(Hypertension. 1995;25:1161-1166.)
© 1995 American Heart Association, Inc.
Articles |
From the Department of Obstetrics and Gynecology, University Hospital, Malmö, Sweden.
| Abstract |
|---|
|
|
|---|
Key Words: pregnancy preeclampsia posture atrial natriuretic peptide renin aldosterone
| Introduction |
|---|
|
|
|---|
Because of a modified action of the RAAS, weaker vasopressor stimuli may lead to stronger responses in preeclamptic compared with normotensive pregnancy. Part of this response may be a modified action of atrial natriuretic peptide (ANP), which is an RAAS antagonist. ANP acts as a blood volume regulatory hormone, involved in the control of water and sodium excretion, blood pressure (BP), vascular resistance, hemoconcentration, and vasopressin secretion.6 7 8 9 ANP has vasorelaxant properties on preconstricted vascular smooth muscle in vitro, for example, after preconstriction by Ang II.10 11 Studies on plasma ANP in normotensive pregnancy have shown a significant increase in midpregnancy12 and late pregnancy,13 a nonsignificant increase,14 or an unchanged concentration.15 There is general agreement that ANP levels are elevated in preeclamptic compared with normotensive pregnancies,13 indicating an altered role of ANP in preeclampsia.
ANP is synthesized in cardiac myocytes, predominantly in the right atrium. Stretching of the myocytes, for example by vascular volume expansion, increased BP, increased sodium intake, or heart arrhythmia, stimulates cytoplasmatic granula to release ANP to the blood.16 In healthy normotensive individuals, physiological concentrations of ANP induce diuresis,17 but higher concentrations are necessary to decrease BP.18 However, after activation of the RAAS, ANP acts as a potent natriuretic and hypotensive agent.19 20
The hemodynamic effects of head-down tilting, as recommended by Trendelenburg in 1870 to facilitate suprapubic cystostomy,21 have been investigated in nonpregnant healthy women as well as in hypertensive individuals. Head-down tilting for 30 minutes induces an increase in cardiopulmonary blood volume that is equal in nonpregnant hypertensive and healthy individuals.22
To our knowledge, the effects of head-down tiltingan increase in central venous blood volume followed by an increase in cardiac filling and ANP releasehave not previously been studied in pregnancy. Head-up tilting results in a significant decrease in plasma ANP in hypertensive and normotensive pregnant women23 and nonpregnant women.24 Most likely, this is due to a reduction in central venous blood volume and right atrial pressure. Short-term intravenous volume expansion in normotensive human pregnancy causes an enhanced ANP response compared with the nonpregnant state,25 and in preeclamptic pregnancy the response is exaggerated compared with normotensive pregnancy.26 Changes in ANP concentration appear to be related to changes in maternal central hemodynamics,27 although during basal conditions there seem to be no correlations between ANP and atrial size.12
The related effects of BP, the RAAS, and ANP after postural stimuli remain unclear in normotensive and preeclamptic pregnancy. Our hypothesis was that head-down tilting to 10° (H-D10°) for 30 minutes would increase the blood volume of the right atrium and cause an exaggerated release of ANP in preeclamptic compared with normotensive pregnancy. We also wanted to reveal whether there is a different interaction between ANP and the RAAS at rest compared with during postural provocation.
| Methods |
|---|
|
|
|---|
After an overnight fast, usual hospital breakfasts and lunches were served; then, no more food or drinking was allowed during the investigation. The subjects remained in bed during the investigation, and blood samples were drawn between noon and 2 PM.
After women had been 30 minutes in the left lateral recumbent (LLR) position, blood samples were drawn for analyses of ANP, PRA, aldosterone, electrolytes, hemoglobin, and hematocrit. A second series of blood samples was collected after a further 30 minutes of H-D10° with the head resting on a small pillow. This position was in no case accompanied by discomfort or signs of cranial congestion. Heart rate and BP were monitored with an automatic sphygmomanometer (Auto BP Monitor, EME Ltd) on the left upper arm, with the cuff positioned at the level of the heart. Mean arterial pressure (MAP) was calculated as Diastolic BP+1/3(Systolic BPDiastolic BP). All women involved in the study were informed in detail about the investigation and its purpose and gave their consent to participate. The study was approved by the Ethics Committee of the Faculty of Medicine, University of Lund.
Chemical Analyses
For determination of PRA and aldosterone concentrations, venous
whole blood was collected in EDTA-prepared glass tubes. For ANP
determination, the tubes also contained the enzyme inhibitor Trasylol.
The tubes were immediately put on ice and centrifuged at room
temperature, and the plasma was stored at -50°C until analyzed.
Radioimmunoassay
Radioimmunoassay of ANP was performed at the Department of
Clinical Neuroscience, Section of Psychiatry and Neurochemistry,
Mölndal Hospital, University of Gothenburg, Sweden. The samples
were analyzed in serial dilutions optimized to the linear part of the
standard curve and corrected for nonspecific binding. Immunoreactive
-human ANP was measured with the use of a rabbit antiserum (A93,
Milab) in a final dilution of 1:24 000. The antiserum showed a 100%
cross-reactivity with
-human ANP-(7-28) and rat
-ANP and less
than 0.1% with
-human ANP-(1-11) and
-human ANP-(18-28). The
detection limit was 10 pmol/L, and the interassay coefficient of
variation was less than 10%.28 PRA and aldosterone
concentrations were routinely analyzed by radioimmunoassay at the
Endocrine Laboratory at Malmö General Hospital.
Statistical Analyses
Values are expressed as mean±SEM. Wilcoxon's signed rank test
was used for matched paired samples, and Scheffé's multiple
range test was used for comparison between multiple groups. Simple
linear regression was used for correlation between two variables. A
value of P<.05 was considered statistically significant.
All values are expressed in SI units.
| Results |
|---|
|
|
|---|
|
MAP, Heart Rate, Hemoglobin, and Hematocrit
There were no significant changes in MAP, systolic BP, or
diastolic BP in the pregnant groups during rest in the LLR position or
during H-D10° tilting (Fig 1). In nonpregnant women,
MAP first decreased and then increased (P=.05) during the
last 10 minutes in the LLR position; after the change to H-D10°, MAP
rose significantly during the last 20 minutes.
|
During the first 10 minutes of H-D10°, heart rate in the preeclamptic group fell significantly to 78±3 beats per minute and then further to 73±3 beats per minute during the last 20 minutes (Fig 2). No significant changes were found in the control groups.
|
Both hemoglobin and hematocrit were lower (P=NS) in pregnancy, with the lowest values observed in the healthy pregnant group (Table).
Atrial Natriuretic Peptide
Resting (basal) plasma ANP concentration was significantly
higher in the healthy pregnant (54±6 pmol/L) and preeclamptic (66±4
pmol/L) groups compared with the nonpregnant group (40±2 pmol/L) (Fig 3, left). The difference between the normotensive and
hypertensive pregnant women was not significant. After 30 minutes of
H-D10°, ANP increased significantly (58±6 pmol/L) in the
normotensive pregnant group, but no changes were found in the other two
groups (Fig 3, left). When all studied women were pooled (n=27), a
correlation (r=.57) was found between baseline MAP and ANP
concentration after 30 minutes of the LLR position (Fig 4).
|
|
Plasma Renin Activity
Resting PRA was significantly higher in the normotensive pregnant
group (6.11±0.81 µg/L per hour) compared with nonpregnant
(1.43±0.22 µg/L per hour) and preeclamptic (2.92±0.61 µg/L per
hour) women (Fig 3, middle). After H-D10°, PRA decreased
significantly in the nonpregnant women to 1.10±0.18 µg/L per
hour.
Aldosterone
Aldosterone concentration at rest was significantly higher in the
normotensive pregnant group (1.27±0.24 nmol/L) compared with both the
nonpregnant (0.31± 0.03 nmol/L) and preeclamptic (0.52±0.09
nmol/L) groups (Fig 3, right). After H-D10°, aldosterone
concentrations decreased in all three groups, although significantly
only in the nonpregnant and preeclamptic groups.
PRA-ANP and Aldosterone-ANP Ratios
PRA-ANP and aldosterone-ANP ratios are shown in the Table. Both
ratios were significantly higher in normotensive pregnant women
compared with nonpregnant and preeclamptic women. In preeclamptic
women, the PRA-ANP ratio but not the aldosterone-ANP ratio was
significantly higher than in the nonpregnant women. During H-D10°,
the PRA-ANP ratio decreased significantly in nonpregnant women (26%)
but not in pregnant women (0.5%) or preeclamptic women (7%). The
aldosterone-ANP ratio declined significantly after H-D10° in the
preeclamptic group (24%) but not in the normotensive pregnant group
(19%).
| Discussion |
|---|
|
|
|---|
To explain the activation of the fourfold to fivefold increase in PRA and aldosterone in normotensive pregnancy, the underfill hypothesis has been proposed.29 In the pregnant baboon, the RAAS is activated weeks before any increase in plasma volume is observed.30 Vascular resistance also decreases before plasma volume expands. A hitherto unidentified vasodilator mechanism is probably activated very early in pregnancy, and RAAS activation might be interpreted as a normal physiological response to counteract hypotension and maintain BP. ANP might represent one of the mechanisms responsible for vasodilation. The relationship between vascular reactivity and ANP suggests that ANP plays a role in the development of low peripheral vascular resistance in pregnancy.31 In addition, the pregnant woman is protected against the potential hypertensive effects of Ang II by vascular refractoriness from as early as 11 weeks of gestation, at the same time as adrenal sensitivity is maintained.32
In preeclamptic women, who often exhibit a low plasma volume, one would expect low ANP levels. Paradoxically, this and other studies13 23 have demonstrated high ANP levels, indicating that the normal balance between ANP and plasma volume is lost in preeclamptic pregnancy. Instead of causing short-term changes of plasma volume, the primary aim of an ANP response might be to counteract the hypertension and vasospasm. This theory is supported by the demonstrated correlation between ANP and MAP in this and other studies.33 The reduced plasma volume may then be interpreted as a secondary side effect caused by increased ANP release. In the present study there was no evidence for any short-term changes in plasma volume, since hemoglobin and hematocrit were not significantly increased (Table). Hence, we were not able to confirm previous reports stating that ANP increases capillary permeability.34
The responses of aldosterone (decrease) and ANP (no change) after H-D10° were similar in the nonpregnant and preeclamptic groups, and the responses of PRA (decrease, P=NS) were similar in the two pregnant groups. Hence, in normotensive pregnancy, ANP might be controlled by mechanisms other than in nonpregnant individuals and in pregnancies complicated by preeclampsia. Another possibility is that the normotensive pregnant woman might be less sensitive to ANP. This might then be a phenomenon similar to the reduced sensitivity mentioned above to Ang II in normotensive pregnancy.5 A lower sensitivity for ANP in normotensive pregnancy might also explain why high levels of ANP are present at the same time as renin and aldosterone increase fivefold to sixfold.
The lack of ANP response after H-D10° tilting in the preeclamptic group is in contrast with previous studies showing an exaggerated ANP release after intravenous volume expansion.26 As in the normotensive pregnant group, an increase in ANP was expected. However, if changes of the RAAS are also taken into account, there is a possible explanation. The changes of the PRA-ANP ratio were not different after tilting in the two pregnant groups. The PRA-ANP ratio decreased by 0.5% (P=NS) in the pregnant group and by 7% (P=NS) in the preeclamptic group, compared with a 25% decrease in the nonpregnant group. Therefore, instead of an increased ANP release, preeclamptic women responded to head-down tilt by a downregulation of PRA, maintaining approximately the same balance between PRA and ANP as healthy pregnant women. However, the balance between aldosterone and ANP in preeclampsia was not maintained compared with normotensive pregnancy. The aldosterone-ANP ratio decreased by 23% (P<.01) in the nonpregnant group, by 19% (P=NS) in the normotensive pregnant group, and by 24% (P<.05) in the preeclamptic group. A lack of correlation between aldosterone and ANP has previously been demonstrated in normotensive pregnancy.35
These results suggested a defective link between PRA and aldosterone in preeclampsia. One reason might be a depletion of angiotensin-converting enzyme stores. We did not measure Ang II or angiotensin-converting enzyme in the present study, but it is known that angiotensin-converting enzyme activity is depressed in pregnancy compared with the nonpregnant state.36 It is probable that activation of the RAAS with excessive conversion of Ang II depletes angiotensin-converting enzyme stores. Angiotensin-converting enzyme might therefore be a rate-limiting factor of RAAS activity in pregnancy.
Although the balance between PRA and ANP was maintained during tilting in the preeclamptic group, the reason for the lack of ANP response is obscure. In accordance with our hypothesis, we had expected a significant increase of ANP concentration in preeclampsia with head-down tilting. One explanation might be that the 10° tilting was too small to imitate a volume load. Another possibility is that the release of ANP from cardiocyte granula was already close to an absolute maximum and the myocytes could not respond to stretching by any accelerated release of ANP. This exhaustion theory remains to be proved.
We have not been able to find any other studies comparing RAAS activation and ANP release during head-down tilting in pregnancy. Comparisons must therefore be restricted to nonpregnant individuals regarding head-down tilting and pregnant individuals regarding head-up tilting. Head-down tilting in normotensive and hypertensive men causes a significant decrease of PRA in both groups although no changes in aldosterone in hypertensive individuals.37 In the present study, nonpregnant women showed the same reactions as previously demonstrated in healthy men, whereas pregnant women maintained their PRA concentration. In contrast to hypertensive men, preeclamptic women showed a significant decrease of aldosterone with head-down tilting.
Changes of RAAS activity would be expected to cause changes of BP if not counteracted. In the two pregnant groups BP remained unchanged during the 30 minutes of H-D10°. A significant change in BP was found only in the nonpregnant group. This was observed in connection with a downregulation of the RAAS and unchanged ANP. Most likely, the stimulus from head-down tilting was not strong enough to imitate a volume load and increase BP in the pregnant groups. Hence, in normotensive pregnancy, ANP release seems to depend on blood volume changes. In preeclampsia, ANP release depends instead on an increase in BP.
In the preeclamptic group, heart rate fell significantly after 10 minutes and again after 30 minutes of H-D10°. This indicates that preeclamptic women primarily reduce cardiac output by decreasing heart rate and that BP is rather resistant to changes in posture. Hence, ANP release in preeclampsia did not change with changes in cardiac output.
In summary, H-D10° for 30 minutes resulted in significantly elevated plasma ANP in normotensive pregnant women that was not accompanied by any changes in PRA, aldosterone, MAP, or heart rate. In contrast, preeclamptic women partly seemed to overcome the increased central blood volume load induced by the head-down tilt position by decreasing heart rate and aldosterone, whereas BP, ANP, and PRA were not significantly changed. In the preeclamptic group as well as in the nonpregnant control group, the response to increased cardiac preload seemed to be downregulation of the RAAS. A significant negative correlation between PRA and ANP responses was found in the normotensive and preeclamptic pregnant groups, but between aldosterone and ANP responses, only in normotensive pregnancy. In normotensive pregnancy, ANP release seemed to depend on blood volume. In preeclampsia the ANP release followed BP and not changes in cardiac output. These results indicate that in women whose pregnancies are complicated by preeclampsia, normal hemodynamic control is lost, with reactions to head-down tilting similar to those in nonpregnant individuals.
| Footnotes |
|---|
Received June 23, 1994; first decision July 27, 1994; accepted January 24, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Bahar, M. Chanimov, M. L. Cohen, M. Friedland, Y. Grinshpon, R. Brenner, I. Shul, R. Datsky, and D. J. Sherman Lateral recumbent head-down posture for epidural catheter insertion reduces intravascular injection Can J Anesth, January 1, 2001; 48(1): 48 - 53. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |