Heart rate variability and fractal dimension during orthostatic challenges

Abstract

Heart rate variability (HRV) spectral analysis has been used as a tool for short-term assessment of parasympathetic (PNS) and sympathetic nervous system (SNS) control of heart rate. However, it has been suggested that the PNS and SNS indicators are superimposed on a broad-band noise spectrum in which the power spectral densities are inversely proportional to their frequency (1/f beta). In this study, we have used coarse-graining spectral analysis to extract the harmonic components for calculation of PNS and SNS indicators and to obtain the slope (beta) of the 1/f beta component to estimate fractal dimension (DF) of a trail of HRV. DF was regarded as an indicator of cardiovascular system complexity. Ten healthy young subjects (6 women and 4 men) were studied in supine rest and with sequential applications of four levels of lower body negative pressure (LBNP; -10, -20, -30, and -50 mmHg) and head-up tilt (HUT; 10, 20, 30, and 70 degrees). In the 20 tests, there were six occurrences of presyncopal symptoms that required the test to be terminated before the planned end point. At low levels of LBNP or HUT, arterial pulse pressure (PP) was not changed from rest, and calculated DF was very high (beta approximately 1.00). At the higher levels of LBNP and HUT, PP decreased. Coincident with this reduction in PP, PNS activity decreased, SNS activity increased, and DF was reduced, each with a significant linear relationship to the change in PP (PNS: r = 0.56; SNS: r = 0.57; DF: r = 0.70, P < 0.01). Each occurrence of presyncope was associated a low PNS indicator as well as DF < 2.50 (beta > or = 1.80). These data indicate that the cardiovascular system is operating at a reduced level of complexity and further suggest that reduced complexity might not be compatible with cardiovascular homeostasis.