The cardiovascular system displays features typical of self-organising systems designed to achieve dynamical stability. In the case of the cardiovascular system, stability is achieved by autonomically mediated control of heart rate, blood pressure and other factors which react rapidly to a range of internal and external stimuli such as acute ischaemia, metabolic imbalance and changes in physical or mental activity. In particular, heart rate varies in a complex reactive manner to these stimuli (which occur even in resting individuals). These variations represent an unique individual pattern that is called in the clinical medicine heart rate variability (HRV).

In the late 70’s, a reduction in HRV was first reported to correlate with increased mortality and/or a number of clinically relevant arrhythmic events in survivors of myocardial infarction (Wolf 1978). Reduced HRV has been shown to be a strong indicator of risk related to the development of a range of cardiovascular (Kleiger 1987, Malik 1990, Malliani 1994, Nolan 1998) and metabolic diseases (O’Brien 1991). E.g., the Ewing battery of reflex HRV tests (Ewing, 1981) became an established set of standardised stimuli with analysis of effects on heart rate and blood pressure in time-domain. Further on, in diabetic patients with late complications such as microproteinuria, a diminished HRV was found predictive for further renal deterioration. As the impact of dysfunction in autonomic control on prognosis in diabetes was recognised, it became clear that the estimation of cardiac autonomic function should be routinely performed in diabetic patients. Analysis of HRV in the frequency /=spectral/-domain allows a more detailed assessment of autonomic function with investigation of effects of sympathetic and parasympathetic subsystems. Even its short-term version proved a significant predictive value. Change in the HRV pattern as measured by spectral analysis is understood by many authors as an early and sensitive marker indicating trespassing the threshold between the health and disease and reduced HRV is seen as a sign of ‘compromised’ health (Dekker 1997).

With the progress of information technology, it is now possible to explore the functioning of autonomic control system simply and non-invasively, using sophisticated analysis of heart rate variability. HRV is quantified here by analysis of variations of the intervals between consecutive normal heartbeats. The usual definition of a heartbeat interval is the time between consecutive R wave peaks. Advances in computer technology have allowed sequential R-R intervals to be measured accurately and recorded in real time. After passing through automated ectopic beat and artefact handling procedures, sophisticated and fast methods of analysis can then be applied to the data to determine HRV measurements which reflect autonomic nervous system activity. Typically, HRV can be measured using add-on software in some standard ECG machines or by dedicated HRV analysers. Instantaneous heart rate is displayed on a computer screen in the form of a bar graph (known as a tachogram), where each bar represents the heart rate associated with an individual sequential R-R interval. A curve passing through the peaks of these bars represents the variation in heart rate. Useful parameters which relate to autonomic processes are determined by further quantitative analysis of this curve.

The HRV analysis method is already an established tool in cardiology and diabetology research and practice, and is increasingly being used in a range of other clinical applications. Some practical examples of HRV will be demonstrated and discussed in the presentation.