Chapter 8. Conclusions

1. This research showed that a dynamical analysis of heart rate behaviour derived from nonlinear mathematics can reveal abnormal patterns of RR interval dynamics which cannot be detected by commonly employed moment statistics of heart rate variability.

2. Approximate entropy showed heart rate tracings to be more predictable in patients with uncomplicated coronary artery disease, but more complex in patients with previous myocardial infarction as compared to healthy controls. This method was not able to differentiate patients with and without ventricular tachyarrhytmias.

3. A short term fractal-like scaling exponent of RR intervals showed more organised behaviour in patients with uncomplicated coronary artery disease. It was not able to differentiate patients with previous myocardial infarction from healthy controls. This measure was markedly reduced in patients with life-threatening arrhythmia and was the best variable to differentiate patients with and without ventricular arrhythmia.

4. Long term power-law slope was normal in patients with uncomplicated coronary artery disease, but significantly steeper before ventricular fibrillation, and it also predicted mortality in a general elderly population.

5. The consequences of a breakdown of fractal-like organisation were seen in ventricular tachyarrhythmia patients, in whom the breakdown of the fractal-like behaviour of RR intervals predicted life-threatening ventricular arrhythmia, and also in elderly people, among whom those with altered fractal-like behaviour of RR intervals had significantly higher mortality rates than those with normal fractal-like heart rate behaviour.