Nonlinear Analysis Of Heart Rate Variability Before And After Coronary Artery Bypass Surgery,

It is now well accepted that heart rate variability (HRV) conveys important clinical information about the dynamic character of heart. From the basic theoretical point of view, it is important to know whether HRV is deterministic or not. However, HRV signals are always contaminated with noise and look like very disorder, it is often hard to draw correct conclusion from some simple techniques of analysis. Babloyantz and Destexhe advocated that HRV of normal subjects was chaos in 1988; In 1990, Goldberger et al confirmed this result, and offered that HRV of some kind of serious patients (e.g. Ventricular fibrillation, etc.) was nearly regular (periodic), but Glass and Malta questioned this advocacy. In the past decade, some people used nonlinear prediction and surrogate data method in analyzing HRV of healthy people and rats, and concluded that there was no evidence for low-dimensional chaos for HRV.In this paper, we analyzed HRV signal of 28 coronary heart disease patients and several normal subjects by make use of surrogate data method, recurrence plot analysis (RPA), singular value decomposition and method of false nearest neighbors. We calculated the correlation dimension^,^), maximum Lyapunov exponent( A j), correlation time( r ), L-Z complexity(CL-z) and correlation coefficient for each HRV and compared them with their corresponding surrogate data. The results show:1). HRV of coronary heart disease patients obeys deterministic dynamical law, they are probably high dimensional chaos with fractal dimension in the range 4-5.2). There are some changes in the value of nonlinear character quantities of HRV after coronary artery bypass operation, but the nonlinear character of HRV remains unchanged.3). The observed HRV of normal subjects are also of high dimensional chaos with values of characteristic quantities different from those of coronary heart disease.