Impaired cardiac autonomic regulation and long-term risk of atrial fibrillation in patients with coronary artery disease

BACKGROUND The influence of autonomic cardiac regulation on long-term risk of new-onset atrial fibrillation (AF) in coronary artery disease (CAD) is not well established. OBJECTIVE The purpose of this study was to evaluate the value of heart rate variability, a marker of cardiac autonomic regulation, in predicting new-onset AF in CAD. METHODS The Innovation to Reduce Cardiovascular Complications of Diabetes at the Intersection study population consisted of 1946 patients with CAD. After exclusions, the present analysis included 1710 patients. Those patients had a 24-hour electrocardiographic recording at baseline. RESULTS A total of 143 cases (8.4%) of new-onset AF were observed during a follow-up of 5.6 +/- 1.5 years. The lower values of the short-term scaling exponent of the detrended fluctuation analysis (DFA1) and the ratio of the low-frequency and high-frequency components of the power spectrum (LF/HF ratio) remained the strongest heart rate variability predictors of the development of AF after relevant adjustments in Cox multivariate regression analysis (P < .001 for both). The accuracy of these parameters in prediction of AF was even better (area under the receiver operating characteristic curve 0.630 and 0.631, respectively) than that of echocardiographic left atrial diameter (area under the curve 0.618). Including DFA1 and LF/HF ratio in the AF risk model increased the C-index from 0.630 (95% confidence interval 0.569-0.692) to 0.666 (95% confidence interval 0.612-0.720). CONCLUSION Impaired cardiac autonomic regulation measured by DFA1 and LF/HF ratio predicts the development of new-onset AF as well as or even better than left atrial diameter in patients with CAD.