Investigation of predictive parameters for recurrence in patients undergoing electrical cardioversion for atrial fibrillation

Background

Electrical cardioversion (ECV) is a technique used to restore sinus rhythm in atrial fibrillation (AF), aiming to reduce AF-related symptoms. Various recurrence predictors, such as non-paroxysmal pattern, prolonged duration, chronic obstructive pulmonary disease, advanced age, or left atrial dilation, have been identified. Hence, the aim of our study was to determine the parameters that could predict AF recurrence in the long term in patients who underwent ECV.

Methods

One hundred seventy-two patients who underwent electrical cardioversion due to AF at our tertiary center were included in the study. Patients with recurrent AF after cardioversion were defined as the AF recurrence (+) group, and those with normal sinus rhythm were defined as the AF recurrence (−) group. Parameters that would predict AF recurrence were investigated between these two groups.

Results

AF recurrence was observed in 91 patients, while 81 patients did not experience any recurrence. The prediction model, constructed using multivariate Cox regression analysis, included six variables: non- high-density lipoprotein (HDL) cholesterol, free T3, mean corpuscular volume (MCV), left atrium anterior–posterior diameter, number of cardioversions, and oral anticoagulant use. ROC curve analysis demonstrated that our prediction model effectively distinguished between patients with arrhythmia relapse and those without, with an AUC value of 0.719 (p < .05).

Conclusions

This study revealed that AF recurrence after ECV can be predicted by our prediction model consisting of patient characteristics, disease factors, and various biochemical parameters. The current study demonstrated a statistically significant association between the number of cardioversions and non-HDL cholesterol levels in patients with recurrent AF.