(Hypertension. 1997;30:1247-1252.)
© 1997 American Heart Association, Inc.
Articles |
From the Department of Obstetrics and Gynecology, D Floor, East Block, Queen's Medical Centre, University Hospital, Nottingham, NG7 2 UH England.
Correspondence to F. Broughton Pipkin, Professor of Perinatal Physiology, Department of Obstetrics and Gynaecology, D Floor, East Block, Queen's Medical Centre, University Hospital, Nottingham, NG7 2UH, England. E-mail Fiona.Broughton-Pipkin{at}nottingham.ac.uk
| Abstract |
|---|
|
|
|---|
20 mm Hg was
observed. The infusion was stopped, and blood pressure was monitored at
2-minute intervals for 30 minutes. There was a significant
diastolic depressor effect after stopping
angiotensin II in the nonpregnant women and those in the
second and third trimesters of pregnancy. Individual women required
differing doses of angiotensin II to evoke the standardized
pressor response. It was thus possible to examine the depressor
response in each group in relation to infused doses of
angiotensin II. In nonpregnant women and in those in the
second and third trimesters of pregnancy, the depressor response was
dose dependent (P<.001). At any given dose, the depressor
response deepened as pregnancy progressed (P<.001). Basal
plasma prostacyclin concentrations rise in pregnancy, and
angiotensin II can stimulate prostacyclin synthesis. This
might mediate the depressor effect. In conclusion, the diminished
pressor response to angiotensin II in normal pregnancy may
be partly due to an increasing depressor effect of the hormone.
Key Words: angiotensin II renin-angiotensin system blood pressure pregnancy, humans
| Introduction |
|---|
|
|
|---|
It has been known for nearly 30 years that pressor responsiveness to Ang II is reduced in normotensive pregnancy, whereas that to noradrenaline is not.11 This suggested the downregulation of Ang II receptors. However, the density of Ang II receptors in mesenteric and uterine arteries has been reported not to differ in nonpregnant and near-term ewes12 ; the Bmax in human uterine arteries is also reported as being unchanged during pregnancy.13 It is not ethically possible to obtain repeated vascular biopsies from pregnant women. The direct study of AT1 receptors in human systemic resistance vessels in pregnancy has thus not been performed. A possible explanation for the decrease in pressor response to Ang II in human pregnancy would be an increased synthesis and/or release of prostacyclin (PGI2) or another vasodilator in response to Ang II. Ang II can release arachidonic acid metabolites such as PGI2 from vascular smooth muscle cell cultures.14 The acute administration of PGI2 results in a significant fall in pressor response to Ang II in second trimester human pregnancy.15 In vitro Ang II evokes significant increases in PGI2 release from the uterine arteries of pregnant but not nonpregnant sheep.16 This effect is mediated through the AT1 receptor.17 No significant increase was observed in the omental vasculature.16
In the course of experiments to investigate possible modulators of the pressor effect of Ang II in human pregnancy,15 18 19 we had repeatedly observed, but not studied or reported on, a fall in both systolic and diastolic pressure to below basal levels on stopping the infusion of Ang II, which was most pronounced during the first 10 minutes. The objectives of this study were therefore to determine (1) whether a significant depressor effect attributable to Ang II was indeed consistently noted in unanesthetized humans; (2) whether any such effect was dose-dependent, indicating indirectly a receptor-mediated rather than a nonspecific effect; and (3) whether any such effect was influenced by pregnancy.
| Methods |
|---|
|
|
|---|
|
Infusion Protocol
The pressor sensitivity to Ang II varies widely in pregnancy. We
therefore aimed to achieve a standard pressor response to 20
mm Hg above basal levels using doubling doses of Ang II from 4
ng/kg · min-1. Nonpregnant women are more
sensitive than pregnant women to Ang II, usually exhibiting a 20-mm Hg
pressor response at 4 ng/kg · min-1. However,
to allow comparison of possible dose as well as baroreflex effects in
pregnant and nonpregnant women, doses of up to 16
ng/kg · min-1 Ang II were given to nonpregnant
women. Up to 64 ng/kg · min-1 were given to
pregnant women.
The protocol used was as described previously18 with only minor modifications. It is described in brief below.
The subjects lay supine throughout the experiment; a pillow was placed under the pregnant subjects to support a right lateral tilt. An indwelling cannula (Venflon 18G, Viggo AB) was placed in the right antecubital fossa through which saline solution was infused continuously at approximately 30 drops per minute. Arterial blood pressure was measured at 2-minute intervals throughout the experiment from the left arm using a Dinamap automatic blood pressure recorder (Critikon Ltd).
Arterial blood pressure was allowed to stabilize for a minimum of 30 minutes after setting up, until random fluctuations in diastolic blood pressure had been within 6 mm Hg for 10 minutes. Ang II (Hypertensin, Ciba) was then infused intravenously through a needle connection into the indwelling cannula of the right arm so that it was flushed in by the saline infusion. The Ang II was made up freshly each day from dry ampules of 2.5 mg; the stock solution was then stored on ice. This was infused with a Vickers pump (Vickers Medical Ltd). The Ang II infusion was discontinued after the last planned infusion step or earlier if a 20-mm Hg rise in diastolic blood pressure had been achieved. A final period of 30 minutes recovery and stabilization of blood pressure then followed.
Calculations
The mean basal systolic and diastolic blood
pressures were calculated over the 10 minutes immediately before the
Ang II infusion was begun. Pressor and depressor effects were
calculated by comparison with this for each patient.
Data are presented as arithmetic mean±SEM. Between- and within-group comparisons were made by ANOVA, with post hoc testing by Scheffé's method, using the Statistical Package for the Social Sciences, version 3X.
| Results |
|---|
|
|
|---|
|
Fig 1
summarizes the change in
systolic pressure during the 30 minutes after stopping Ang II
infusion in the four study groups. There was no evidence of any
depressor effect in any group, with statistically significant rises
above basal being maintained in most groups (P<.001 for all
except group 3). There was a significant effect (P<.0001;
ANOVA) of gestation group on the level of this rise; patients in the
third trimester showed the least rise. Fig 2
shows the change in
diastolic pressure over the same period. The nonpregnant
patients and those in the second and third trimesters showed
significant and maintained falls in diastolic pressure
throughout this period. Again, there was an effect of gestation age,
with the response being greatest in the third trimester (group 3;
P<.01).
|
|
The comparable maximum evoked pressor effect in the four groups was
achieved by significantly different doses of Ang II (Table 2
). We
therefore considered only data from subjects exposed to a maximum dose
of 16 ng Ang II/kg · min-1.
Table 3
shows the maximum evoked pressor
effect in these patients. There was again no evidence of a
systolic depressor effect, but the diastolic
pressure had fallen to significantly below basal (P<.05) by
6 to 8 minutes after cessation of infusion in all groups except group 1
and was significantly reduced overall in these groups
(P<.02; Fig 3
). There was no
evidence for an effect of gestation per se. However, when a comparable
analysis was performed on data from subjects receiving 32 ng
Ang II/kg · min-1, not only were significant depressor
effects again demonstrated in the second and third trimesters
(P<.001 for both), but there was also a significant effect
of gestation age (Fig 4
;
P<.001 overall).
|
|
|
Comparison of Figs 3
and 4
suggested a dose-dependent depressor effect
at each gestation stage. This was confirmed for the nonpregnant
subjects (Fig 5
; P<.001
comparing maximal doses of 8 and 16
ng/kg · min-1) and in the third trimester,
comparing maximal doses of 16, 32, and 64
ng/kg · min-1 (P<.001; Fig 6
). Interestingly, at this gestation, the
highest necessary dose of Ang II was also associated with a significant
systolic depressor effect (P<.001); there was a
significant difference between the effects of the three doses
(P<.001).
|
|
| Discussion |
|---|
|
|
|---|
A fall in blood pressure to levels below basal on cessation of the infusion of a potent pressor agent could be due either to a transient disequilibrium in baroreflex response or to a longer-acting, receptor-mediated depressor effect, initially masked by the pressor effect. Acutely raising plasma Ang II concentrations for a minimum of 30 minutes can evoke a degree of central resetting of the blood pressure. Rapid baroreceptor resetting can be observed in animals after as little as 30 minute infusion of a pressor agent.21 By standardizing pressor responses as far as we were able in human subjects, we attempted to minimize differences in baroreflex response. However, because Ang II exerts direct effects on the baroreceptors,22 an effect at this level cannot be ruled out.
The elegant studies of Scheuer and Perrone8 suggest
strongly that in the anesthetized rat, the depressor effect
after the bolus injection of Ang II is genuinely agonist mediated via
AT2 receptors. The data presented here support the
hypothesis of an agonist-mediated response. Both in nonpregnant (Fig 5
)
and pregnant (Fig 6
) subjects, the depressor effect was significantly
dose related. The duration of effect was also greater at higher doses.
Thus, in the third trimester, the depressor effect after 16
ng/kg · min-1 was ended by 12 minutes after
stopping the infusion (Fig 6
), whereas in the groups requiring 32 and
64 ng/kg · min-1 to evoke a 20-mm Hg rise in
pressure, the depressor effect was still apparent at the end of the
experiment.
In the original observations of the depressor effect of Ang II,7 it was shown that the simultaneous administration of cyclooxygenase inhibitors potentiated the pressor effects and abolished the depressor effects of Ang II. We have shown previously15 that infused PGI2 blunts the pressor response to Ang II in human pregnancy. However, Ang II has been reported to not significantly stimulate prostacyclin synthesis by ovine omental arteries in pregnancy.16 In any case, Ang II in the circulation of pregnant and nonpregnant women has a half-life of around 50 seconds,23 whereas the observed depressor effect was of much longer duration. It could therefore be supposed that a biologically active metabolite of Ang II had been produced by degradation, which has a long half-life in the circulation. Ang II is degraded in vivo by a variety of amino peptidases, and some C- and N-terminal fragments have biological activity. Amino peptidase concentrations are raised in human pregnancy,24 but the metabolic clearance rate of Ang II appears to be largely unchanged.23 25 The [desAsp1]Ang II fragment (angiotensin 2-8), Ang III, has only about one third the pressor activity of Ang II in humans.26 Assays of sufficient sensitivity to identify this fragment show that in the nonpregnant subjects, it is present at about one fifth of the concentration of Ang II.27 28 The only study of such sensitivity including pregnant women29 reported a similar proportion in five pregnant women. Both Ang II and Ang III have been reported to be depressors on direct central administration in low doses.30
The N-terminal fragment [desPhe8]Ang II (Ang-[1-7]) can
be synthesized in vivo and in vitro by the degradation of either Ang I
or Ang II by endopeptidases such as prolyl or neutral
endopeptidase 24.11
(enkephalinase).31 32 Ang-(1-7) stimulates
prostaglandin synthesis in cultured vascular
endothelial cells33 and human
astrocytes34 and stimulates renal prostacyclin release
from the isolated perfused rat kidney.35 It has also been
reported to evoke the release of nitric oxide from porcine
coronary artery rings36 and from feline mesenteric
vasculature.37 It is thus not surprising that it should
appear to act as a vasodilator and depressor agent both
peripherally37 38 and
centrally.21 22 There appears to be no published
information concerning Ang-(1-7) concentrations during pregnancy or
comparative rates of Ang II metabolism to its various
fragments. The rate of further breakdown of the large fragments is also
unknown at this time, although human placental subcellular fractions
have been reported to break down Ang-(2-8) rapidly.24
Ang-(1-7) may perhaps be a more likely candidate as a depressor
fragment than Ang-(2-8) in view of its known depressor
effect.8 Basal PGI2 synthesis is also enhanced
in normal pregnancy.39 Because Ang-(1-7) appears to exert
its depressor effect partly through stimulation of the vasodilator
prostanoids, this might contribute to the greater depressor effect seen
in pregnancy (Fig 2
). Interestingly, Ang-(1-7) has been reported to
directly antagonize the pressor effects of Ang II in spontaneously
hypertensive rats.40
Ang II normally exerts a negative feedback on renin release, and we have shown previously that this effect is maintained in human pregnancy.19 It is thus possible that the relatively high plasma concentrations of Ang II reached during infusion suppressed renin release to such an extent that preinfusion concentrations of endogenous Ang II had not been reattained by the end of the experiment. We have no data with which to test this hypothesis.
One other possibility can be suggested. Mesenteric vascular Ang II receptors are downregulated in rabbit pregnancy41 ; rabbit systemic vascular Ang II receptors are of the AT1 subtype.42 However, uterine and mesenteric Ang II vascular smooth muscle receptors from sheep are not downregulated in pregnancy.12 43 The Bmax in human uterine vessels has also been reported as being unchanged in pregnancy,13 although there appear to be no studies of the Ang II receptors in human mesenteric vessels in pregnancy. The observation that uterine vascular smooth muscle AT2 receptors are not downregulated in human pregnancy,13 without information on other systemic vessels, does not necessarily close the question because the responses of pregnant human uterine and other systemic vasculature differ markedly to various agonists.44
In conclusion, we have shown that on ending an Ang II infusion, there is a depressor effect in nonpregnant and pregnant women. This effect increases with gestation age and may partly explain the blunted pressor response to Ang II in normal pregnancy.
| Acknowledgments |
|---|
Received May 31, 1996; first decision July 11, 1996; accepted April 24, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. B. Ahmed, A. K. Kang, K. D. Burns, C. R. J. Kennedy, V. Lai, D. C. Cattran, J. W. Scholey, and J. A. Miller Effects of Oral Contraceptive Use on the Renal and Systemic Vascular Response to Angiotensin II Infusion J. Am. Soc. Nephrol., March 1, 2004; 15(3): 780 - 786. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. P. Rowe and B. Dixon Angiotensin III Depressor Action in the Conscious Rabbit Is Blocked by Losartan but not PD 123319 Hypertension, January 1, 2000; 35(1): 130 - 134. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |