Autonomic response to cardiac dysfunction in chronic heart failure: A risk predictor based on autonomic information flow

Background: Chronic heart failure (CHF) is associated with a complex dysfunction of cardiac, cardiovascular, autonomic, and other mechanisms. Autonomic information flow (AIF) characteristics calculated from heart rate patterns were recently found as promising predictors of outcome in several cardiovascular diseases. Aim: To assess the prognostic value of AIF indices in CHF patients. Methods: We analyzed 24-hour Halter recordings from 200 consecutive CHF patients in sinus rhythm and computed AIF over the shortest possible interval of an interbeat series, namely over one heart beat interval (BDnn), and over longer intervals (12.5-166.7 seconds, PDmVLF), which reflect slower heart rate modulations. End-point for survival analysis over three years (Cox model) was total cardiac death. A prognostic model was built (backward elimination) considering known clinical and functional risk factors, and the ability of AIF indices to add prognostic information to this model assessed. Results: Out of candidate predictors, New York Heart Association class, left ventricular ejection fraction, peak VO2. and systolic pressure were selected as the variables with the highest joint predictive value. When entered into this model, both BDnn and PDmVLF added prognostic information (HR (95%CI): 1.76 (1.00-3.09), P = 0.05, 1.73 (1.05-2.85), P = 0.031 respectively). High risk was associated with reduced fast AIF and increased slower AIF. Conclusion: In CHF patients, AIF indices provide prognostic information independent of known risk factors.