Spectral Analysis of Baseline Electrocardiogram During Atrial Fibrillation Predicts Response to Antiarrhythmic Drug Therapy in Patients With Persistent Atrial Fibrillation.

Abstract

We evaluated the ability of spectral analysis of the baseline ECG during atrial fibrillation (AF) to predict the response of persistent AF to antiarrhythmic drug therapy. Patients with persistent AF who were admitted for dofetilide loading were prospectively enrolled in the study. Atrial activity was extracted from the ECG using an Independent Component Analysis method and then subjected to a Modified Periodogram. The regularity index was computed as the ratio of the power in the dominant frequency and all its harmonics to the total power in the spectrum. Patients were followed at 1 month, 3 months and every 3 months thereafter. Of 28 patients enrolled in the study, 14 (50%) converted acutely to sinus rhythm during the 3-day hospital loading period. The clinical and echocardiographic characteristics of patients with and without acute pharmacologic conversion were similar. The regularity index was significantly higher in those who converted to sinus rhythm compared to those who did not (0.71±0.20 vs. 0.38±0.13, respectively; P < 0.0001). A regularity index ≥0.44 had a 79% sensitivity and 93% specificity to predict acute conversion and was associated with a nearly 5-fold increase in the acute conversion rate (odds ratio = 4.89; 95% confidence interval 1.74-13.75; P = 0.003). The regularity index was the only independent predictor of acute conversion. Neither acute conversion, nor the regularity index predicted sinus rhythm maintenance, after a median follow-up of 10 months. Increased regularity index predicts acute conversion of persistent AF during dofetilide loading, but does not predict long-term sinus rhythm maintenance.