| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 1996;27:1299-1304.)
© 1996 American Heart Association, Inc.
Articles |
From the Research Forum (I.S.) and Department of Internal Medicine (A.M., S.K., H.A.), Ullevål University Hospital, and Stovner Health Centre (E.S.), Oslo, Norway.
Correspondence to Ingebjørg Seljeflot, Medical Outpatient Clinic, Department of Medicine, Ullevål University Hospital, N-0407 Oslo, Norway.
| Abstract |
|---|
|
|
|---|
Key Words: angiotensin II fibrinolysis insulin catecholamines hyperinsulinism
| Introduction |
|---|
|
|
|---|
The mechanisms regulating the different components in the fibrinolytic system are still not clarified. Insulin has been suggested as one of the regulators of PAI-1 on the basis of cell culture experiments7 8 9 and results from clinical studies showing strong correlations between plasma levels of PAI-1 and insulin3 4 5 and insulin resistance.10 However, several studies examining the effect of an acute increase in blood insulin level with the use of a glucose clamp technique or OGTT have not confirmed insulin per se as a stimulator of PAI-1 release.11 12 13 But under these circumstances, several different physiological reactions are involved, including vasodilation and vasoconstriction. As stimulating effects of some vasoactive hormones on TPA release from the endothelium are well established,14 this could confer a link between insulin and fibrinolysis in acute situations.
An increase in norepinephrine has been demonstrated during both the glucose clamp procedure15 and oral glucose loading,16 17 and results from several studies clearly show that catecholamines (norepinephrine and epinephrine) are involved in the regulation of fibrinolysis, mostly by a stimulating effect on TPA secretion from the vasculature.18 19 20
Some in vitro data have suggested that Ang II, a potent vasoconstrictor, is involved in the mechanisms of fibrinolytic regulation by synthesis of both TPA and PAI-1,21 and clinical use of angiotensin-converting enzyme inhibitors has been associated with a reduction in plasma levels of PAI-1.22 Furthermore, Ang II may be a mediator of catecholamine secretion, especially norepinephrine, as the level of circulating norepinephrine has been reported to be decreased in patients treated with Ang II receptor blockers.23 The expected increase in norepinephrine during hyperinsulinemia thus might be suppressed during treatment with Ang II receptor antagonists.
We are pursuing the hypothesis that Ang II could be a mediator of TPA as well as of norepinephrine release. We have previously demonstrated decreased levels of PAI-1 and increased levels of TPA during an OGTT,13 and the same profiles have been shown during glucose clamp procedures,12 24 both of which lead to acute hyperinsulinemic states. Ang II receptor antagonism could affect these fibrinolytic reactions and thereby act prothrombotically, either directly by a reduced effect of Ang II or by a decrease in norepinephrine, both possible stimulators of TPA secretion.
The aim of the present study was to investigate the effects of selective Ang II receptor blockade on fibrinolysis and some relevant hormones in the basal situation and two different models of acute hyperinsulinemia: OGTT and euglycemic glucose clamp.
| Methods |
|---|
|
|
|---|
Subjects
Twenty subjects (7 women and 13 men; mean age, 48 years; range,
24 to 79) with essential hypertension (blood pressure >140/95 mm Hg
and self-assessed home diastolic pressure >90 mm Hg)
were recruited from general practice physicians. Seventeen subjects
were previously untreated for their hypertension, and 3 had been
without medication the previous 3 to 7 weeks. The study was
approved by the Regional Ethics Committee, and all participants gave
their written, informed consent.
Procedures
All subjects underwent the OGTT 3 to 4 days before the glucose
clamp. The same protocol was used after 4 weeks of treatment. All
procedures were started with subjects in a fasting state between 8 and
9 AM.
OGTT Procedure
The first blood samples were collected from an antecubital vein
before the subjects were given 75 g glucose monohydrate in 400 mL
of water orally. Venous blood samples then were drawn 1, 2, and 3 hours
after intake. The subjects were in a quiet state during the study.
Glucose Clamp Procedure
Before the glucose clamp procedure was started, venous blood was
arterialized as follows: On the right arm, an antecubital
vein was cannulated with a short
polytetrafluoroethylene catheter (Venflon
17G, Viggo AB) and the arm placed in a heating sleeve (Thermal Vascular
Dilatator, Swetron AB) with the temperature set to 52°C. This cannula
was used for sampling of arterialized venous blood during
the procedure. On the left arm, an antecubital vein was cannulated for
later infusion of insulin and glucose. The subjects rested in the
supine position for 20 minutes before baseline blood collection.
The euglycemic hyperinsulinemic glucose clamp procedure was performed with a modification of the method described by DeFronzo et al25 and previously detailed.26 Insulin was infused at a fixed rate of 1 mU/kg per minute. Blood samples were taken every 5 minutes for determination of blood glucose concentration with Reflolux II (Boehringer Mannheim GmbH). In addition, blood samples were collected at baseline and after 30, 60, 90, and 120 minutes.
The glucose disposal rate (milligrams per kilogram per minute) was calculated from the amount of glucose infused from 100 to 120 minutes, and the mean serum insulin concentration was determined from the two samples obtained during the last 20 minutes.
Blood Sampling Procedures and Laboratory Methods
Serum glucose was determined enzymatically with a glucose
dehydrogenase method (HoffmannLa Roche), and insulin was measured by
a radioimmunoassay with a specific antibody (Linco Research Inc) and an
intra-assay coefficient of variation less than 9% at all levels.
Citrated plasma (Becton Dickinson Vacutainer tubes containing 0.129 mmol/L trisodium citrate in a 1:10 dilution) was collected and separated within 30 minutes by centrifugation at 2500g for 20 minutes at 4°C for determinations of PAI-1 activity, PAI-1 antigen, and TPA antigen. Acidified plasma for measurement of TPA activity was obtained with Stabilyte tubes (Biopool AB) as described by Rånby et al.27 PAI-1 and TPA activities were measured amidolytically. Enzyme-linked immunosorbent assay methods with a double-antibody technique were used for determinations of PAI-1 (measuring free PAI-1 as well as in complex with TPA) and TPA antigen (measuring free TPA as well as in complex with PAI-1). Commercial kits (Biopool AB) were used (Spectrolyse/pL, Spectrolyse/Fibrin, TintElize PAI-1, and TintElize TPA, respectively). The interassay coefficients of variation were 4.5% for PAI-1 activity, 8.0% for TPA activity, 7.8% for PAI-1 antigen, and 6.5% for TPA antigen. Plasma epinephrine and norepinephrine were determined by the radioenzymatic technique of Peuler and Johnson as previously detailed.28 Interassay coefficients of variation were 10% for both determinations. For all hormones and fibrinolytic parameters, the same procedures for sample preparations were used with both venous blood during the OGTT and arterialized venous blood from the glucose clamp procedure.
Statistics
For group differences during the test period, the Mann-Whitney
rank sum test was used. For differences from baseline to various time
points, Student's t test for paired data was used. For
correlation analysis, the Pearson correlation coefficient was
estimated. A two-tailed value of P
.05 was considered
statistically significant. Data are presented as mean±SE.
| Results |
|---|
|
|
|---|
|
Insulin and Catecholamine Responses
All subjects were good responders to glucose intake in the OGTT.
The measured values of insulin 1 hour after intake did not differ
between the treatment regimens (910±465 pmol/L during losartan
and 833±436 pmol/L during placebo, P=.499). The mean levels
of insulin during the last 30 minutes of the clamp procedure were
1011±59 pmol/L during losartan and 927±52 pmol/L during
placebo (P=.09), whereas the glucose disposal rate was
6.7±0.6 mg/kg per minute during losartan and 6.2±0.5 mg/kg
per minute during placebo (P=.41).
Changes from baseline did not differ between the treatment regimens
with regard to catecholamines in either of the test models
(Fig 1
). Norepinephrine levels did not
increase significantly from baseline to various time points during
either of the procedures. Norepinephrine showed no changes
during the procedures except during the OGTT, when a significant
increase after 3 hours appeared with both treatment regimens (Fig 1
).
|
Fibrinolytic Response
Table 2
shows the results of the fibrinolytic
variables during the OGTT and glucose clamp. The treatment regimens
did not differ significantly with regard to changes from baseline to
different time points in any of the variables measured (Table 2
).
Although TPA activity was significantly higher during placebo compared
with losartan at 2 hours (P<.05) and a similar
difference of borderline significance occurred at 1 hour
(P=.072) during the OGTT, the changes from baseline were not
significantly different between the two regimens (P=.232 and
P=.288, respectively).
|
During the OGTT, PAI-1 activity decreased significantly
(P<.001), as did PAI-1 antigen (P<.001), and
TPA activity increased significantly (P<.01) after 1 hour
with both treatment regimens; these changes were sustained throughout
the test period. During the glucose clamp procedure, fibrinolytic
variables after 30 minutes did not change, whereas after 1 hour,
the same changes as during the OGTT occurred and were sustained
throughout the procedure. TPA antigen did not change significantly from
baseline during any of the procedures (Fig 2
).
|
The values of PAI-1 activity, PAI-1 antigen, and TPA activity were systematically lower in the arterialized blood samples during the glucose clamp compared with the venous blood samples during the OGTT.
The correlation coefficient between insulin and PAI-1 activity at baseline was .62 (P<.05).
| Discussion |
|---|
|
|
|---|
Blood pressure was satisfactorily reduced by losartan, without any adverse effects on lipids or other measured metabolic variables. Thus, Ang II receptor antagonism probably was effectively achieved.
We have confirmed previous observations of increased fibrinolytic activity during acute hyperinsulinemia.11 12 13 24 Although we did not have any control subjects receiving saline infusion in the present study, the previously published studies12 13 have demonstrated that the rapid effects under these conditions cannot be accounted for by diurnal changes.29
To our knowledge, no studies have been published that compare the two
models of acute hyperinsulinemia presently used
with regard to their effects on fibrinolysis and
catecholamines. As we have found that the same profiles in
fibrinolytic variables appear in the same subject in the two
models, the effects should be related to insulin itself or to other
variables similarly changed during the procedures. We have
investigated the effects of 4 weeks of blockade of the Ang II receptor
with losartan and thereby indirectly the effect of Ang II on
fibrinolysis on the basis of the hypothesis that Ang II
might be a potential stimulator of TPA21 and
catecholamines.23 If this hypothesis were
correct, the increase in TPA during the procedures should have been
suppressed during losartan treatment. We could not convincingly
demonstrate such an effect (Fig 2C
, OGTT). Although losartan
treatment resulted in significantly lower levels in TPA activity
compared with placebo during the OGTT, the differences in changes from
baseline between the treatment regimens were not statistically
significant.
No differences in PAI-1 levels were encountered either in basal conditions or during the hyperinsulinemic procedures between the treatment regimens. Thus, our findings do not support those of Ridker et al,30 who demonstrated a dose-dependent increase in PAI-1 during Ang II infusion. However, this increase was shown with a very different model.
In endothelial cell cultures exposed to Ang II, an increase in PAI-1 synthesis and release has been demonstrated, suggesting that Ang II promotes increased plasma levels of PAI-1 in humans.31 However, there are conflicting results,32 and stimulatory effects of Ang II on the synthesis of both TPA and PAI-1 in cultured aortic smooth muscle cells have also been reported.21 In the study of Vaughan et al,31 inhibition of AT1 and AT2 receptors did not prevent the expression of PAI-1, suggesting that other angiotensin receptor subtypes are involved in the regulation of fibrinolysis. Thus, the endothelial angiotensin receptor is probably not of the AT1 subtype. This could explain the lack of effect on fibrinolysis in our study by losartan, which is a selective AT1 receptor antagonist. Although this assumption is strengthened by the recent findings of Kerins et al,33 entirely different results have also been published.34 Thus, the mechanisms by which the renin-angiotensin system may interact with fibrinolysis are still debatable.
No differences were observed between the two treatment regimens in norepinephrine and epinephrine levels. These results are not necessarily contradictory to those of Moan et al,23 who found a significant reduction in norepinephrine levels after 4 weeks of treatment with losartan. In that study, the patients had severe hypertension, and they had been treated with other hypertensive drugs until 3 days before entering the study.
Increased norepinephrine levels in acute hyperinsulinemic states have been described in several studies,15 16 17 35 but the results are conflicting.18 24 36 37 In the present study, norepinephrine did not increase significantly during either of the procedures, and a similar effect occurred with both treatment regimens, suggesting no effect of Ang II on norepinephrine levels during hyperinsulinemic states. Recently, Jern et al20 demonstrated an acute increase of TPA during norepinephrine infusion in humans that could point to norepinephrine as a direct mediator of TPA release in the current situations. However, with regard to the present results, it seems unlikely that norepinephrine has a regulatory role in fibrinolysis in these hyperinsulinemic states. This is in accordance with the findings of Landin et al.24
Epinephrine, a well-known stimulator of fibrinolysis,18 19 did not change during either of the procedures until 2 hours. This is in accordance with several other studies15 17 24 37 and confirms the suggestion that the increased fibrinolytic activity during acute hyperinsulinemia is probably not due to epinephrine. The increased levels of epinephrine after 3 hours during the OGTT are not readily explainable, but a physiological epinephrine response to the decreased glucose level would be expected at this time.
The reason for the differences in the levels of the basal fibrinolytic variables between venous and arterialized blood is not clear. One explanation could be a longer resting period before baseline sampling in the clamp procedure. In addition, the clamp procedure was performed with subjects in the supine position, whereas the OGTT was undertaken with subjects sitting.
Taken together, these data show that Ang II receptor blockade with losartan for 4 weeks did not affect the circulating levels of fibrinolytic variables either in basal conditions or during acute hyperinsulinemic states. The hypothesis that losartan could act prothrombotically by giving a smaller increase in TPA than expected during acute hyperinsulinemia could not be confirmed. However, more direct effects on fibrinolysis or the involvement of receptor subtypes other than AT1 cannot be ruled out. The increased fibrinolysis during acute hyperinsulinemia could not be explained by the action of norepinephrine. It seems reasonable to speculate on the vasoactive effects of insulin itself in playing a regulatory role in these situations, since insulin has the potential of vasodilation as well as vasoconstriction.38 The two models of acute hyperinsulinemia showed the same profiles in the measured variables. Thus, both models seem suitable for examinations during acute hyperinsulinemic states.
| Selected Abbreviations and Acronyms |
|---|
|
|
| Acknowledgments |
|---|
Received October 18, 1995; first decision December 5, 1995; accepted February 15, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. Gojanovic, F. Feihl, L. Liaudet, and B. Waeber Review: Concomitant calcium entry blockade and inhibition of the renin-angiotensin system: a rational and effective means for treating hypertension Journal of Renin-Angiotensin-Aldosterone System, March 1, 2008; 9(1): 1 - 9. [Abstract] [PDF] |
||||
![]() |
A. W.J.H. Dielis, M. Smid, H. M.H. Spronk, K. Hamulyak, A. A. Kroon, H. ten Cate, and P. W. de Leeuw The Prothrombotic Paradox of Hypertension: Role of the Renin-Angiotensin and Kallikrein-Kinin Systems Hypertension, December 1, 2005; 46(6): 1236 - 1242. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Sawathiparnich, L. J. Murphey, S. Kumar, D. E. Vaughan, and N. J. Brown Effect of Combined AT1 Receptor and Aldosterone Receptor Antagonism on Plasminogen Activator Inhibitor-1 J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3867 - 3873. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. J. Brown, S. Kumar, C. A. Painter, and D. E. Vaughan ACE Inhibition Versus Angiotensin Type 1 Receptor Antagonism: Differential Effects on PAI-1 Over Time Hypertension, December 1, 2002; 40(6): 859 - 865. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. C Felmeden and G. Y. Lip The renin-angiotensin-aldosterone system and fibrinolysis Journal of Renin-Angiotensin-Aldosterone System, September 1, 2000; 1(3): 240 - 244. [PDF] |
||||
![]() |
E. Chabielska, T. Matys, I. Kucharewicz, D. Pawlak, R. Rolkowski, and W. Buczko The involvement of AT2-receptor in the antithrombotic effect of losartan in renal hypertensive rats Journal of Renin-Angiotensin-Aldosterone System, September 1, 2000; 1(3): 263 - 267. [Abstract] [PDF] |
||||
![]() |
N. J. Brown, M. Agirbasli, and D. E. Vaughan Comparative Effect of Angiotensin-Converting Enzyme Inhibition and Angiotensin II Type 1 Receptor Antagonism on Plasma Fibrinolytic Balance in Humans Hypertension, August 1, 1999; 34(2): 285 - 290. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Cesari and G. P. Rossi Plasminogen Activator Inhibitor Type 1 in Ischemic Cardiomyopathy Arterioscler. Thromb. Vasc. Biol., June 1, 1999; 19(6): 1378 - 1386. [Full Text] [PDF] |
||||
![]() |
Correction Hypertension, November 1, 1996; 28(5): 919 - 919. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |