OBJECTIVES: Power spectrum analysis decomposes the heart rate signal into its frequency components and facilitates separation of sympathetic (low frequency) and parasympathetic (high frequency) activity. In congestive heart failure, augmented sympathetic tone and decreased parasympathetic tone were found. Autonomic nervous system was normalized 6 months after myocardial infarction. So we compared the autonomic nervous system activity by the heart rate variability in congestive heart failure and old myocardial infarction. METHODS: The protocol involved 20 healthy subjects (Group 1), 5 congestive heart failure patients not caused by myocardial infarction (Group 2), 4 congestive heart failure patients due to myocardial infarction and 11 old myocardial infarction patients without heart failure. We took 24 hour Holter monitoring by Del Mar Avionic tape recorder. All Holter tapes were analyzed with use of Model 563 Stratascan Holter Analysis System. We computed power spectra on each 256 sec segment of each hour during 24 hour recording. So, RR interval, SD of RR interval by time domain, and LF, HF, LF/HF ratio, Total PSD by frequency domain were measured. RESULTS: In congestive heart failure, nocturnal HF peak and diurnal variation of LF/HF ratio was decreased relative to healthy subjects. Nocturnal HF peak in old myocardial infarction was not visualized. All of LF, HF and Total PSD in congestive heart failure and old myocardial infarction patients relative to healthy subjects. CONCLUSION: On heart rate variability analysis using by 24 hour Holter monitoring, abnormal autonomic nervous activity was demonstrated in congestive heart failure and old myocardial infarction patients relative to healthy subjects.