ABNORMALITIES IN BEAT-TO-BEAT COMPLEXITY OF HEART-RATE DYNAMICS IN PATIENTS WITH A PREVIOUS MYOCARDIAL-INFARCTION

Objectives. The purpose of this research was to study possible abnorma lities in the beat to beat complexity of heart rate dynamics in patien ts with a previous myocardial infarction. Background. Analysis of appr oximate entropy of time series data provides information on the comple xity of both deterministic and random processes. It has been proposed that regularity or loss of complexity of RR interval dynamics may be r elated to pathologic states, but this hypothesis has not been well tes ted in cardiovascular disorders. Methods. Approximate entropy and conv entional time and frequency domain measures of RR interval variability were compared between 40 healthy subjects with no evidence of heart d isease and 40 patients with coronary artery disease and a previous Q w ave myocardial infarction. The groups were matched with respect to age , and cardiac medication was discontinued in the patients with coronar y artery disease before the 24-h electrocardiographic recordings. Resu lts. Approximate entropy was significantly higher in the postinfarctio n patients (1.21 +/- 0.18 [mean +/- SD]) than in the healthy subjects (1.05 +/- 0.11, p < 0.001), whereas the standard deviation of RR inter vals (63 +/- 19 vs. 86 +/- 23 ms, p < 0.001) and the very low, low and high frequency spectral components were lower (p < 0.01, p < 0.001, p < 0.05, respectively). Approximate entropy was not related to the tim e domain or the spectral components of heart rate variability and was more commonly abnormal in postinfarction patients (62.5%) than any lin ear measure (from 20% to 42.5%) when the 90% percentile of the values obtained for healthy subjects was defined as the normal range for each measure. Conclusions. Despite reduced linear measures of heart rate v ariability, the unpredictability or randomness of beat to beat heart r ate dynamics is increased in patients with a previous myocardial infar ction. Complexity analysis of RR interval dynamics may provide useful information on abnormalities in heart rate behavior that are not easil y detected by the commonly used moment statistics.