Pharmacological validation of heart rate variability measures that assess sympathetic or vagal tone or the sympathovagal balance [ECG analysis]

Heart rate variability (HRV) analysis is commonly used to assess sympathetic or vagal tone, or the sympathovagal balance (SVB). Thorough validation, however, is lacking. According to the Rosenblueth-Simeone model, heart rate (HR) equals m/spl middot/n/spl middot/HR/sub 0/ (m=sympathetic factor, n=vagal factor, HR/sub 0/=intrinsic heart rate); m, n, and HR/sub 0/ can be determined by two-stage autonomic blockade. SVB is defined as m/spl middot/n. The current study aims at validating HRV measures with values of m, n, and m/spl middot/n. The authors analyzed the ECG of 21 healthy males (ages 24.9/spl plusmn/3.0 yrs) in three 5-min episodes. In episode 1 they computed HR, and the following HRV measures: SDNN, CV, pNN50, and the absolute and relative spectral powers VLF (0.01-0.05 Hz), HF (0.05-0.15 Hz), and HF (0.15-0.40 Hz). Prior to episodes 2 and 3 complete sympathetic and vagal blockade was instituted, respectively, for measurement of m, n, and HR/sub 0/. Linear correlations between m, n, m/spl middot/n, and all HRV measures, were computed. For all HRV measures the authors also determined the maximal mean performance (MMP; averaged sensitivity and specificity) to discriminate high from low values of m, n, and m/spl middot/n. Significant correlations >0.43 or <-0.43 were obtained for n vs. HR (r=0.67), and pNN50 (r=-0.66), and for m/spl middot/n vs. HR (r=0.73), SDNN (r=-0.46) and pNN50 (r=-0.67). Maximal mean performances >0.75 were found for n vs. HR (MMP=0.76), SDNN (MMP=0.80), CV (MMP=0.76), and pNN50 (MMP=0.76), and for m/spl middot/n vs. SDNN (MMP=0.80), CV (MMP=0.76), and pNN50 (MMP=0.76). These results demonstrate that time domain HRV measures superiorly rank the subjects on vagal tone or SVB. However, performance is weak. Therefore, reconsideration of current HRV interpretation seems to be indicated.<>