INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LOW-FREQUENCY ARTERIAL PRESSURE oscillations or "Mayer waves" were first described in 1877 (14), yet mechanisms underlying these low-frequency rhythms have remained elusive. Because the magnitude of systolic pressure fluctuations is directly associated with severity of end-organ damage (renal, heart, and brain) and subsequent cardiovascular complications (18), understanding mechanisms of arterial blood pressure oscillations assumes clinical importance.

In an effort to explain arterial pressure rhythms occurring at frequencies slower than respiration, at least two hypotheses have been advanced. First, because of intrinsic delays in effector responses to norepinephrine, increases of sympathetic nerve firings are manifested in cardiac and vascular responses with a time delay of ~10 s (28). With this construct, waxing and waning of arterial pressures around 0.1 Hz are likely driven by arterial baroreceptor input (9). Second, low-frequency arterial pressure rhythms have been recorded in patients with cervical spinal cord lesions (10) and in isolated mesenteric arteries (21), suggesting the possibility that the peripheral vasculature possesses intrinsic rhythmicity independent of baroreflex feedback loops. The notion that Mayer waves reflect vascular sympathetic outflow has recently been challenged (27), and therefore factors that may contribute directly to vascular rhythmicity warrant further investigation.

Increases and decreases of blood flow associated with changes in vascular shear stress activate and inhibit production of nitric oxide (NO), which directly affects peripheral vascular resistance (25). Inhibition of NO synthesis increases low-frequency arterial pressure variability in dogs (15) and rats (16), suggesting a primary role of NO in buffering changes of arterial pressure at the frequency of Mayer waves. Castellano and co-workers (3) tested the influence of NO inhibition in humans and reported reductions of low-frequency arterial pressure oscillations. However, such reductions might be expected if blockade of NO results in hypertension and subsequent sympathetic neural inhibition (25), as occurred in the study by Castellano et al. To our knowledge, the potential of NO to buffer arterial pressure oscillations in humans has not been resolved.

Therefore, to better understand interactions between NO and arterial pressure oscillations in humans, we used passive head-up tilting to increase rather than decrease sympathetic nerve activity with systemic inhibition of NO synthase (NOS). We tested the hypothesis that inhibition of NOS with N^G-monomethyl-L-arginine (L-NMMA) increases arterial pressure oscillations at the low frequency on assumption of the upright posture.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. We studied eight healthy subjects (6 men and 2 women; height 173 ± 4 cm, weight 73 ± 4 kg, age 31 ± 3 yr). All subjects were nonsmokers; refrained from caffeine, alcohol, and heavy exercise 24 h before the study; and had no history of cardiovascular or autonomic dysfunction. This study was conducted in accordance with the Declaration of Helsinki (1989) of the World Medical Association and was approved by the Institutional Review Boards of the University of Texas Southwestern Medical Center and the Presbyterian Hospital of Dallas. Each subject gave written, informed consent before participating.

Instrumentation. We continuously monitored heart rate with electrocardiography, beat-by-beat arterial pressure with finger photoplethysmography (Finapres, Ohmeda, Aurora, CO), brachial cuff pressure with standard sphygmomanometry (Suntech), and respiratory excursions with piezoelectric transducers (Pneumotrace, Morro Bay, CA). We positioned the Finapres transducer at heart level to avoid spurious recordings due to changes of hydrostatic pressure gradients during changes of posture. Beat-to-beat recordings of arterial pressure were used for spectral analysis and cuff pressures were used to compare absolute arterial pressures between experimental conditions.

Protocol. We studied all subjects in a quiet, temperature-controlled (25°C) laboratory in the morning, at least 2 h postprandial. After instrumentation, subjects rested quietly in the supine position for at least 30 min. We collected 6 min of data with subjects in the supine position. The subjects were then passively tilted to a 60° head-up position. After 2 min of stabilization, we collected 6 min of data during tilt. The subjects were then returned to the supine position for a recovery of ~45 min. After this test, systemic NOS was blocked by intravenous infusion of L-NMMA with a loading dose of 5 mg/kg for 15 min and a maintenance dose of 5 µg · kg¹ · min¹ through the study. Previous work has demonstrated stable blood levels and sustained NOS inhibition with use of this regimen in human subjects (13). After 30 min of L-NMMA infusion, 6 min of data were collected in the supine position, followed by a repeat of the 60° head-up-tilt protocol (2-min stabilization, 6-min data collection).

Data analysis. Electrocardiogram and arterial pressure waveforms were sampled at 1 kHz and digitized at 12 bits with an analog-to-digital converter (Das-20, Metrabyte). Respiratory excursions were sampled simultaneously with electrocardiogram and arterial pressure signals at 10 Hz. Digitized signals were stored on a laboratory computer and processed with a customized software program for R wave and systolic and diastolic pressure detection. Beat-to-beat R-R interval and systolic and diastolic pressures were linearly interpolated and resampled at 2 Hz for spectral analysis. The time series of R-R interval and systolic and diastolic pressures were first trend eliminated with third-order polynomial fitting and then subdivided into 256-point segments with 50% overlap for spectral estimation. Fast Fourier transforms were then implemented with each Hanning-windowed data segment and then averaged to calculate auto-spectra, cross-spectra, coherence, and transfer functions.

Statistics. Mean values for dependent variables of interest were examined with a 2 (condition; drug vs. no drug) × 2 (position; supine vs. tilt) factorial analysis of variance with repeated measures on both factors. In the event of significant interactions, we decided a priori to probe further with simple main effects analysis. We considered differences to be significant when P  0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Inhibition of NOS with L-NMMA infusion in the supine position significantly increased systolic (P = 0.03) and diastolic (P = 0.002) pressures but did not affect R-R interval, transfer function gain, or blood pressure variabilities.

Passive 60° head-up tilting significantly decreased R-R intervals (P = 0.0001), R-R interval variability at the high frequency (P = 0.03), and transfer function gain between systolic pressure and R-R interval at both the low (P = 0.007) and high (P = 0.002) frequencies. Tilting significantly increased systolic and diastolic pressure variabilities at both low and high frequencies (systolic: low-frequency power, P = 0.0001 and high-frequency power, P = 0.049; diastolic: low-frequency power, P = 0.003 and high-frequency power, P = 0.02) but did not affect mean cuff pressures or the coherence between systolic pressure and R-R interval. Passive 60° head-up tilting after inhibition of NOS had no additional effect on any of the variables we measured compared with the responses to head-up tilt without L-NMMA. Specifically, systolic arterial pressure oscillations at the frequency of Mayer waves increased similarly with head-up tilt both before and after L-NMMA infusion (P = 0.43).

Figure 1 shows representative arterial pressure oscillations with tilt before and after L-NMMA infusion. Numerical results for all dependent variables of interest across all measurement conditions are shown in Table 1.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Systolic pressure oscillations for 1 representative subject during supine (top) and head-up-tilt (bottom) measurements. L-NMMA, NG-monomethyl-L-arginine.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We tilted subjects to a 60° head-up position before and after systemic inhibition of NOS to test the hypothesis that NO contributes importantly to the dynamic regulation of arterial pressure rhythms. Our two primary findings are 1) L-NMMA infusion increases arterial pressure in the supine position, suggesting that either central or direct modulation of vascular tone by NO contributes to steady-state arterial pressure control, and 2) L-NMMA infusion has no effect on arterial pressure oscillations either supine or during tilt, suggesting that NO contributes minimally, if at all, to the modulation of low-frequency blood pressure rhythms.

Effects of NO on supine, resting humans. Afferent and efferent signals controlling arterial pressure derive, initially, from prevailing cardiac output and peripheral vascular resistance. Because humans exist in a "closed-loop" environment, it is not possible to know with certainty which signal precedes which, but it is clear that numerous mechanisms controlling either cardiac output and/or peripheral resistance generate and control steady-state arterial pressure and arterial pressure oscillations. In addition to reflex autonomic neural control, vascular tone is determined importantly by local factors, including tonic release of endothelium NO, which is a powerful vasodilator (6). NO is produced by the enzyme NOS, whose precursor is L-arginine, and can be blocked by analogs of L-arginine, including L-NMMA. In the present study, inhibition of NOS with L-NMMA significantly increased supine arterial pressures, confirming a role for NO in the control of steady-state pressure. Others have reported similar results (11, 12). Of note, significant changes of arterial pressures after L-NMMA infusion in the present study did not translate into changes of vagal-cardiac control or arterial barororeflex sensitivity, as estimated by R-R interval oscillations at the high frequency and transfer function analysis between systolic pressures and R-R intervals. The coherence function, which reflects the strength of the linear association between input and output signals at specific frequencies, was significant [by convention 0.50 (8)] before and after L-NMMA infusion (Table 1), suggesting that inhibition of NO does not uncouple the usual relation between systolic pressure R-R interval responses. R-R intervals were unchanged despite the significant increases of steady-state arterial pressure after L-NMMA, which may reflect a central influence of NO (29) resulting in rapid baroreflex resetting. The notion of rapid resetting of arterial baroreflexes is not without precedent. For example, carotid-cardiac baroreflex gain is unchanged during exercise despite increases of carotid sinus pressure (20).

Effects of NO during 60° head-up tilt. Moving from the supine to standing position causes a large portion of the circulating blood volume to move below the level of the heart, and this volume must be pumped against a hydrostatic gradient to maintain arterial pressure and cerebral perfusion. Accordingly, compensatory mechanisms have evolved such as venous backflow valves, skeletal and respiratory muscle pumps, and autonomic reflexes that respond to reductions in ventricular stroke volumes (and thereby changes in arterial pulse amplitude) by increasing heart rate and peripheral vascular resistance secondary to vagal withdrawal and sympathetic stimulation. We found, as others have before (7, 26), that passive head-up tilting decreases high-frequency R-R interval spectral power and transfer function gain at low and high frequencies. These data are consistent with a general withdrawal of vagal neural control during tilt.

Summary. Inhibition of NOS increases arterial pressure oscillations at low frequencies in rats (16) and dogs (15) and increases steady-state arterial pressures in humans (11, 12). On the basis of this evidence and evidence that the peripheral vasculature seems to possess intrinsic rhythmicity (10, 21, 30), we tested the hypothesis that NO contributes importantly to the control of arterial pressure rhythms in humans. Our data confirm a primary role for NO in maintaining stable arterial pressures at rest but provide evidence against the notion that dynamic oscillations of pressure, including naturally occurring rhythms at the frequency of Mayer waves, depend critically on tonic NO activity.