Establishment and validation of a 3-month prediction model for poor functional outcomes in patients with acute...

Area of Science:

• Neurology
• Cardiology
• Medical Informatics

Background:

• Cardiogenic cerebral embolism (CCE) is a significant health risk.
• Existing prognostic prediction models for CCE are limited.
• Non-valvular atrial fibrillation (NVAF) is a common cause of CCE.

Purpose of the Study:

• To develop and validate prognostic prediction models for poor functional outcomes at 3 months in acute CCE patients with NVAF.
• To identify key predictors of poor functional outcomes in this patient population.

Main Methods:

• A development cohort of 730 CCE patients was used.
• External validation was performed on regional (118 patients) and time-sequential (63 patients) cohorts.
• Least Absolute Shrinkage and Selection Operator (LASSO) regression was employed for variable selection after addressing missing values with Multiple Imputation by Chained Equations (MICE).

Main Results:

• The N-ABCD2 model, incorporating NIHSS score, Age, BNP, CRP, D-dimer, and antithrombotic medication, was established.
• The model demonstrated strong predictive performance with Area Under the Curve (AUC) values of 0.878 (development), 0.918 (external regional), and 0.839 (external time-sequential).

Conclusions:

• The N-ABCD2 model accurately predicts 3-month poor functional outcomes in CCE patients with NVAF.
• The model's reliance on objective and readily available clinical variables enhances its practical utility.