Predictors of Atrial Fibrillation in Heart Failure Patients with Indications for ICD Implantation

Background/Objectives: Atrial fibrillation (AF) is a prevalent arrhythmia that significantly complicates the management of heart failure (HF) patients, particularly those who have implantable cardioverter-defibrillators (ICDs). The interplay between AF and inappropriate ICD therapy poses a critical challenge in optimizing patient outcomes, as inappropriate shocks can lead to increased morbidity, psychological distress, and a reduced quality of life. We aimed to explore the various clinical and demographic predictors of AF in HF patients with indications for ICD implantation. Methods: This study included 122 patients who were indicated for ICD implantation and had undergone transthoracic echocardiography (TE). We evaluated the relationships between clinical and demographic factors and the occurrence of AF, which was recorded either before ICD implantation or during the follow-up period afterward. From our findings, we established predictors and a risk model for AF. Results: Out of 122 HF patients with ICDs, 52 (42.6%) experienced an episode of AF either prior to ICD implantation or during a follow-up period of 20.5 [6.0; 53.0] months, as recorded by the ICDs' endogram. Patients with AF were older compared to those without AF (p < 0.001). Additionally, they exhibited a higher left ventricular early diastolic filling rate (LVE) (p = 0.006) and a greater left atrial index (LAI) (p = 0.002). These three factors-age, LVE and LAI-were found to be independently associated with AF in both univariable and multivariable logistic regression analyses. The final model, including age, LVE, and LAI, showed a good discrimination capability with an AUC of 0.775. At a cutoff value of >0.47, the model achieved a sensitivity of 67.3% and a specificity of 77.2% in identifying HF patients with ICDs at risk for AF. Conclusions: This study found that 42.6% of HF patients with ICDs experienced AF, with older age, higher LVE, and greater LAI identified as significant predictors.