Searching for Chaos in Fibrillation a

Ventricular fibrillation (VF) is a state of disorganized, ineffective contraction of the heart’s two main pumping chambers. Ventricular fibrillation has traditionally been described as “turbulent” and chaotic,’ and in recent years there has been speculation that fibrillation may be an instance of deterministic chaos in the context of nonlinear dynamical systems theory. The idea that fibrillation is chaos has received indirect support both from mathematical models and experimental observations. A simple deterministic24 model of cardiac electrical activity displays fibrillation-like activity. This suggests that the seeming randomness of fibrillation may arise from a deterministic dynamical system. Experiments oriented toward investigating the transition from a normal heart rhythm to fibrillation have shown that there is a correlation between decreased electrical stability of the heart and a “period doubling” in cardiac r h ~ t h m . ~ The experiments begin with the heart in an approximately periodic rhythm. The changes in this rhythm are observed as a parameter of the system-such as temperature-is changed. At several points in the experiment, the stability of the pattern of electrical activity in the heart is probed by applying a series of electrical perturbations to the heart. The amount of electrical current needed to induce VF is taken as a measure of electrical stability: the greater the required current, the greater the stability. It is tempting to see such experiments as analogous to numerical “experiments” in which a parameter of a dynamical system is gradually changed, and the resulting bifurcation behavior is observed. One widely observed route to chaos in such numerical experiments is a cascade of period doublings that turn a periodic system into an aperiodic and chaotic one. Although the observed doubling in the period of the heart’s rhythm may be such a period doubling on the way to chaos, the analogy between the cardiac electrical instability experiments and the numerical experiments is not perfect. In particular, the type of stability tested in the cardiac experiments has to do with the proximity of different basins of attraction, and not the pitchfork bifurcations associated with the period-doubling route to chaos6.’