Development and Validation of a Model to Predict Secondary Arrhythmia in Patients With Epilepsy

Area of Science:

• Cardiology
• Neurology
• Medical Informatics

Background:

• Epilepsy patients exhibit increased susceptibility to arrhythmias, potentially worsening prognosis.
• Early identification of arrhythmia risk in epilepsy is crucial for timely intervention.
• A clinical prognostic model was developed to address this unmet clinical need.

Purpose of the Study:

• To develop and validate a clinical prediction model for assessing arrhythmia comorbidity risk in epilepsy patients.
• To facilitate early clinical intervention and improve patient outcomes.
• To provide clinicians with a tool for risk stratification.

Main Methods:

• Retrospective collection of clinical data from 495 epilepsy patients (January 2022 - February 2025).
• Development and validation datasets created using a 7:3 ratio split.
• Logistic regression model constructed using LASSO regression for variable selection; model performance evaluated using AUC, C-index, calibration curves, and decision curve analysis.

Main Results:

• The predictive model demonstrated good discriminative ability with an overall C-index of 0.752 (95% CI: 0.701-0.804).
• Sensitivity and specificity were 74.6% and 68.1%, respectively.
• A nomogram was developed for visual representation of the predictive model.

Conclusions:

• The developed predictive model accurately assesses arrhythmia risk in epilepsy patients.
• The model aids clinicians in early detection and intervention strategies.
• Implementation of this model can potentially improve the prognosis of epilepsy patients with comorbid arrhythmias.