(Hypertension. 2000;36:543.)
© 2000 American Heart Association, Inc.
Scientific Contributions |
From the Departments of Clinical Physiology (M.J., P.F.), Clinical Neurophysiology (M.E.), Cardiology (B.R.), and Nephrology (H.H., G.J.), Göteborg University, Sahlgrenska University Hospital, Göteborg, Sweden; and National Institutes of Neurological Disorders and Stroke, Bethesda, Md (G.E.).
Correspondence to Mats Johansson, MD, PhD, Department of Clinical Physiology, Sahlgrenska University Hospital, SE 413 45 Göteborg, Sweden. E-mail m.johansson{at}medfak.gu.se
| Abstract |
|---|
|
|
|---|
Key Words: hypertension, renovascular sympathetic nervous system renin-angiotensin system
| Introduction |
|---|
|
|
|---|
The aim of the present study was to assess the effect of short-term ACE inhibition on the adrenergic drive in patients with hypertension and renal artery stenosis. Given that animal experiments indicate different effects on renal nerve activity and heart rate in response to Ang II,9 we also wanted to explore whether there is regional differentiation in the sympathetic response to ACE inhibition and nonspecific vasodilatation, with the latter used as a positive control test.
Patients were divided into 2 groups given either a single dose of the ACE inhibitor enalaprilat or a single dose of the nonspecific vasodilator dihydralazine, in an attempt to reduce blood pressure to a similar extent. In this model, the former drug prevents Ang II production and the latter increases Ang II plasma concentrations. The degree of sympathetic activity was estimated by means of NE isotope dilution and direct recordings of sympathetic nerve traffic in the peroneal nerve.
| Methods |
|---|
|
|
|---|
The study group comprised patients with hypertension undergoing a
clinical investigation for renovascular hypertension, involving renal
vein blood sampling for assessment of plasma renin activity (PRA)
(Table). The present study population
represents a subset of patients referred to in previous
publications.2 10 All patients had hypertension and renal
artery stenosis
50% according to angiography. Overall, 57
patients were examined. Nine patients were excluded because of vagal
reactions during the catheterization procedure, and
therefore 48 patients were included in the study.
|
Catheterization
Subjects were studied in the morning in a
catheterization laboratory. Subjects refrained from
smoking and coffee drinking 12 hours before the study. All patients
were hospitalized 4 days before catheterization, and
while they were inpatients they were kept on a low salt diet (40
mmol/24 h). Diuretics and calcium channel blockers were given
during the hospital stay, but other antihypertensive medications were
withdrawn 2 days before the investigation. The proportion of patients
medicated with calcium channel blockers and diuretics did not
differ between the 2 intervention groups. Calcium channel blockers were
used by 83% of the patients in the enalaprilat group and by 70% in
the dihydralazine group. Diuretics were taken by 65%
of the patients in the enalaprilat group and by 70% in the
dihydralazine group. No antihypertensive drugs were given for
the 12 hours preceding the investigation. A cannula was introduced
percutaneously into a left radial artery for blood
pressure monitoring and blood sampling. Both renal veins were
catheterized via femoral veins, and the renal vein catheters were
positioned under fluoroscopic control, with their positions confirmed
by means of oxygen saturation.
Infusions
Para-aminohippurate (PAH) (Merck Sharp & Dohme), with
dosing depending on estimated glomerular filtration rate,
and tracer doses of L-2,5,6-[3H]NE
(40 to 60 Ci/mmol, New England Nuclear) were infused into a
peripheral vein. An infusion rate of 1.0 to 1.5 µCi/min
of [3H]NE was used.
Sympathetic Nerve Recordings
Multiunit postganglionic sympathetic nerve activity was
recorded with a tungsten microelectrode, with a tip diameter of a
few micrometers, inserted into a muscle-innervating
fascicle of the peroneal nerve at the fibular head. A reference
electrode was inserted subcutaneously 1 to 2 cm from the
recording electrode. Details regarding the recording
technique and the criteria for muscle sympathetic nerve activity have
been described previously.11 The number of muscle
sympathetic nerve activity bursts in the mean voltage neurogram, which
occur in bursts strictly coupled to the cardiac rhythm, were counted by
inspection of the mean voltage neurogram. Total muscle sympathetic
nerve activity is the cumulative burst area per minute (arbitrary
units) in the mean voltage neurogram and is presented as
percentage of prestimulus control.
Two independent laboratory colleagues, who were not part of the study and had no knowledge of the study protocol, performed the analysis. For interindividual comparisons of baseline activity, muscle sympathetic nerve activity is presented as burst frequency (bursts per minute) calculated from a 5-minute resting period and during a 60-second period starting 6 minutes after enalaprilat or dihydralazine administration.12 Total muscle sympathetic nerve activity was determined for each individual during successive 60-second periods up to 8 minutes after enalaprilat or dihydralazine administration. Muscle sympathetic nerve activity recordings were available for 11 of 11 subjects up to 6 minutes and for 9 of 11 patients at 8 minutes.
Experimental Protocol
Patients were divided into 2 groups given
intravenously either a single dose of 1.25 mg enalaprilat
(n=27) or a single dose of 6.25 mg dihydralazine (n=21) during
3 minutes. All blood samples were taken simultaneously from
a radial artery and both renal vein(s) at steady state, at least 30
minutes after the [3H]NE and PAH infusions were
started. Sampling was performed at baseline and 30 minutes after drug
administration. Samples were collected into ice-chilled tubes
containing heparin or EDTA and glutathione, including a renin
inhibitor for the Ang II tubes. Plasma was separated by
centrifugation and stored at -80°C until assayed for
catecholamines, PRA, and Ang II. Renal plasma flow was
derived from total infusion clearance of PAH corrected for renal
fractional extraction, and separate renal plasma flow was assessed by
gamma camera renography.
Of the 11 patients in whom muscle sympathetic nerve activity was recorded in addition to measurements of NE spillover, 6 received enalaprilat and 5 dihydralazine. Muscle sympathetic nerve activity was measured at baseline and continuously up to 8 minutes after drug administration. In 5 patients to whom enalaprilat was administered, muscle sympathetic nerve activity was recorded up to 30 minutes after drug administration.
Assays
Catecholamines were extracted from plasma (1 mL) and
samples of infusate (10 µL) with the use of alumina adsorption and
separated by high-performance liquid
chromatography. Timed collection of
[3H] eluate leaving the electrochemical cell
permitted separation of [3H]NE for subsequent
counting by liquid scintillation spectroscopy.13
Interassay coefficients of variation were 4.6% for
endogenous NE and 3.2% for
[3H]NE.
PRA was measured according to Giese et al,14 with the use of radioimmunoassay for angiotensin I (Ang I). EDTA was added to inhibit ACE. The interassay coefficient of variation for the renin analysis was 8.8%. The reference values are 0.2 to 2.0 ng Ang I per milliliter per hour. Ang II was assayed according to Kappelgaard et al15 and Morton and Webb.16 Bestatin was used to specifically inhibit aminopeptidase B and leucine aminopeptidase. The interassay coefficient of variation was 5.1%. The radioactive recovery for 125I-Ang II was 87% and the biological recovery 82% in duplicate determinations.
Calculations
Renal NE spillover was estimated with the use of Ficks
principle corrected for the fractional extraction across the kidney,
as follows:
![]() |
Total body NE spillover was measured by the radiotracer
method17 and calculated according to the following
formula:
![]() |
![]() |
Statistical Methods
Results are expressed as mean±SEM values. Students
t tests for paired and unpaired observations were used.
Parameters not normally distributed were transformed
logarithmically before the parametric test. If a nonnormal
distribution was retained, the Wilcoxon signed rank test for
paired comparisons and the Mann-Whitney U test for unpaired
comparisons were used. We assessed the relation between 2 variables
by calculating the rank correlation coefficient according to Spearman.
Effects of enalaprilat and dihydralazine on total muscle
sympathetic nerve activity over time were compared with an ANOVA for
repeated measurements, with time after drug administration as
within-subjects factor and drug as between-group factor. Statistical
significance was defined as P<0.05.
| Results |
|---|
|
|
|---|
Arterial pressure decreased (P<0.01) to a similar extent after enalaprilat and dihydralazine, whereas heart rate increased only after dihydralazine and remained unchanged after enalaprilat administration (Figure 1, top panel). Renal plasma flow increased to a similar extent after both enalaprilat and dihydralazine administration (Figure 1; P<0.01 for both).
|
PRA increased after both enalaprilat and dihydralazine stimulation, with a more pronounced increase for the former drug (348% increase after enalaprilat versus 134% increase after dihydralazine). As expected, plasma Ang II concentrations decreased by 68% after enalaprilat, whereas dihydralazine administration resulted in a 73% increase. The magnitude of the blood pressure reduction was correlated with the preintervention PRA (r=0.53, P<0.01) in enalaprilat-treated patients, whereas no correlation was found in the dihydralazine group.
Arterial plasma NE concentrations increased after
dihydralazine (from 3.36±0.39 to 5.03±0.68 pmol/mL;
P<0.01), whereas they remained unchanged after enalaprilat
administration (4.32±0.44 before and 4.56±0.37 pmol/mL after
enalaprilat). Total body NE spillover increased by 48%
(P<0.01) after dihydralazine, whereas no change was
seen after enalaprilat administration (Figure 1). Renal NE
spillover increased by 44% after enalaprilat (P<0.01;
Figure 1, bottom panel) and tended to increase after
dihydralazine administration (
15%). There was no
correlation between the change in plasma flow and the change in renal
NE spillover in either the enalaprilat or the dihydralazine
group.
Muscle sympathetic nerve activity measured at baseline and 6 minutes after dihydralazine increased from 51±6 to 59±7 bursts per minute (P<0.05), whereas it remained unchanged after enalaprilat administration, at 56±3 before and 57±3 bursts per minute after (Figure 1, bottom panel). Muscle sympathetic nerve activity also remained unchanged 30 minutes after enalaprilat administration in the 5 patients in whom extended recording periods were performed. Total muscle sympathetic nerve activity (cumulative burst area per minute) increased after dihydralazine (P=0.01; Figure 2), whereas it remained unchanged after enalaprilat administration. There were positive correlations between the change in total body NE spillover and Ang II (r=0.43; P<0.01) after drug administration (with enalaprilat- and dihydralazine-treated patients considered together). Similarly, the change in muscle sympathetic nerve activity after drug administration correlated with the changes in Ang II (r=0.64; P<0.05), whereas no relationship was found between the change in renal NE spillover and Ang II levels.
|
| Discussion |
|---|
|
|
|---|
Although the present study cannot delineate the specific mechanism responsible for the differentiated effect of ACE inhibition on sympathetic nerve activity, there are several possible mechanisms. First, Ang II acts presynaptically in sympathetic nerves to facilitate NE release.8 This mechanism may explain the inhibited sympathetic response after enalaprilat since blood pressure reduction was accompanied by a fall in Ang II plasma concentrations. However, directly recorded muscle sympathetic nerve activity increased after dihydralazine, whereas no change was seen after enalaprilat. This is consistent with a central site of action.
Enalaprilat can reach cerebral tissue through regions without an intact blood-brain barrier and thus directly reduce the activity of the tissue-bound renin-angiotensin system. A reduced circulating level of Ang II, crossing the blood-brain barrier, is another effect of ACE inhibition. Matsumara et al18 found a blunted baroreflex inhibition of heart rate when Ang II was infused into the vertebral artery compared with intravenous infusion of Ang II, indicating a central effect. Furthermore, inhibition of ACE causes accumulation of bradykinin and increased formation of prostaglandin and nitric oxide, which also may affect central sympathetic outflow.19
Although there is some disagreement as to whether Ang II modulates baroreceptor reflex control of heart rate by changing the sensitivity or resetting the reflex control of heart rate to a higher blood pressure level, the latter mechanism is now generally accepted.4 20 Consequently, the arterial pressure increase due to infusion of Ang II is followed by a modest bradycardia, contrasting with a marked bradycardia in response to infusion of phenylephrine. When an ACE inhibitor is administered, plasma concentrations of Ang II fall and the baroreflex curve resets to a lower arterial pressure.20 A resetting of the baroreflex control of heart rate provides an explanation for the lack of heart rate increase after enalaprilat administration in the present study.
Unchanged heart rate and arterial plasma NE concentrations after enalaprilat, in conjunction with an increase of both these variables after dihydralazine administration, are in agreement with the findings of Herlitz et al.21 Moreover, Ligtenberg et al4 recently reported a blunted muscle sympathetic nerve response to short-term treatment with enalapril compared with the calcium channel blocker amlodipine. Rongen et al22 reported reduced blood pressure and unchanged total body NE spillover and heart rate after a 1-week treatment of losartan in healthy subjects. Thus, the present findings of unchanged overall and muscle sympathetic nerve activity after short-term ACE inhibition are in agreement with previous results.
The novel findings of a differentiated sympathetic response pattern after enalaprilat in patients with renal artery stenosis corroborate the results of animal experiments.9 23 In conscious rabbits, Kumagai and Reid9 found that both an ACE inhibitor and an angiotensin type I receptor antagonist decreased arterial pressure and increased renal sympathetic nerve activity, whereas heart rate was unchanged. Thus, Ang II resets the baroreflex control of heart rate, leaving the control of renal nerve activity unaffected. In contrast, DiBona et al24 found decreased renal sympathetic nerve activity after intravenous angiotensin type I receptor blockade, suggesting that endogenous Ang II tonically influences renal baroreflex control in rats on a low or normal but not on a high sodium diet. These contrasting results in animal studies may be due to different experimental approaches. DiBona et al24 used infusions of methoxamine after losartan administration to restore arterial pressure to baseline values before measuring renal sympathetic nerve activity. Experimental data indicate that arterial baroreceptor activity may respond to constriction and stiffening of aortic smooth muscle, leaving arterial pressure unaffected.25 Hence, arterial baroreceptors may have been activated by methoxamine, although arterial pressure and filling pressure of the left heart were similar to baseline values.
In agreement with Noll et al,26 who found unchanged muscle sympathetic nerve activity in conjunction with reduced diastolic blood pressure after oral short-term administration of captopril to healthy subjects, we found unchanged muscle sympathetic nerve activity after intravenous administration of enalaprilat, suggesting resetting of the baroreflex control of muscle sympathetic nerve activity to lower blood pressures. This notion is further supported by the findings of Matsukawa et al,27 who examined muscle sympathetic nerve activity in healthy subjects during equipotent infusions of Ang II and phenylephrine. They found a dose-dependent reduction of muscle sympathetic nerve activity after Ang II that was smaller than during the infusion of phenylephrine.
When the effects of dihydralazine and enalaprilat are compared, it is important to consider the effects on hemodynamics. The former drug is a vasodilator that preferentially dilates arterial blood vessels, whereas the latter also dilates capacitance vessels, thereby reducing the filling pressures of the heart.28 Hence, it is conceivable that enalaprilat exerts a more pronounced unloading effect on cardiac low-pressure baroreceptors than dihydralazine. Unloading of cardiac low-pressure baroreceptors preferentially activates renal sympathetic nerve activity with little effect on abdominal or peripheral sympathetic outflows.29 In support of this contention, we have found that short-term enalaprilat administration to healthy subjects causes a slight reduction in mean arterial pressure, whereas left ventricular filling pressure is markedly reduced (M. Johansson et al, unpublished data, 1999).
Study Limitations
Regional plasma flow affects NE spillover measurements, and an
increase in plasma flow may increase regional NE spillover by a
washout effect.30 In the present study, renal plasma
flow increased by only 17% after enalaprilat, whereas renal NE
spillover showed a 44% increase. Moreover, there was no
correlation between the change in renal NE spillover and the change
in renal plasma flow after enalaprilat administration, and therefore a
major washout effect on renal NE spillover by increased renal
plasma flow after enalaprilat administration seems unlikely.
Furthermore, reduced blood flow and glomerular filtration
rate in stenotic kidneys may have affected the
estimation of renal sympathetic nerve activity by the isotope dilution
method. However, we obtained similar results by analyzing the material
after excluding stenotic kidneys (renal NE spillover
increased by 46% in nonstenotic kidneys after enalaprilat;
P=0.01). Although the isotope dilution method has its
limitations, we advocate that it is the best method available for
providing an index of efferent renal sympathetic nerve activity in
humans.
To keep blood pressure under control, treatments with a calcium channel blocker and/or diuretics (80% and 69% of the study population, respectively) were sustained during the examination. These treatments may have affected both the baseline values for sympathetic nerve activity and the response to the drug intervention. Moreover, treatment with diuretics is known to activate the renin-angiotensin system. Although the results of long-term treatment with calcium channel blockers on sympathetic activity have been variable, there is a possibility of sympathetic reflex activation as a response to the vasodilatory effect of these drugs. However, total body NE spillover did not differ among patients treated or not treated with diuretics or a calcium channel blocker. Hence, an important drug or drug withdrawal effect on overall sympathetic nerve activity seems unlikely.
The design of the present study did not mimic completely the clinical situation of long-term treatment with ACE inhibitors, but the results may still have clinical implications because of the effects of reduced salt and water excretion by increased renal sympathetic nerve activity.31
In conclusion, the blood pressure reduction after short-term administration of an ACE inhibitor to hypertensives with renal artery stenosis is associated with lack of overall and muscle sympathetic reflex activation but preserved reflex increases in renal sympathetic activity. These data thus suggest a differentially regulated sympathetic outflow in this condition. Moreover, the concomitant reduction of Ang II plasma concentrations does not support a stimulatory effect of endogenous Ang II on renal sympathetic nerve activity in these patients.
| Acknowledgments |
|---|
Received November 3, 1999; first decision December 8, 1999; accepted April 26, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. Grisk and R. Rettig Interactions between the sympathetic nervous system and the kidneys in arterial hypertension Cardiovasc Res, February 1, 2004; 61(2): 238 - 246. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. F. DiBona The Sympathetic Nervous System and Hypertension: Recent Developments Hypertension, February 1, 2004; 43(2): 147 - 150. [Full Text] [PDF] |
||||
![]() |
T. W. Lameris, S. de Zeeuw, D. J. Duncker, G. Alberts, F. Boomsma, P. D. Verdouw, and A. H. van den Meiracker Exogenous Angiotensin II Does Not Facilitate Norepinephrine Release in the Heart Hypertension, October 1, 2002; 40(4): 491 - 497. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |