Heart rate variability (HRV) has recently become a popular noninvasive research tool in cardiology. Clinical assessment of HRV is frequently based on standard long-term ambulatory electrocardiograms, whereas physiologic studies employ spectral analysis of short-term recordings under controlled conditions. From a general point of view, HRV can be used in clinical practice to estimate (1) the integrity of cardiac autonomic innervation, (2) the physiologic status of cardiac autonomic activity, and (3) the vulnerability to various cardiac arrhythmias resulting from autonomic imbalance. Clinical relevance of HRV has been clearly demonstrated in only two clinical conditions: (1) impaired HRV can be used alone or in a combination with other factors to predict risk of arrhythmic events after acute myocardial infarction, and (2) decrease in HRV is a useful clinical marker for evolving diabetic neuropathy. Substantial advances of our knowledge are required to establish and promote clinical applications in other areas of clinical medicine. To accomplish this task, proper hypotheses should be studied and appropriate techniques selected.