REPRODUCIBILITY OF CONVENTIONAL AND POWER SPECTRAL MEASUREMENTS OF CARDIOVASCULAR SYMPATHETIC ACTIVATION IN NORMAL SUBJECTS

1. Normal subjects (n = 5; age 20-42 years; mean resting blood pressur e (+/- 1 s.d.) 116 +/- 21/61 +/- 11 mmHg) underwent cardiovascular ref lex testing five times each. On every occasion systolic blood pressure (SEP) responses to sustained handgrip (GRIP) and cold presser (COLD) tests were measured and continuous non-invasive SEP and heart period ( RRINT) data were analysed in the frequency domain using fast Fourier t ransforms. Power spectral (PS) density estimates of high frequency/tot al power (HF%; a measure of vagal activity), low frequency/HF ratio (L F/HF; a measure mainly of cardiovascular sympathetic activity for hear t period) and low frequency/total power (LF%; a proposed measure of sy mpathetic activity for SEP) at rest, during and 2 min after the end of stimuli were calculated, 2, The data from the rest and recovery perio ds did not differ and showed that cardiovascular recovery to baseline measures following sympathetic stimulation occurred within 2 min, 3, T here was a significant rise in SEP with GRIP and COLD, The LF/HF(RRINT ) rose significantly with GRIP, but not with COLD, The LF%(SBP) did no t change significantly with GRIP or COLD, 4. The SEP and PS analyses s howed low intra-individual reproducibility of responses to reflex test s, with coefficients of variation for PS measures at rest of 25-41% an d on sympathetic stimulation of up to 80%, 5, The high variability of these observations indicates that PS methods may not be suitable for t he analysis of transient cardiovascular reflexes.