Bispectral analysis as a tool to investigate dynamics of cardiorespiratory physiology.

INTRODUCTION: Recent studies show that cardiorespiratory variables are nonlinearly coupled. In the present study, we tested the feasibility of using bispectral analysis to estimate changes in nonlinear phase coupling. METHODS: We stimulated chemo- and baro-reflexes by switching inspired gas between air and air with 5% CO2 added during supine and 70 degrees head-up tilt. Bispectra were estimated using Fourier transform of the triple cumulant and were integrated between 0.04 and 0.3 Hz. Variables analyzed included cardiac RR intervals, systolic BP (SBP), tidal volume (VT), end tidal pressure of CO2 (PETCO2) and mean cerebral blood flow velocity (CBFM). Data were collected in 30 healthy adults who were then grouped according to whether they become presyncopal (PS) during tilt or remained normal with no signs or symptoms of presyncope (NPS). RESULTS: Six subjects (20%) developed presyncope. During tilt with room air breathing, phase coupling in RR, SBP, and between RR-SBP and RR-VT was lower in PS than in NPS, while phase coupling between CBFM-SBP was higher. In contrast, during tilt with 5% inhaled CO2, phase coupling between CBFM-PETCO2 was lower for PS than for NPS. DISCUSSION: Perturbations to baro- and chemo-reflexes led to changes in nonlinear phase coupling that were altered in presyncope. Although physiological interpretation of changes in phase coupling are less than clear at this time, our results show that such nonlinear indexes may provide a helpful perspective in understanding the complex phenomenon of orthostatically mediated syncope.

Transfer function analysis of gender-related differences in cerebral autoregulation.

Effects of gender in cardiovascular regulation are widely investigated, however differences in cerebral autoregulation (CA) between genders remains less explored. In the present study, we used spectral analysis to analyze differences in cerebral autoregulation between 10 men and 10 age-matched women. Subjects rested in a supine position (20 min) and then were passively tilted to a 70-degree head up tilt position (30 min). During the first 10 min of supine they breathed only room air (Supine Control) while during the second 10 min of supine they breathed room air or room air plus 5% CO2 in a Pseudo Random Binary Sequence (PRBS) (Supine PRBS). During the first 10 min of tilt, subjects breathed room air or room air plus 5% CO2 switched in a PRBS (Tilt PRBS). For the remaining 20 min of tilt, they breathed only room air (Tilt Control). Blood pressure (BP) and cerebral blood flow velocity (CBFV) were non-invasively recorded using Finapres and Transcranial Doppler. Coherence and transfer function between mean BP and mean CBFV were estimated. During Supine Control, women had higher coherence (p < 0.05) and transfer function gain (p < 0.01) within 0.03-0.10 Hz and 0.22-0.31 Hz than men. During tilt, coherence within 0.05-0.26 Hz were higher in men than in women (P < 0.02). Although the reasons for these gender-related differences in cerebral autoregulation remain unclear, our results suggest that in investigating cerebral autoregulation, gender may need to be considered as a factor.

Orthostatic modification of ventilatory dynamic response to carbon dioxide perturbations.

In order to determine whether changes in ventilatory control contribute to the observed decrease in arterial partial pressure of carbon dioxide (PaCO(2)) during head up tilt, we assessed ventilatory dynamic sensitivity to changes in PaCO(2) during supine and 70 degrees passive head up tilt. In 24 adult normals, we stimulated the ventilatory control system by switching inspired CO(2) between room air and room air+5% CO(2) in a pseudo random binary sequence. A Box-Jenkins model was used to compute ventilatory response to CO(2). Airflow, CO(2), non-invasive beat by beat blood pressure, ECG and cerebral blood flow velocity (Doppler) were recorded. During tilt, sensitivity of the ventilatory controller to CO(2) disturbance increased (from 0.45 to 0.72 L/min/mm Hg, p<0.005); minute ventilation increased (7.63 to 8.47 L/min, p<0.01), end tidal CO(2) (ETCO(2)), cerebral blood flow velocity (CBF) and baroreflex sensitivity decreased (46.9 to 42.9 mm Hg, p<0.001; 84.9 to 72.9 cm/s, p<0.001; and 17.6 to 5.5 ms/mm Hg, p<0.001). The primary observation from our study was that the sensitivity of ventilatory control system to perturbations in ETCO(2) increased during tilt. Taken together with decrease in mean levels of ETCO(2) and an increase in minute ventilation, these results suggest that during tilt, a change in the regulated level or 'set point' of PaCO(2) may occur.

Assessment of ventilatory sensitivity to carbon dioxide changes during orthostasis.

Investigators have anecdotally reported changes in respiratory pattern preceding symptoms of orthostatic intolerance for several years. Evidence from recent studies support these observations, and show that alterations in respiration often precede pre-syncope during orthostasis. These observations suggest the possibility that changes in interaction between the chemo and baro-reflex control systems may produce phasic or dynamic changes in respiratory and hemodynamic parameters such that these changes are important in orthostasis intolerance. Our objective in the present study was to develop a method to obtain a quantitative index of ventilatory sensitivity to changes in inspired during orthostasis. Based on an approach previously developed by one of the authors (EB), we used, during supine and 70 degrees head up tilt, pseudo-random binary changes in inspired CO2 to quantify ventilatory sensitivity to CO2 disturbances. A pneumotach was used to measure airflow. From these recordings, we used a prediction error based systems identification algorithm to quantify ventilatory impulse response to CO2 stimulus. Results from 7 subjects showed that the integrated ventilatory response (area under impulse response curve) to CO2 was larger during tilt than that during supine (mean [std] 5.3 [2.6] Vs 3.0 [1.6] ml/min/0.01LCO2). These results suggest that the changes in ventilation due to disturbances in CO2 are likely to be larger during orthostasis than those during supine. Such an elevated response would indicate that the role of chemo-reflex control of breathing in cardiovascular instability culminating in orthostatic intolerance is likely to be important and should be further investigated.