(Hypertension. 1995;25:1167-1171.)
© 1995 American Heart Association, Inc.
Articles |
From the Division of Cardiology, Toronto Hospital and Centre for Cardiovascular Research, University of Toronto (Canada).
Correspondence to Dr John S. Floras, Division of Cardiology, Mount Sinai Hospital, Suite 1615, 600 University Ave, Toronto, Ontario M5G 1X5, Canada.
| Abstract |
|---|
|
|
|---|
Key Words: pressoreceptors blood pressure heart rate peptides, atrial natriuretic
| Introduction |
|---|
|
|
|---|
Among the seemingly random variations in heart rate and blood pressure that occur at rest in conscious animals13 and humans,12 14 15 16 17 18 one can identify by invasive or noninvasive methods sequences of three or more beats in which systolic pressure and pulse intervals change in parallel, either increasing (Up sequences) or decreasing (Down sequences).12 16 17 18 Calculation of the slope of the regression equation relating RR interval to the preceding level of systolic pressure during these sequences yields an estimate of the gain of the spontaneous baroreceptorheart rate reflex (or baroreflex sensitivity, BRS) (in milliseconds per millimeter of mercury [ms · mm Hg-1]) quantitatively similar to that obtained in response to bolus administration of vasoactive drugs by the ramp method.16 17 These brisk sinus node responses to ramp increases and decreases in arterial pressure are mediated by vagal cholinergic activation or withdrawal, respectively,19 and tend to be greater when blood pressure rises than when it falls.20
Recently, using spectral analysis of heart rate variability, we have documented a reduction in both total power and high-frequency spectral power (absolute units) during ANF infusion.21 The implication of this decrease in total spectral power is that ANF reduces spontaneous variations in heart rate in humans. Because power in the high-frequency component of the heart rate variability spectrum was attenuated in these studies, the mechanism responsible for this decreased spontaneous variation would appear to be a reduction in the vagal modulation of heart rate.22 We reasoned that if ANF interfered with the vagal modulation of heart rate, it should attenuate chronotropic responses to ramp increases and decreases in arterial pressure. Therefore, the aim of the present experiment was to test the hypothesis that ANF reduces the spontaneous BRS for heart rate in humans.
| Methods |
|---|
|
|
|---|
Procedures
All studies occurred at the same time of day. Subjects lay
supine in a quiet, motionless, and awake, eye-opened state. An
intravenous catheter was placed in a left forearm vein for infusions. A
finger cuff was placed on the left index finger for continuous
noninvasive beat-by-beat recording of blood pressure23
(Ohmeda 2300 Finapres). Respiratory movements were detected by a
pneumobelt coupled to a P23 1D transducer (Gould Inc). Lead II of the
electrocardiogram was recorded continuously and inscribed by an ink
recorder onto paper, along with mean heart rate, the blood pressure
waveform, and breathing frequency.
Protocol
Each subject was studied on 2 separate days and received either
human ANF (99-126, IAF Biochem) or vehicle (isotonic saline) allocated
at random and in a double-blind manner. After a 15-minute baseline
period, ANF was administered as in our previous
protocols,1 21 first as a bolus of 16 nmol (50 µg) in 10
mL over 3 minutes and then at a rate of 16 pmol/kg per minute (50 ng/kg
per minute) at 0.8 mL/min. The principal hemodynamic effects of this
dose, which increases plasma concentrations to approximately 160 pmol/L
(500 pg/mL),1 are reductions in central venous pressure
and diastolic pressure.1 Saline was given as a bolus of 10
mL over 3 minutes and then at a rate of 0.8 mL/min.
Calculation of Spontaneous BRS
The analog output of the electrocardiographic meter was
discriminated to yield a train of rectangular impulses corresponding to
the QRS spikes. The impulse train was processed on a real-time basis
with a microcomputer at a sampling frequency of 1000 Hz and stored
sequentially for data analysis. Immediately after detecting an R
wave, the computer algorithm tracked the pressure channel and
identified the systolic and diastolic pressures as the highest and
lowest values occurring before the next R wave. Over the last 7 minutes
of each baseline period, and from 16 to 23 minutes into the respective
infusions, sequences of three or more beats in which the systolic
pressure and the RR interval of the immediately following beat changed
in the same direction (either increasing [Up] or decreasing [Down])
were identified and classified as baroreflex sequences. As reported by
others,16 17 18 most of these were three-beat sequences, but
some were four-, five-, or six-beat sequences. The direction of each
sequence (Up or Down) was identified, and a linear regression, relating
RR interval to the systolic pressure of the antecedent cardiac cycle,
was derived for each individual sequence. For each time period
(baseline or infusion), the mean value of the slope of spontaneous BRS
was calculated for each subject, and the number of sequences per minute
per subject was also recorded. Sequences were not accepted for
analysis if the correlation coefficient between systolic pressure
and the subsequent RR interval was less than .85. The number of
baroreflex sequences obtained for each subject was expressed as the
number of sequences per minute per heart rate. Because of possible
hysteresis in blood pressureheart rate relationships,20
the spontaneous BRS for Up and Down sequences for each subject were
determined separately.
Statistical Analysis
Means and their standard errors are reported throughout. A
two-way ANOVA was used to test for significant differences between mean
baseline or preinfusion values for each subject, as well as values
obtained during the respective ANF or saline infusions, and to compare
values obtained during the ANF infusion with values obtained during the
saline infusion. The Student-Newman-Keuls test was used for post hoc
comparison of intervention effects over time against initial values and
for comparison of interaction effects. Statistical significance was
accepted at a value of P<.05.
| Results |
|---|
|
|
|---|
|
|
|
Systolic pressure and respiratory rate did not change during either infusion (Table 1). Saline did not change the mean RR interval (Table 1), whereas ANF infusion decreased mean RR interval from 952±35 to 930±40 ms (P<.04). There was a significant between-group effect of ANF infusion (P<.04) on RR interval: ie, heart rate was greater during ANF infusion than during saline.
Tables 2 and 3 summarize group means of spontaneous baroreflex data obtained during saline and ANF infusions. The number of baroreflex sequences (total, Up, or Down) did not change during either infusion. Mean slopes of the spontaneous BRS for heart rate (BRS) for Up and Down sequences assessed separately and together were stable during the saline infusion. In contrast, ANF lowered the mean BRS slope for all sequences from 32.7±4.8 to 23.1±2.8 ms · mm Hg-1 (P<.04) (Fig 1). Spontaneous BRS for Up sequences decreased from 32.9±5.5 to 24.2±3.3 ms · mm Hg-1 (P<.05). There was no change in the baroreflex slope for Down sequences compared with baseline values before ANF infusion (from 27.9±4.2 to 23.0±3.2 ms · mm Hg-1 [P>.05]) (Fig 2). However, there was a significant ANF-saline interaction for the slope of these Down sequences (F=6.9, P<.05); ie, the slope for Down sequences was significantly shallower during ANF infusion than during saline (35.0±5.2 ms · mm Hg-1).
|
|
| Discussion |
|---|
|
|
|---|
The spontaneous baroreflex method applied in the present experiments has a number of advantages over other methods for estimating the arterial baroreflex control of heart rate. The neck chamber device, for example, is cumbersome, provokes some anxiety, and requires breath holding by the subject to characterize the slope of the heart rate response and perturbs only one afferent component of the reflex under study.19 Consequently, any responses to carotid baroreceptor stimulation are countered by the activity of aortic baroreceptor afferents.9 19 More importantly, heart rate responses to selective carotid activation and deactivation are less than those observed when both sets of arterial baroreceptors act in concert. For example, Ebert and Cowley9 recorded baroreflex slopes of 2 to 4 ms · mm Hg-1 in healthy subjects compared with resting values of 25 ms · mm Hg-1 in the present study or slopes of 15 to 40 ms · mm Hg-1 that would be anticipated from responses to bolus injection of phenylephrine.3 4 19 31 Consequently, the neck collar method has less likelihood of detecting a significant influence of potential modulatory interventions in the baroreceptorheart rate reflex, as might be elicited by ANF, even if these were present.
These limitations may explain why results of previous experiments with the use of invasive or obtrusive methods of testing the arterial baroreflex control of heart rate have been inconsistent. Volpe et al4 reported that ANF (50 ng/kg per minute, or approximately 16 pmol/kg per minute) augmented the reflex bradycardic response to phenylephrine and attenuated the tachycardic response to nitroglycerin in young healthy volunteers, but this neural interaction with ANF, similar to that observed in rats,29 was abolished by pretreatment with the angiotensin-converting enzyme inhibitor enalapril. The reflex increase in heart rate elicited by +20 mm Hg neck pressure was also attenuated by ANF (at a lower dose than in this study: 15 to 25 ng/kg per minute, or approximately 5 to 8 pmol/kg per minute), but the peptide did not affect the reflex bradycardic response to -40 mm Hg of neck suction nor the magnitude of spontaneous respiratory sinus arrhythmia.9 Moreover, attenuated tachycardia was not observed by Volpe et al, who applied more intense neck pressure (+60 mm Hg) during a higher dose of ANF (approximately 160 pmol/kg followed by 16 pmol/kg per minute).30 Also, in our previous experiments, ANF (at the same dose as in the present study) augmented the reflex increase in heart rate in response to hypotensive lower body negative pressure (-40 mm Hg),1 an intervention that also lowers central venous pressure.
ANF reduced the magnitude of the vagally mediated heart rate responses to spontaneous rises and falls in arterial pressure. The bradycardic response to rises in blood pressure fell from 32.9±5.4 to 24.3±3.3 ms · mm Hg-1. Compared with preinfusion baseline, the effect of ANF on the tachycardic response to spontaneous decreases in blood pressure was less striking (from 27.9±4.2 to 23.0±3.2 ms · mm Hg-1).
Because the principal interaction appeared to be with responses to
parasympathetic activation rather than withdrawal, ANF might act on
vagal cholinergic neuroeffector mechanisms, ie, by inhibiting the
release or postjunctional action of acetylcholine. However, when
compared with saline, ANF infusion also reduced the slope of the
spontaneous BRS for heart rate in response to falls in blood pressure
(P<.05), suggesting an additional, perhaps central, effect
of ANF on this reflex. Atchison et al28 32 33 have shown
that ANF augments the parasympathetic regulation of heart rate in rats
through a prejunctional mechanism analogous to
1
antagonism but has no direct effect on the rate of sinoatrial action
potential discharge. Our observations do not establish any evidence for
a similar action by ANF in humans. However, the presence of
immunoreactive ANF receptors for this peptide in and near brain sites
involved in cardiovascular regulation34 35 36 provides the
opportunity for ANF to influence the neural control of heart rate and
vascular resistance via a central effect. For example, in Wistar rats
microinjection of exogenous ANF into the caudal nucleus tractus
solitarius increases the firing rates of nucleus tractus solitarius
neurons,34 reduces arterial blood pressure,34
and blunts the baroreceptor control of heart rate.35
The ANF dose given in the present study increases plasma concentrations beyond the normal physiological range in young adults, to levels observed in healthy elderly subjects and in conditions such as congestive heart failure.37 38 However, replication of our previous infusions allows us to interpret our present observations within the framework of the hemodynamic and neuromodulatory effects of ANF documented in previous experiments.1 21 Systolic pressure was unaffected, indicating that the ANF dose administered in these experiments did not alter the set point of the arterial baroreflex. In addition, as the mean slopes of chronotropic responses to increases and decreases in arterial blood pressure were similar under both saline and ANF conditions (Fig 2, Table 1), our observations relate to the gain of this reflex across the linear portion of its operating curve. Finally, these effects of ANF, on either heart rate variability or spontaneous BRS, cannot be attributed to changes in breathing frequency, as this variable did not alter.
These observations allow us to conclude that, in addition to inhibiting sympathetic outflow to muscle1 and the sinoatrial node,21 this ANF dose attenuates the spontaneous baroreflex regulation of heart rate by the parasympathetic nervous system. The latter likely contributes to the mild positive chronotropic effects of ANF infusion in humans.1 10 11 21
| Acknowledgments |
|---|
Received July 8, 1994; first decision November 22, 1994; accepted February 6, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Kasama, T. Toyama, H. Kumakura, Y. Takayama, T. Ishikawa, S. Ichikawa, T. Suzuki, and M. Kurabayashi Effects of Intravenous Atrial Natriuretic Peptide on Cardiac Sympathetic Nerve Activity in Patients with Decompensated Congestive Heart Failure J. Nucl. Med., July 1, 2004; 45(7): 1108 - 1113. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Gori, J. S. Floras, and J. D. Parker Effects of nitroglycerin treatment on baroreflex sensitivity andshort-term heart rate variability in humans J. Am. Coll. Cardiol., December 4, 2002; 40(11): 2000 - 2005. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. L. Abramson, S.-i. Ando, C. F. Notarius, G. A. Rongen, and J. S. Floras Effect of Atrial Natriuretic Peptide on Muscle Sympathetic Activity and Its Reflex Control in Human Heart Failure Circulation, April 13, 1999; 99(14): 1810 - 1815. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Sigaudo, J.-O. Fortrat, A.-M. Allevard, A. Maillet, J.-M. Cottet-Emard, A. Vouillarmet, R. L. Hughson, G. Gauquelin-Koch, and C. Gharib Changes in the sympathetic nervous system induced by 42 days of head-down bed rest Am J Physiol Heart Circ Physiol, June 1, 1998; 274(6): H1875 - H1884. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Willenbrock, H. Stauss, M. Scheuermann, K. J. Osterziel, T. Unger, and R. Dietz Effect of chronic volume overload on baroreflex control of heart rate and sympathetic nerve activity Am J Physiol Heart Circ Physiol, December 1, 1997; 273(6): H2580 - H2585. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-i. Ando, H. R. Dajani, B. L. Senn, G. E. Newton, and J. S. Floras Sympathetic Alternans: Evidence for Arterial Baroreflex Control of Muscle Sympathetic Nerve Activity in Congestive Heart Failure Circulation, January 21, 1997; 95(2): 316 - 319. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |