Viable Prediction for Atrial Fibrillation Recurrence After Catheter Ablation

How to Cite

Baljepally, V., Raffa, J., & Zhao, X. (2021). Viable Prediction for Atrial Fibrillation Recurrence After Catheter Ablation. Vanderbilt Undergraduate Research Journal, 11(1).


Introduction: Atrial fibrillation (AF) is the most common heart rhythm abnormality and the leading cause of stroke. Radiofrequency catheter ablation is used to treat AF but recurrence can occur after the ablation procedure, requiring repeat procedure. A new model to predict AF recurrence after ablation was developed through multivariate analysis.

Methods: The variables include demographic, electrocardiographic, echocardiographic, and clinical parameters. In a retrospective review (n=82), 41 patients who underwent repeat ablation for recurrent AF were compared to 41 controls that underwent ablation only once.

Results: Of the analyzed parameters, age, female gender and left atrial enlargement were not predictive, but P wave duration (PWD) and obstructive sleep apnea (OSA) were significant predictors of repeat ablation (p-value = 0.0003 and 0.0023, respectively). Based on the analyses, a simple decision tree model was developed, achieving a prediction accuracy of 87% (sensitivity=83%, specificity=90%).

Conclusion: The developed PWD and OSA 2-predictor model has good accuracy and sensitivity, both of which make it a viable prediction model for AF recurrence after catheter ablation. The developed model will help doctors: (1) Avoid repeat procedure in patients at high risk of recurrence by exploring alternative treatments (2) Reduce costs by avoiding repeat procedure (3) Correct underlying issues prior to procedure in those at high risk (4) Objectively inform patients about recurrence so they can make an informed decision about whether to undergo the procedure. Adopting predictive models such as ours may therefore improve quality care and reduce costs for AF patients undergoing ablation.
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Copyright (c) 2021 Array