year AIS, and 1-year MACE, with gradient boosting models showing strong predictive ability (AUCs ranging from 0.7475 to 0.8736).

Conclusions:

• The TCAB score, particularly when enhanced by machine learning, demonstrates robust predictive power for short-term and 1-year cerebrovascular and cardiovascular events post-CABG.
• The TCAB score serves as a practical tool for guiding preoperative revascularization decisions and intraoperative embolic protection strategies.