Evaluation of Nonlinear Dynamics of Ventricular Repolarization in Normal Subjects and in Patients After Myocardial Infarction

Background: An application of the chaos theory to cardiology has gained growing interest during the last decade. It was shown that low dimensional chaos could be observed either in single cardiac cells, excitable cell assemblies, or in the whole heart and cardiovascular system. The aim of our study was to characterize the behavior of interbeat RT intervals variation in healthy subjects and in patients after myocardial infarction (Ml) in terms of the chaos theory. Methods: Twenty‐five healthy volunteers, aged 24 ± 7 years comprised the control group and 67 patients after Ml, aged 52 ± 10 years comprised the study group. Mean time after an acute Ml was 18 ± 17 months (range 3–60). The study was performed in supine position during spontaneous breathing at 10–12 a.m. ECGs were registrated from X Y Z leads. The signal was gained with a low noise amplifier. After A/D conversion (16 bit, 2000 Hz) and crude R wave centering, the fiducial points of R wave and of T wave were determined in the lead with the maximal T wave amplitude (at least 0.25 mV). RT interval variability analysis was performed in time domain to measure beat‐to‐beat RT interval duration differences from the mean (standard deviation of mean RT interval, SD‐RT, ms). Nonlinear analysis was based on self similarity test. Self similarity properties were quantified by calculation of Hurst coefficient (H), then fractal dimension (Drt) was drawn, according to the equation Drt = 2 — H. Results: The differences between the control and the study groups in respect to mean RR interval (917.5 ± 115.1 ms vs 945.5 ± 152.6 ms, respectively) and to RT interval variability (SD‐RT 2.67 ± 1.67 ms vs 3.49 ± 1.85 ms, respectively) were not significant. In the controls the magnitude of RT variability (SD‐RT) correlated with the mean RR interval (Spearman R = 0.402, P < 0.05) and with the magnitude of RR interval variability (SD‐RR) (R = 0.643, P < 0.001). In the Ml patients, no significant correlations were found (R = 0.061, P > 0.1, and R = 0.231, P > 0.1, respectively). When the magnitude of RT interval variability was normalized to the SD‐RR, a significantly greater relative SD‐RT was found in the Ml patients. Fractal dimension of RT interval variability Drt was 1.76 ± 0.12 in the controls. In the Ml patients Drt was significantly greater 1.87 ± 0.13 (P < 0.001, Kolmogorov‐Smirnov test). The fractal dimension of RT variability did not correlate with the mean RR interval or SD‐RR in the controls. In the study group, weak, but significant negative correlation between Drt and RRI was observed (R =−0.298, P < 0.05). When the upper limit of normal (i.e., Drt = 1.880) was set arbitrarily, abnormal (i.e., Drt > 1.88) fractal dimensions of RT variability were recognized in 41 Ml patients. In 13 of them, the left ventricular ejection fraction was < 40%, and in 7 patients the nonsustained ventricular tachycardia was present, while among the 26 patients with fractal dimension ≤ 1.880, only 2 had low ejection fraction and none experienced ventricular tachycardia. Conclusion: Our study indicates that methods drawn from the chaos theory might be a novel, useful tool in quantifying ventricular repolarization.