Cross-spectral analysis of heart rate and blood pressure modulations.

The cross-spectral analysis of heart rate (HR) and blood pressure (BP) variabilities provides "amplitude" and "phase" related measures. Compared to the amplitude measure, that is the baroreflex gain, the phase related measure characterizing the time lag between HR and BP oscillations has been studied to a much lesser extent. A population of 103 patients (73 men, 30 women, aged 53 +/- 12, range 20-82 years) referred for the management of coronary artery disease and/or hypertension were studied. In each subject, electrocardiogram and BP recordings were obtained in the supine and sitting positions of 5 minutes of rest (spontaneous respiration), 3 minutes of controlled respiration at 0.1 Hz (slow-controlled respiration), and 3 minutes of controlled respiration at 0.33 Hz (fast-controlled respiration). The frequency of maximum coherence (above the arbitrary threshold of 0.5) of BP and RR interval variabilities was searched between 0.033-0.133 Hz and 0.200-0.400 Hz to obtain baroreflex gain and phase shift in low and high frequency bands, respectively. Mean phase shifts of -79.1 and -67.0 degrees (-2.4 and -2.1 s) were found during slow-controlled respiration in the supine and sitting body positions, respectively. The mean phase shift between systolic BP and RR interval in the low frequency band was found between 83 and -109 degrees for body positions and respiration regimes. The actual baroreflex related time lag between systolic BP and RR variations was found between 3.5 and 5.1 seconds. The study concludes that the appropriate, and not always easy, selection of the frequency of maximum coherence between BP and HR oscillation is crucial for an accurate cross-spectral assessment of baroreflex sensitivity.

Stability of the noninvasive baroreflex sensitivity assessment using cross-spectral analysis of heart rate and arterial blood pressure variabilities.

BACKGROUND: Depressed baroreflex sensitivity (BRS), usually estimated using the invasive phenylephrine method or the nitroprusside test, is significantly and independently associated with an increased risk of malignant ventricular arrhythmias and sudden cardiac death in patients surviving acute myocardial infarction. Several investigators have compared the standard phenylephrine test and different noninvasive methods. HYPOTHESIS: This study evaluated the influence of different body positions with different breathing regimes on cross-spectral baroreflex indices (coherence between the spectral densities of blood pressure and cardiac cycle variabilities) in both low- and high-frequency bands. METHODS: The data were obtained in 103 patients (73 males, aged 53 +/- 12 years) with coronary artery disease and/or hypertension. Simultaneous electrocardiographic and noninvasive blood pressure recordings were obtained in each subject in both supine and sitting positions during both spontaneous and slow and fast controlled respiration (0.1 and 0.33 Hz). RESULTS: The results show a significant bias and disagreement between noninvasive baroreflex sensitivity (BRS) indices. The mean values of the baroreflex in low frequency ranged from 5.0 +/- 5.3 to 10.1 +/- 7.9 ms/mmHg, while in high frequency, the mean values ranged from 6.6 +/- 6.1 to 10.1 +/- 7.9 ms/mmHg. The limits of agreement ranged from +/-1.7 to +/-4.1 ms/mmHg with bias from -1.0 to +0.7 ms/mmHg. CONCLUSION: A comprehensive comparison of different methods shows that BRS estimated in low-frequency band in sitting position during spontaneous respiration is the most representative part of the global baroreflex gain.