OBJECTIVE: The heart rate variability (HRV) has been used in patients with heart failure as a non-invasive method and provided neuro-cardiovascular evaluation. This study was planned to determine the efficacy and prognostic value of 24-hour and short time HRV in autonomic dysfunction in patients with congestive heart failure. METHODS: Forty-six patients with symptomatic or asymptomatic left ventricular dysfunction (ejection fraction <40%) were included to the study. In the study group, 16 patients were in NYHA class I (35%), 19 - were in NYHA class II (41%) and 11 - were in NYHA class III (24%). In the first day, HRV was evaluated from the 24-hour Holter recordings. Following day; we assessed the HRV during: (1) 10 min of supine resting, (2) 10 min of regular breathing at a frequency of 20 acts/min, and (3) 10 min of passive orthostatism after tilting 80 degrees with tilt table. RESULTS: Twenty-four hour and short time recordings of HRV showed significant decrease in long-term LF/HF24 (LF- low frequency, HF- high frequency), and short-term LF/HFsupine, LF/HFbreathing and LF/HFtilt ratios in patients with NYHA class III when compared with the patients in NYHA class I-II (p=0.0001, p=0.01, p=0.03, p=0.0001, respectively). During 446+/-186 days of follow-up, cardiovascular end-points occurred in 20 patients. In Cox multivariate analysis, significant predictors of cardiac mortality and morbidity were, reduced LF/HF ratio (HR=0.4, 95% CI 0.31-0.73, p=0.001) in the 24-hour recordings and low left ventricular ejection fraction (HR=0.9, 95% CI 0.83-0.99, p=0.03). CONCLUSION: Our study demonstrated that both of the methods were useful for assessment of cardiac autonomic dysfunction and only 24-hour recordings of HRV had a prognostic value.