Author + information
- Received October 21, 1999
- Revision received April 11, 2000
- Accepted June 15, 2000
- Published online November 1, 2000.
- Craig Cheetham, BSca,
- Julie Collis, BSca,
- Gerard O’Driscoll, MB, BCh, BAO, FRACPb,§,
- Kim Stanton, MB, BS, FRACP∥,
- Roger Taylor, MB, BS, FRACPb,e,* ( and )
- Daniel Green, PhDa,b,§
- ↵*Reprint requests and correspondence: Professor R.R. Taylor, Department of Cardiology, Royal Perth Hospital, Box X2213, GPO, Perth 6847, Western Australia
The present study examined the effect on forearm endothelial function of an angiotensin II type 1 receptor antagonist, losartan, in subjects with non-insulin-dependent diabetes mellitus (NIDDM).
Angiotensin-converting enzyme (ACE) inhibition with enalapril improves acetylcholine (ACh)-dependent endothelial function in patients with NIDDM. This could be mediated through angiotensin II and the type 1 receptor or could be due to inhibition of kininase II and a bradykinin preserving effect. It is therefore relevant to determine whether a type 1 receptor antagonist improves endothelial function.
The influence of losartan (50 mg daily for four weeks) on endothelium-dependent and independent vasodilator function was determined in 9 NIDDM subjects using a double-blinded placebo-controlled crossover protocol. Forearm blood flow was measured using strain-gauge plethysmography.
Losartan significantly decreased infused arm vascular resistance in response to three incremental doses of intrabrachial acetylcholine (p < 0.05, ANOVA). The forearm blood flow ratio (flow in infused to noninfused arm) was also increased (p < 0.01). Responses to sodium nitroprusside and monomethyl arginine were not significantly changed.
Losartan administration at 50 mg per day improved endothelium-dependent dilation of resistance vessels in patients with NIDDM. That is, blockade of the angiotensin II type 1 receptors improves endothelial function in NIDDM. At least some of the similarly beneficial effect of ACE inhibition is probably mediated also through the angiotensin II-type 1 receptor pathway. The use of a type 1 receptor antagonist seems a reasonable alternative to an ACE inhibitor to maintain endothelial function in NIDDM subjects.
Angiotensin-converting enzyme (ACE) inhibitors improve endothelial NO-related vasodilator function in patients with depressed function, including those with both insulin-dependent (1) and non-insulin-dependent (2) diabetes mellitus (NIDDM). The mechanisms are not well defined and may involve greater NO production and/or reduced NO inactivation by oxygen-derived free radicals (3) or by other compounds. Since the introduction of angiotensin II type 1 receptor blocking drugs there has been considerable speculation on the relative merits of these drugs and ACE inhibitors, especially in hypertension (4–6) and cardiac failure (7). The receptor blocking drugs can potentially produce greater inhibition of effects mediated by angiotensin II. During the administration of a competitive ACE inhibitor, angiotensin II can still be produced from angiotensin I, probably attributable both to the higher concentrations of angiotensin I that accompany ACE inhibition and to conversion by other enzyme systems such as chymase. In contrast, type 1 receptor inhibition with currently used drugs is partly noncompetitive, or insurmountable, although the degree varies with the particular inhibitor. Furthermore, it is speculated that type 1 receptor inhibition might result in buildup of angiotensin II with stimulation of type 2 receptors and beneficial effects. These considerations would favor the type 1 receptor blocking drugs over ACE inhibitors. On the other hand, ACE inhibitors additionally inhibit the breakdown of bradykinin, and there is considerable evidence that increase in release of NO, induced by the accumulation of bradykinin, contributes to the augmented endothelium dependent vasodilator responses resulting from ACE inhibition (8–12).
As a clear improvement in endothelium-dependent vasodilation has been found to result from four weeks administration of the ACE inhibitor, enalapril, in NIDDM (2), it was of interest to examine the effect of the type 1 receptor antagonist, losartan in a similar group.
Nine subjects (7 men, 2 women, average age 54 ± 2 years) with NIDDM without evidence of microvascular or macrovascular complications were recruited. They undertook a screening program consisting of a medical history and examination, and a hematological and biochemical profile, including measurement of blood glucose, glycated hemoglobin, serum electrolytes, urea and creatinine, uric acid, liver function and serum lipids. The following were excluded: smokers, those with renal impairment or proteinuria, hepatic impairment, gout or hyperuricemia, more than mild hypercholesterolemia (total cholesterol >6.0 mmol/l) or hypertension (systolic BP >160 mm Hg). The average time since diagnosis of diabetes was 5 ± 1 years. None of the patients were receiving lipid-lowering therapy or taking vitamin supplements or ACE inhibitors. Eight patients were taking metformin and, additionally, two were taking glibenclamide and one metformin plus glipizide. Medications remained unchanged during the study. None had significant microalbuminuria on quantitative assessment (24-h excretion using nephelometric method) or significant retinopathy (full-field photography). The mean glycated hemoglobin at entry was 7.6 ± 0.5% (normal range = 4.3% to 6.0%), indicating moderate to good glycemic control. The study protocol was approved by the Royal Perth Hospital Ethics Committee and subjects gave written, informed consent.
The effect of four weeks of angiotensin II type 1 receptor blockade was studied using a randomized, double-blind, placebo-controlled crossover protocol. Subjects were randomized to receive losartan 50 mg daily (Merck, Sharp & Dohme, Australia) or a similarly packaged placebo. The validity of the active drug/placebo randomization was checked by tablet analysis using an HPLC qualitative method. Forearm vascular function was studied after four weeks, following which crossover of therapy occurred with restudy four weeks later. The procedures were conducted, on average, 3.0 ± 0.6 h after the study medication and, for individual subjects, at the same time of the day for the repeat study after crossover. Subjects were required to refrain from drinking alcohol or caffeine-containing beverages for 12 h before the procedure. At each visit the biochemical and hematological parameters were repeated. There were no adverse side effects.
Vascular function assessment
Investigations were conducted in a quiet, temperature-controlled laboratory with subjects lying supine and both forearms supported above heart level. A 20-gauge arterial cannula (Arrow, Reading, Pennsylvania) was introduced into the brachial artery of the nondominant arm under local anesthesia with <2 ml of 1% lidocaine (Astra Pharmaceuticals, Australia) to transduce pressure, for the infusion of drugs or physiological saline and for sampling of arterial blood. Forearm blood flow (FBF, ml/100 ml forearm/min) was measured simultaneously in both arms by gallium/indium strain gauge (SG24, Medasonics, Mountain View, California) plethysmography. Wrist cuffs, connected to a flow-regulated source of compressed air, and arm cuffs, connected to a rapid inflation device (E20, D.E. Hokanson, Bellevue, W.A., Australia), were placed on each limb. Output from the strain gauges passed through an amplifier (SPG 16, Medasonics) and was sampled by an on-line microcomputer at 75 Hz before being displayed on a monitor in real time. A software program coordinated the acquisition, storage and display of data as well as inflation and deflation of the arm cuffs, ensuring that blood flow measures were synchronized with cuff inflation during recording periods. Intra-arterial pressure was measured continuously (Transpac, Abbot Laboratories, Illinois) throughout the study. Drug infusions were administered using a constant rate infusion pump (IVAC 770, IVAC Corporation, California) in the protocol indicated below.
Baseline measurements started at least 25 min after cannulation of the brachial artery. Blood flow measurements were taken by inflating the wrist cuffs to 220 mm Hg to exclude the hands from the circulation, and by rapidly inflating the upper arm cuffs to 45 mm Hg for 10 out of every 15 s throughout the baseline and drug infusion periods. Output from the strain gauges was stored and the average of the last five flow measurements from each period was used for analysis. Between infusions, the cuffs were deflated, allowing at least 15 min for forearm blood flow to recover from the preceding infusion before further baseline measures were recorded.
All solutions were prepared aseptically from sterile stock solutions or ampoules immediately before infusion into the brachial artery. Acetylcholine (ACh; Miochol; Johnson & Johnson, Australia) was infused at 10, 20 and 40 μg/min each for 3 min, followed by sodium nitroprusside (SNP; David Bull Laboratories, Australia) at 2, 4 and 8 μg/min each for 3 min, and then NG-monomethyl-L-arginine (LNMMA; Clinalfa, Switzerland) at 2, 4 and 8 μmol/min each for 5 min.
Although the low doses of drugs infused in the study produce negligible systemic effects and showed no effect on blood pressure or heart rate, it is still desirable to account for any possible variation in overall hemodynamics as a cause of the flow changes seen in the infused forearm. Thus, FBF was measured simultaneously in both arms, although only one arm was infused, and the noninfused arm served as a control. As in earlier studies (1,2), forearm blood flow in the infused arm is described as a ratio to that in the non-infused arm. Changes in these ratios during ACh, SNP and LNMMA infusions are expressed as percentage changes from the baseline immediately preceding each drug administration. In addition, FBF is expressed in absolute units (ml/100 ml/min), and vascular resistance was calculated in the infused arm as the ratio of mean arterial pressure to forearm blood flow and expressed in the units mm Hg per ml/100 ml tissue/min. All blood flow measures were analyzed by two blinded investigators.
Results are expressed as means ± SE. The responses after losartan therapy were compared to placebo responses using two-way analysis of variance (ANOVA) with repeated measures performed on the three dose levels of ACh, SNP and LNMMA. Responses at each level of drug infusion were compared using Student t test (two-sided). A p value of <0.05 was considered statistically significant.
There were no differences in blood glucose, glycated hemoglobin or serum lipids between losartan and placebo treatment (Table 1). Although blood pressure was lower at the time of FBF measurement following losartan therapy, the difference was of borderline statistical significance (p = 0.05).
Absolute FBF data recorded in the infused and noninfused limbs at baseline and during the infusion of ACh, SNP and LNMMA at three dose levels during placebo and losartan administration, are presented in Table 2. There was a trend to a greater increase in absolute FBF in the infused limb in response to ACh after losartan than after placebo, but this did not reach significance (p = 0.08, ANOVA; p = 0.08, 0.18, 0.10, by t tests respectively at low, intermediate and high ACh infusion rates). Although the data suggest a trend toward a greater response to SNP, this was not significant (p = 0.2, ANOVA; p = 0.29, 0.09, 0.24 by t tests) and the response to LNMMA was not altered by losartan (p = 0.5, ANOVA; p = 0.19, 0.57, 0.97). When the infused limb data were expressed in terms of forearm vascular resistance (FVR, Table 3) , losartan was associated with a significant increase in the response to ACh (p = 0.04, ANOVA; p = 0.03, 0.17, 0.14) and the responses to SNP and LNMMA were not significantly altered (p = 0.2, ANOVA; p = 0.35, 0.16, 0.27 for SNP and p = 0.6, ANOVA; p = 0.11, 0.49, 0.76 for LNMMA).
As described in the Methods, it is optimal to analyze the data in terms of FBF ratios, that is, the ratio of flow in the infused arm to that in the noninfused arm, and to refer these to the similarly derived baseline ratios preceding each set of drug infusions. Figures 1 through 3⇓⇓⇓ present the percentage changes in these ratios from their baselines, in response to ACh, SNP and LNMMA. Losartan administration significantly augmented responses to ACh (p = 0.01, ANOVA; p < 0.02, <0.02, <0.05 at the 3 dose levels, respectively), while there were no differences in response to SNP (p = 0.6, ANOVA; p > 0.4 at each infusion level by t tests) or LNMMA (p = 0.5, ANOVA; p > 0.2 at each infusion level).
Losartan in a dose of 50 mg once daily has approximately the same effect on blood pressure in hypertensives as does enalapril 20 mg daily (4,5). The effect of losartan to improve Ach-stimulated vasodilator endothelial function, which is largely NO-mediated, was comparable to that found with enalapril in similar NIDDM subjects (2).
Several authors have found endothelium-dependent responses to be depressed in NIDDM (13–16), but we did not compare our subjects with a normal group.
Although the responses to SNP were not significantly altered by losartan, they could be considered as equivocal in terms of absolute FBF. Such would be consistent with reports that endothelium-independent vasodilator responses to SNP (14,15) and to glyceryl trinitrate (13) can also be depressed in NIDDM. However, our SNP results do not substantiate improvement in endothelium-independent dilation from losartan.
Pharmacokinetics of Losartan
The receptor inhibition of losartan itself is competitive but that of its dominantly active metabolite, which has a peak plasma concentration approximately 3 h to 4 h after losartan administration, is partly noncompetitive or insurmountable. Although the effect on blood pressure in hypertensives is waning by 24 h, substantial effect remains (5,6). Our studies of FBF would have been conducted at a time of near maximum drug effect, as indicated by the slightly lower blood pressure at the time of the study conducted on losartan. Evidence has been presented by others that the hypotensive effect is not responsible for the beneficial effect on endothelial function of ACE inhibition with captopril (17).
Previous studies of type 1 receptor inhibition and endothelial function
Although many clinical studies have found ACE inhibition to improve endothelial function in a variety of conditions (1,2,17–20), there have been few such studies of the effect of type 1 receptor inhibition. In a recent preliminary communication, improvement in flow-mediated dilation (FMD) of the brachial artery, a largely NO-dependent response, as is the dilator response to ACh, was found after 2 months of losartan 25 or 50 mg daily in patients with coronary artery disease (21).
Mechanism of action; inhibition of kininase II with ACE inhibition
The role of the inhibition of kininase II and the preservation of bradykinin in the beneficial actions of the ACE inhibitors is controversial, but there is substantial evidence that, in some circumstances, it is important (8–12). Exposure of in vitro preparations to ACE inhibitors greatly enhances exogenous bradykinin induced vascular relaxation and acute ACE inhibition increases the coronary vasodilation resulting from administration of bradykinin into the human coronary circulation, an effect mediated through NO (22). Furthermore, experimental administration of a bradykinin receptor inhibitor abolishes or greatly reduces the potentiation by ACE inhibitors of muscarinic induced endothelium dependent responses (8,10,12). It is interesting to note that the inhibition of endothelium dependent vasodilation by oxidized LDL is reversed by ACE inhibition, the effect of which in the rat aorta was reported to be abolished by bradykinin receptor inhibition with HOE 140 (icatibrant), whereas losartan did not protect against the effect of oxidized LDL (12). Finally, in favor of an important action of ACE inhibition through bradykinin and NO production, and most persuasively in man, the bradykinin receptor inhibitor, icatibrant, not only substantially reduced the hypotensive effect of captopril (23), but completely abolished the increase in FMD of the radial artery induced by acute ACE inhibition with quinaprilat (11).
Mechanism of action; inhibition of superoxide production with type 1 receptor and ACE inhibition
Despite the above, a recent study suggests that ACE inhibition with quinapril does not increase NO production but rather decreases it; it is proposed that ACE inhibition reduces superoxide production, leading to reduced inactivation of NO and a compensatory decrease in NO production (24). Enhanced inactivation of NO by superoxide is now considered to be a major mechanism for depression of NO-related endothelial function (25). Superoxide production by membrane NADPH/NADP oxidase is stimulated by angiotensin II, believed to be via the type 1 receptor. The increase in superoxide production and depression of ACh-induced vascular relaxation by angiotensin II has been found to be inhibited by losartan (26), and angiotensin II macrophage-mediated oxidation of LDL was inhibited by the receptor antagonist saralasin (27). This evidence implicates a type 1 receptor and superoxide involvement in an angiotensin II-induced deleterious effect on endothelial function and provides an explanation for the beneficial effect of losartan. At the same time, it does not deny additional involvement of the bradykinin pathway in the action of ACE inhibition.
Other mechanisms relating endothelial function and NIDDM
In NIDDM, additional mechanisms may be operative to lead to depression of NO-related endothelial function, including the inactivation of NO by advanced glycation products as others have discussed (14). Also, apparently, insulin can stimulate the production of NO through the insulin receptor, the effector pathway having some commonality with that for glucose transport (28). Insulin resistance in NIDDM could lead to impaired NO production, as could defective metabolism of tetrahydrobiopterin. This compound is a cofactor for NO-synthase, its concentration is reduced in the diabetic rat (29), its metabolism is dependent upon the oxidant state and it can improve endothelial function in experimental diabetes (30).
Conclusions and implications
Even though the relative importance of the various possible mechanisms leading to depressed endothelial function in clinical NIDDM remains to be elucidated, our study shows that blockade of the angiotensin II type 1 receptor results in demonstrable improvement. This would not necessarily pertain to other conditions associated with depressed endothelial function in which the mechanisms might be different. However, in NIDDM, therapy with a type 1 receptor blocking drug would appear to be a reasonable alternative to an ACE inhibitor to maintain endothelial function. Such an approach could be particularly indicated in the presence of side effects such as cough seen more frequently with ACE than with type 1 receptor inhibition. Theoretical considerations suggest that a combined approach, as suggested in the management of other conditions, should be evaluated.
Losartan and placebo were supplied by Merck, Sharpe and Dohme (Australia). We thank Mr. Peter Hackett, LRSC, AAIMLS, Pharmacology and Toxicology, Path Center, for technical assistance.
☆ This study was supported by the Arnold Yeldham and Mary Raine Medical Research Foundation.
- angiotensin converting enzyme
- forearm blood flow
- flow mediated dilation
- forearm vascular resistance
- low density lipoprotein
- noninsulin-dependent diabetes mellitus
- nitric oxide
- sodium nitroprusside
- Received October 21, 1999.
- Revision received April 11, 2000.
- Accepted June 15, 2000.
- American College of Cardiology
- O’Driscoll J.G,
- Green D.J,
- Maiorana A,
- Stanton K,
- Colreavy F,
- Taylor R.R
- Rajagopalan S,
- Harrison D.G
- Johnston C.I
- Zazinger J,
- Zheng X,
- Bassenge E
- Aush-Schwelk W,
- Duske E,
- Claus M,
- et al.
- Hornig B,
- Kohler C,
- Drexler H
- Williams S.B,
- Cusco J.A,
- Roddy M.A,
- Johnstone M.T,
- Creager M.A
- Mancini G.B,
- Henry G.C.H,
- Macaya C,
- et al.
- Nakamura M,
- Funakoshi T,
- Arakawa N,
- Yoshida H,
- Makita S,
- Hiramori K
- Prasad A,
- Narayanan S,
- Mincemoyer R,
- Tupas-Habib T,
- Ellahham S,
- Quyyumi A.A
- Kuga T,
- Mohri M,
- Egashira K,
- et al.