(Circulation. 2001;103:263.)
© 2001 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Section of Hypertension (F.E., C.L.L., E.A.), University of Texas Medical Branch at Galveston; Hypertension and Atherosclerosis Section (H.G., M.R.B.), Boston University School of Medicine, Boston, Mass; and the MRC Multidisciplinary Research Group on Hypertension (E.L.S.), Clinical Research Institute of Montreal, University of Montreal, Montreal, Quebec, Canada.
Correspondence to Fernando Elijovich, MD, Michigan State University, Medical Education and Research Center, 333 Bostwick Ave, Grand Rapids, MI 49503. E-mail Fernando_Elijovich{at}grmerc.net
| Abstract |
|---|
|
|
|---|
Methods and
ResultsPlasma endothelin was measured in 47
patients with essential hypertension. Endothelin, catecholamine, and
plasma renin activity (PRA) responses to 24-hour sodium deprivation
(
Na) were assessed in 29 of these patients. Endothelin was higher in
hypertensive patients (4.6±0.2 fmol/mL) than in 20 control subjects
(3.3±0.3 fmol/mL, P<0.002),
was correlated with BP, and was negatively associated with PRA
(P<0.04). Salt-sensitive,
salt-resistant, and indeterminate groups were defined by the tertiles
of the t statistic for the
difference in BP before and after
Na. Systolic BP falls were
-15±1, -2±2, and -9±1 mm Hg, respectively. PRA, its response
to
Na, and its level after
Na were lowest (albeit nonsignificant)
in salt-sensitive patients. Baseline catecholamine and endothelin
levels did not differ among the groups. In response to
Na,
catecholamines increased more significantly in salt-sensitive patients
(+2.4±0.9 nmol/L) than in the other groups (0.4±0.2 and 0.7±0.2
nmol/L for indeterminate and salt-resistant groups, respectively;
P<0.03), whereas endothelin
increased in the salt-sensitive group (0.8±0.3 fmol/mL), decreased in
the salt-resistant group (-0.4±0.3 fmol/mL), and sustained minimal
change in the indeterminate group (0.2±0.3 fmol/mL)
(P<0.04). Thus, endothelin
levels in the salt-depleted state were highest in salt-sensitive
patients (5.2±0.4 fmol/mL) versus the other groups (3.4±0.4 and
4.4±0.4 fmol/mL for salt-resistant and indeterminate groups,
respectively) (P<0.02).
Changes in endothelin during
Na and levels after
Na were
correlated with changes in catecholamines
(P<0.02).
ConclusionsOur data suggest that salt-depleted salt-sensitive hypertensives with blunted renin responses exhibit enhanced catecholamine-stimulated endothelin levels and may therefore respond better than unselected patients with essential hypertension to endothelin receptor blockers.
Key Words: endothelin sodium catecholamines renin hypertension
| Introduction |
|---|
|
|
|---|
In human essential hypertension, circulating levels of endothelins have been found to be elevated by some6 7 but not all8 9 investigators. Expression of endothelin-1 in the endothelium of small resistance arteries is enhanced in patients with moderate to severe hypertension.10 Because endothelins have a predominantly paracrine action, with secretion from the endothelium toward the medial layer of the vessel,11 their plasma levels primarily reflect spillover to the circulation, not necessarily the degree of activation of the endothelin system. However, unequivocal evidence for a role of endothelins in human hypertension was provided by the demonstration that bosentan, a specific antagonist of endothelin type A (ETA)/endothelin type B (ETB) receptors, significantly lowers BP in these patients.12
Analogous to the data in rats, salt sensitivity of BP in humans may constitute a phenotype that predicts BP dependence on the vasoconstrictor action of endothelins. For example, plasma endothelin is higher in (1) hypertensive African American subjects than in hypertensive white subjects,13 (2) low-renin essential hypertensive subjects than in their normal- or high-renin counterparts,14 (3) obese hypertensive subjects than in obese normotensive subjects,15 and (4) salt-sensitive (SS) hypertensive subjects than in salt-resistant (SR) hypertensive subjects, classified with a dietary protocol.16 Furthermore, endothelin is correlated with plasma insulin levels in essential hypertensive subjects.17
Despite the apparent relationship between salt sensitivity
of BP and a role for endothelin in the development or maintenance of
these forms of hypertension, there are no studies to date exploring the
regulation of plasma endothelin by changes in salt balance in human
essential hypertension. We investigated the effect of salt deprivation
(
Na) on plasma endothelin levels in salt-replete essential
hypertensive patients. The BPs of participating subjects were
classified as SS or SR according to their response to an established
protocol of acute
Na in an inpatient
setting.18
| Methods |
|---|
|
|
|---|
Twenty-nine patients were admitted to the General Clinical
Research Center of the University of Texas Medical Branch to undergo a
protocol of
Na after salt loading, similar to that described by
Weinbergers group (Grim et
al18 ). In brief, on
awakening the morning after admission, they were placed on a diet
containing 160 mEq NaCl and were given a 2-L infusion of normal saline
from 8:00 AM to 12:00 noon
(HiNa day). On the following day, the diet was changed to 10 mEq NaCl
per 24 hours, and furosemide (40 mg) was given orally at 8:00
AM, 12:00 noon, and 4:00
PM (LoNa day).
Blood samples for routine tests, plasma renin activity
(PRA), plasma catecholamines (epinephrine and norepinephrine), and
plasma endothelin were obtained each morning at 8:00
AM, immediately before
sodium loading, immediately before furosemide, and after the
Na day.
PRA was measured by
radioimmunoassay,19 and
plasma catecholamines were measured by radioenzymatic assay (BioTrak
TRK 995, Amersham).
Plasma immunoreactive endothelin was measured by radioimmunoassay as previously described.8 10 In brief, blood was collected in potassium EDTA tubes, and endothelin was extracted from plasma by passage through C18 Sep-Pak cartridges (Waters Associates). The antibody against endothelin-1 was from Peninsula. The minimum detectable concentration of endothelin was 0.4 fmol/mL. Recovery of 5 fmol/mL endothelin-1 added to plasma was 75%. Cross-reactivity of the antibody was 17% with big endothelin-1, 7% with endothelin-2 and -3, and none with unrelated peptides. The interassay and intra-assay coefficients of variation were <15% and <10%, respectively.
BP was continuously monitored over the entire
hospitalization with the Spacelabs monitor, as described above. The
number of readings in the 29 patients for the period of noon to 10
PM was as follows: 37±1
for HiNa and 33±1 for LoNa. The effect of
Na on BP was assessed by
2 different methods: (1) the absolute BP change in mm Hg from the
salt-replete to the salt-depleted state (mean of all readings of the
LoNa minus that of the HiNa days), and (2) the value of the
t statistic (unpaired Student
t test) for the comparison, in
each patient, of all systolic BP readings of the HiNa and LoNa days
(noon to 10 PM). This
parameter,
![]() |
Na, respectively. Hence, the more salt sensitive
the patient is, the larger is the
t value. Patients were
classified into SS (n=10), indeterminate (IN, n=10), and SR (n=9)
groups, according to whether their
t values fell within the upper,
middle, or lower tertiles of the distribution of the
t statistic in the total
population, respectively. ECG left ventricular hypertrophy (LVH) was assessed by the Cornell ([RaVL+SV3 mm] · QRS ms) and Sokolow-Lyon (SV2+RV5 mm) indices, with maximum normal cutoffs being 2440 mm · ms and 35 mm, respectively.
Data are presented as mean±SE. Comparisons of means between 2 groups were carried out by unpaired Student t test, and those between 3 groups were carried out by 1-way ANOVA followed by contrasting of means with the Tukey-Kramer test. Correlation coefficients were calculated with the Pearson method. All statistical tests and fitting of regression lines were performed with the JMP software (version 3.0.2) of the SAS Institute. A value of P<0.05 was used to reject the null hypothesis.
| Results |
|---|
|
|
|---|
|
Plasma levels of immunoreactive endothelin were higher in
the whole group of hypertensive patients (4.6±0.2 fmol/mL) and also in
the subset of 29 who participated in salt-sensitivity studies (4.4±0.3
fmol/L) than in 20 normotensive control subjects from our laboratory
(3.3±0.3 fmol/L, P<0.002).
However,
Figure 1
shows that there was overlap of endothelin levels
between hypertensive and normotensive subjects.
|
Baseline plasma endothelin of hypertensive patients
exhibited a positive correlation with systolic and diastolic BPs
obtained in the office and with the average systolic and diastolic BPs
obtained with ambulatory monitors. The variability of plasma endothelin
was explained better by ambulatory than by office BPs
(Figure 2
).
|
Plasma endothelin levels were higher in male (5.2±0.4
fmol/mL) than in female (4.3±0.2 fmol/mL) hypertensive patients
(P<0.04), but not different
between white (4.8±0.2 fmol/mL) and African American (4.0±0.5
fmol/mL) patients (P=NS, not
shown). The sex difference in plasma endothelin was probably due to
higher BPs in males (159±5/102±3 mm Hg) than in females
(152±3/93±2 mm Hg)
(P=NS/P<0.01).
Baseline plasma endothelin (1) was not higher in patients with
hyperlipidemia (LDL cholesterol >3.36 mmol/L) or LVH (Cornell >2440
mm · ms) than in their counterparts and (2) was not correlated with
age, serum creatinine, serum lipids, or indices for LVH in the ECG. In
contrast, there was a significant negative correlation between baseline
plasma endothelin and PRA
(r=-0.32,
P<0.04), best fit by the
logarithmic function shown in
Figure 3
.
|
Baseline systolic and diastolic BPs of patients classified
as SS (155±7/96±4 mm Hg), IN (152±7/93±4 mm Hg), and SR
(153±7/91±4 mm Hg) were not significantly different. The systolic BP
changes of these 3 groups in response to
Na were -15±1, -9±1,
and -2±2 mm Hg, respectively.
Table 2
shows PRA, catecholamine, and endothelin
levels in the SS, IN, and SR groups before and after
Na (ie, in the
salt-replete and salt-depleted states) and also their changes in
response to salt depletion. Baseline PRA (not shown) and PRA in the
salt-replete and salt-depleted states were somewhat lower in the SS
group than in the other 2 groups. Also, the PRA response to salt
depletion was blunted in the SS group. However, none of these
differences reached statistical significance. Baseline plasma
catecholamine levels (not shown) and the levels after salt loading were
not different among the groups. However, their responses to salt
depletion were significantly larger in the SS group, leading to higher
(albeit nonsignificant) plasma levels in salt-depleted SS patients.
Finally, baseline endothelin (not shown) and endothelin after salt
loading were not different among the SS, IN, and SR groups. However,
the responses of plasma endothelin to salt depletion were in opposite
directions for the SS (increase) and SR (decrease) groups. The change
in the IN group was not statistically different from zero. The ANOVA
for the mean changes in the 3 groups was significant, and the
Tukey-Kramer test confirmed that this was due to the difference between
the SS and SR groups. As a result of this, plasma endothelin in
salt-depleted SS patients was significantly higher than that in the
other 2 groups.
|
To confirm that different endothelin responses to
Na in
the SS, IN, and SR groups were not due to the arbitrary cutoffs
(tertiles of the t statistic)
chosen to define these groups, we analyzed the relationship between
individual salt sensitivity of BP and endothelin response to salt
depletion as continuous variables. The scatterplots and regression
analyses in
Figure 4
show that there were continuous significant
relationships between the change in plasma endothelin and either the
fall in systolic BP or the t
statistic for this fall during
Na. The variability of the plasma
endothelin responses to
Na was explained better by the
t statistic than by the
absolute BP fall in mm Hg (21% versus 14%), supporting the view that
the former is probably a better metric for salt sensitivity of BP (see
Methods).
|
Finally, the changes in plasma endothelin due to
Na, as
well as the resulting plasma endothelin levels in the salt-depleted
state, were correlated with the changes in plasma catecholamines due to
Na
(Figure 5
). No such relationships were observed between
endothelin and renin or between catecholamines and
renin.
|
| Discussion |
|---|
|
|
|---|
We found no other correlates of plasma endothelin in our patients. Slightly increased levels in males were probably due to higher BPs in subjects of that sex. The lack of racial differences13 was perhaps due to equal severity of hypertension or to equal prevalence of salt sensitivity of BP in our African American and white patients. Although significant correlations between ECG left ventricular mass and plasma endothelin have previously been reported in obese hypertensive patients,15 we could not confirm these by ECG indices of LVH. This may be due to less accurate assessment of left ventricular size or less variability in left ventricular mass in our mildly hypertensive patients.
Acute changes in salt intake uncovered marked differences in hormonal responses between SS and SR patients. We cannot speculate whether these differences would have been elicited by a chronic (eg, dietary) protocol of salt loading and unloading.
Although the groups were relatively small and differences in
renin did not reach statistical significance, our SS patients had
somewhat lower baseline PRA and blunted renin responses to
Na. These
2 characteristics have been repeatedly observed in salt-sensitive
hypertension (see review21 ).
PRA responses to acute22 or
to dietary23 salt depletion
predict the BP fall in these patients. Because their PRA responses to
other stimuli (eg,
orthostasis24 ) are also
blunted, the most likely explanation for these observations is an
intrinsic defect in the responsiveness of the renin-angiotensin system
and not an effect of preexisting volume expansion on the renin response
to salt depletion.
The SS subjects in the present study had normal baseline catecholamines but an exaggerated increase in response to salt depletion. A role for hyperactivity of the sympathetic nervous system in salt-sensitive hypertension with a blunted renin-angiotensin system is controversial.21 However, these patients have exaggerated orthostatic rises in plasma catecholamines and impaired suppression of their plasma levels by a salt load.24 These observations, taken together with ours, suggest that overactivity of the sympathetic nervous system is present, either as a primary abnormality or as compensation for impaired responsiveness of the renin-angiotensin system in the salt-replete and salt-depleted states of SS hypertensive patients.
It has been suggested that the endothelin system is stimulated in a compensatory manner when a blunted renin-angiotensin system is incapable of maintaining BP.14 Although our SS patients did not have higher plasma endothelin compared with SR subjects, there was a significant negative correlation between endothelin and PRA in all patients. This could be explained by endothelin-induced inhibition of renin release25 but is more likely to represent higher endothelin levels secondary to a low renin state, as proposed by others.14 26
Finally, as demonstrated in the rat kidney, the normal response to salt depletion is inhibition of endothelin generation.27 This probably accounts for compensatory reduction of ETB-induced natriuresis. Therefore, the decrease of plasma endothelin by salt depletion in SR subjects is consistent with a physiological response, whereas the increase in SS patients suggests an abnormal response, perhaps triggered by blunting of the pressor role of the renin-angiotensin system in these subjects.
The mechanism by which plasma endothelin levels increase
during
Na in SS patients is unknown. Correlations between
catecholamine and endothelin responses to
Na suggest that synthesis,
release, or spillover of endothelin may be stimulated by catecholamines
in the vasculature or
neurohypophysis.28 Previous
observations supporting this include the following: (1) required
integrity of the endothelium for norepinephrine to exert its full
vasoconstrictor effect,29
(2) increased expression of endothelin-1 mRNA in response to
norepinephrine in the left ventricle of
rats,30 and (3)
costimulation of plasma norepinephrine and endothelin by mental stress
in normotensive offspring of hypertensive
patients.31
An alternative explanation for increased plasma endothelin in salt-depleted SS subjects could be an impairment of its clearance by ETB receptors.32 This would constitute a specific abnormality, because the normal response of renal ETB expression to volume depletion is upregulation.27 Decreased expression of vasodilator/clearance ETB receptors would be consistent with impaired endothelium-dependent vasodilation in SS hypertensive patients,16 33 but it has not been reported. Abnormal downregulation of ETB during salt depletion could be due to decreased natriuretic peptides34 or increased plasma catecholamines.35 Decreased urinary excretion of endothelin in SS hypertensive patients36 suggests preserved, rather than impaired, ETB clearance of endothelin, although it could also be caused by diminished endothelin synthesis by the renal medulla, analogous to observations in SHR.37
Regardless of its mechanism, increased plasma endothelin in salt-depleted SS hypertension has therapeutic implications. The response of unselected hypertensive patients to bosentan, the ETA/ETB receptor blocker, was significant but not greater than the response of normotensive humans.12 This could imply a lack of a specific abnormality of the endothelin system in essential hypertension and simple removal of normal endothelin-dependent vasoconstrictor tone by bosentan. Alternatively, we suggest that the modest reduction of BP in the bosentan trial was due to lack of targeting of therapy to a specific hypertensive phenotype in which endothelin may play a specific pathogenic vasoconstrictor role. Our data suggest that salt-depleted (diet or diuretics) SS hypertensive patients should be the target population for endothelin receptor blockers.
Whether these patients will benefit more from pure ETA versus combined ETA/ETB blockers is unresolved. In heart failure, overexpression of vasoconstrictor ETB receptors in the smooth muscle predicts a more powerful action for ETA/ETB blockers.38 A recent demonstration of ETB-mediated vasoconstriction in the forearm vascular bed of essential hypertensive patients suggests that this may also be the case in hypertension.39
In conclusion, SS hypertensive patients exhibit significantly increased plasma endothelin levels when in the salt-depleted state, probably via stimulation by catecholamines. This predicts an enhanced antihypertensive action for ETA or ETA/ETB blockers in this group of patients, a suggestion to be confirmed with the appropriate therapeutic trials.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received June 12, 2000; revision received August 21, 2000; accepted August 21, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. J Epstein Efficacy and Safety of Darusentan: A Novel Endothelin Receptor Antagonist Ann. Pharmacother., July 1, 2008; 42(7): 1060 - 1069. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Laffer, R. J. Bolterman, J. C. Romero, and F. Elijovich Effect of Salt on Isoprostanes in Salt-Sensitive Essential Hypertension Hypertension, March 1, 2006; 47(3): 434 - 440. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Gates, H. Tanaka, W. R. Hiatt, and D. R. Seals Dietary Sodium Restriction Rapidly Improves Large Elastic Artery Compliance in Older Adults With Systolic Hypertension Hypertension, July 1, 2004; 44(1): 35 - 41. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-S. Zheng, X.-Q. Yang, K. J. Lookingland, G. D. Fink, C. Hesslinger, G. Kapatos, I. Kovesdi, and A. F. Chen Gene Transfer of Human Guanosine 5'-Triphosphate Cyclohydrolase I Restores Vascular Tetrahydrobiopterin Level and Endothelial Function in Low Renin Hypertension Circulation, September 9, 2003; 108(10): 1238 - 1245. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Laffer, M. Laniado-Schwartzman, M.-H. Wang, A. Nasjletti, and F. Elijovich 20-HETE and Furosemide-Induced Natriuresis in Salt-Sensitive Essential Hypertension Hypertension, March 1, 2003; 41(3): 703 - 708. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Li, G. D. Fink, S. W. Watts, C. A. Northcott, J. J. Galligan, P. J. Pagano, and A. F. Chen Endothelin-1 Increases Vascular Superoxide via EndothelinA-NADPH Oxidase Pathway in Low-Renin Hypertension Circulation, February 25, 2003; 107(7): 1053 - 1058. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Rothermund, R. Vetter, M. Dieterich, P. Kossmehl, O. Gogebakan, C. Yagil, Y. Yagil, and R. Kreutz Endothelin-A Receptor Blockade Prevents Left Ventricular Hypertrophy and Dysfunction in Salt-Sensitive Experimental Hypertension Circulation, October 29, 2002; 106(18): 2305 - 2308. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |