Power Spectral Analysis to Predict Death after Acute Myocardial infarction
After the initial discovery that SDNN predicted death, Bigger, Fleiss, Rottman, Kleiger proposed a 5-year project to better define the use of RR variability to predict death and sustained arrhythmic events in coronary heart disease. Bigger et al. proposed to use several 24-hour power spectral measures in the studies because these measures were mutually exclusive and all-inclusive. It was hoped that cardiovascular conditions, e.g., myocardial infarction and heart failure, would cause selective decreases in power spectral bands, especially those bands that are modulated by the autonomic nervous system. For example, it was postulated that myocardial infarction would selectively decrease high frequency power of the 24-hour power spectrum. As the proposed studies were done, it became clear that the initial hypothesis was incorrect. Myocardial infarction caused about the same fractional reduction in all frequency bands of the 24-hour power spectrum. Bigger et al. (38) studied 715 patients from MPIP who had a Halter recording done two weeks after myocardial infarction to establish the associations between six frequency domain (power spectral) measures of RR period variability and mortality during four years of follow-up before and after adjusting for five previously established post infarction risk predictors.
Power Spectral Measures of RR Variability. The six measures were calculated from power spectral analysis of continuous 24-hour ECG recordings using the CHRONOS algorithms. Using 24-hour power spectral density, they calculated the power within four frequency bands: (1) <0.0033 Hz, ultra low frequency (ULF) power; (2) 0.0033 to <0.04 Hz, very low frequency (VLF) power, which shows a relative increase in patients with congestive heart failure (39) and is the lowest frequency band that can be estimated by our 5-minute method (19); (3) 0.04 to <0.15 Hz, low frequency (LF) power, which reflects modulation of sympathetic or parasympathetic tone by baroreflex activity (40); and (4) 0.15 to 0.40 Hz, high frequency (HF) power, which reflects modulation of vagal tone, primarily by breathing (41,42). In addition, they calculated total power (power ffl.40 Hz) and the ratio of low to high frequency power, a measure that has been used as an indicator of sympathovagal balance (31). High values for the ratio suggest predominance of sympathetic nervous activity. The 24-hour recordings were digitized without benefit of phase lock loop; therefore, flutter and wow could cause small (<4%) increases in HF power (0.15-0.40 Hz).
Correlations Among Time and Frequency Domain Measures of RR Variability. Table I lists some of the variables that have been used to assess RR variability in cardiovascular diseases. In the 715 myocardial infarction patients. Bigger et al. determined the correlations among time and frequency domain measures of RR variability and the predictive value of time domain measures of RR variability for death during follow-up after acute myocardial infarction (37). Each frequency domain measure of RR variability had one or two corresponding variables in the time domain that correlated with it so strongly (r >0.90) that the variables were essentially equivalent: ultra low frequency power with SDNN and SDANN index, very low frequency power and low frequency power with SDNN index, and high frequency power with r-MSSD and pNN50. As expected from theoretical considerations, SDNN and the square root of total power were almost perfectly correlated. Time domain measures of RR interval variability, especially those that measure ultra low or low frequency power, were strongly and independently associated with death during followup in this sample.
Univariate Association of RR Variability and Death. The association between the six measures of RR variability and three mortality end points were evaluated: death from all causes, cardiac death, and arrhythmic death by the Hinkle-Thaler definition (43). Results using all-cause mortality as the endpoint are shown in Figure 7. Each measure of RR variability had a significant and at least moderately strong univariate association with all-cause mortality, cardiac death, and arrhythmic death. Power in the lower frequency bands - ultra low frequency (ULF) and very low frequency (VLF) power - had stronger associations with all three mortality end points than power in the higher frequency bands - low frequency (LF) and high frequency (HF) power. The 24-hour total power also had a significant and strong association with all three mortality end points. VLF power was the only variable that was more strongly associated with arrhythmic death than with cardiac death or all-cause mortality (38).
