Comparison of time- and frequency domain-based measures of cardiac parasympathetic activity in Holter recordings after myocardial infarction

Abstract The standard deviation of normal RR intervals predicts mortality after acute myocardial infarction independently of left ventricular function and ventricular arrhythmias. 1 Myers et al 2 suggested that this measure of heart period variability is a broad-band measure composed of high and low frequency components, and suggested that patients with malignant arrhythmias have reduced power in the high frequency band of the heart period or heart rate power spectrum. The high frequency band of the heart rate power spectrum reflects vagally mediated respiratory sinus arrhythmia. 3–5 Also, 2 time domain measures—the percent of absolute differences between successive normal RR intervals that exceed 50 ms (pNN50) and the root mean square successive difference (r-MSSD)—have been used as measures of tonic vagal activity. 6 The objective of this study was to determine the strength of association among 3 measures of tonic vagal activity: the high frequency band (0.15 to 0.50 Hz) in the heart period power spectrum, pNN50 and r-MSSD. We expected strong correlations among these 3 variables, and hoped that the correlations would be strong enough to permit their interchangeable use for assessing tonic vagal activity in long duration electrocardiographic recordings. We expected weaker correlations of these 3 measures with measures of total heart rate or heart period variability, such as the standard deviation of normal RR intervals or the total power in the heart period power spectrum.