Author + information
- Received February 8, 1999
- Revision received August 24, 1999
- Accepted October 18, 1999
- Published online February 1, 2000.
- Yukihito Higashi, MD, PhD∗,* (, )
- Shota Sasaki, MD∗,
- Keigo Nakagawa, MD∗,
- Tomohiro Ueda, MD∗,
- Atsunori Yoshimizu, MD∗,
- Satoshi Kurisu, MD∗,
- Hideo Matsuura, MD, PhD∗,
- Goro Kajiyama, MD, PhD∗ and
- Tetsuya Oshima, MD, PhD†
- ↵*Reprint requests and correspondence: Dr. Yukihito Higashi, Hiroshima University School of Medicine, First Department of Internal Medicine, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan
The purpose of this study was to compare the effect of different antihypertensive agents, calcium antagonists, angiotensin-converting enzyme (ACE) inhibitors, beta-blockers and diuretic agents on endothelial function.
Endothelial dysfunction is a component of essential hypertension, and various antihypertensive drugs may be able to restore normal function.
Forearm blood flow (FBF) was measured in 296 patients with essential hypertension, including 46 untreated subjects using strain-gauge plethysmography during reactive hyperemia and after sublingual administration of nitroglycerin (NTG). Forty-seven normotensive subjects were similarly evaluated as control subjects.
The FBF during reactive hyperemia in the 296 hypertensive patients was significantly less than that in age-matched normotensive subjects. The increase in FBF after administration of sublingual NTG was similar in both groups. Systolic and diastolic blood pressures and forearm vascular resistance were greater in the untreated group than in the four treated groups and did not differ with respect to the antihypertensive agent used. The maximal FBF response from reactive hyperemia was significantly greater in the ACE inhibitor–treated group than in the group treated with calcium antagonists, beta-blockers, diuretic agents, or nothing (40.5 ± 5.2 vs. 32.9 ± 5.8, 34.0 ± 5.6, 32.1 ± 5.9, and 31.9 ± 5.8 ml/min per 100 ml tissue, p < 0.05, respectively). Reactive hyperemia was similar in the calcium antagonist, beta-blocker, diuretic and untreated groups, and changes in FBF after sublingual NTG administration were similar in all groups. The infusion of NG-monomethyl-l-arginine, a nitric oxide (NO) synthase inhibitor, abolished the enhancement of reactive hyperemia in hypertensive patients treated with ACE inhibitors.
These findings suggest that ACE inhibitors augment reactive hyperemia, an index of endothelium-dependent vasorelaxation, in patients with essential hypertension. This augmentation may be due to increases in NO.
Essential hypertension is associated with alterations in endothelial function. Endothelium-dependent vasodilation in response to substances such as acetylcholine, bradykinin and substance P and reactive hyperemia was reduced in brachial (1–3), coronary (4)and renal arteries (5,6)in patients with essential hypertension. Impairment of endothelial function has been shown to play an important role in the development and maintenance of hypertension (7).
Several studies have demonstrated the restoration of endothelial function in essential hypertensive patients through the administration of antihypertensive agents (8,9), especially angiotensin-converting enzyme (ACE) inhibitors. Recently, we also have reported that a 12-week treatment with ACE inhibitor, but not calcium channel antagonist, improves endothelium-dependent renovascular relaxation in patients with essential hypertension (10). In contrast, Creager et al. (11)have shown that effective antihypertensive therapy, including ACE inhibitors, did not restore impaired endothelium-dependent vasodilation in the forearm circulation of hypertensive patients. Although the results of experimental studies have supported the beneficial effect of ACE inhibitors on endothelial function (12,13), data on the forearm circulation of humans are controversial. In addition, the effect of other classes of antihypertensive agents on endothelial function has not been fully evaluated in human hypertension. It may be clinically important to select an appropriate antihypertensive agent that is effective in improving endothelial dysfunction in patients with established essential hypertension.
Thus, we compared the effects of the most frequently prescribed classes of antihypertensive agents on endothelial function in patients with essential hypertension. We measured the response of forearm blood flow (FBF) to reactive hyperemia, an index of endothelium-dependent vasodilation, and to the sublingual administration of nitroglycerin (NTG), an index of endothelium-independent vasodilation.
The study group consisted of 296 Japanese patients with essential hypertension (162 men and 134 women; mean age 52 ± 9 years) and 47 normotensive subjects (27 men and 20 women; mean age 48 ± 12 years). Hypertensionwas defined as a systolic blood pressure >160 mm Hg and/or a diastolic blood pressure >95 mm Hg, determined in a sitting position on at least three different occasions. Patients receiving any antihypertensive medications were considered to be hypertensive. Patients with secondary forms of hypertension, as suggested by clinical and biochemical examinations, were excluded from the study. Patients with a history of cardiovascular or cerebrovascular disease, diabetes mellitus or hepatic or renal disease were excluded. Patients receiving monotherapy with either calcium antagonists, ACE inhibitors, beta-blockers or diuretic agents were randomly recruited. All patients receiving antihypertensive drugs were treated for >24 weeks. Normotensionwas defined as a systolic blood pressure <140 mm Hg and a diastolic blood pressure <80 mm Hg. The normotensive control subjects had no history of serious disease and took no medication for at least four weeks before the study. The study protocol was approved by the Ethics Committee of the First Department of Internal Medicine of Hiroshima University. Informed consent for participation was obtained from all subjects.
Measurement of FBF
The FBF was measured using a mercury-filled Silastic strain-gauge plethysmograph (EC-5R, D.E. Hokanson, Inc., Issaquah, Washington) as previously described (1,2). The strain-gauge was attached to the left upper arm supported above the right atrium and connected to a plethysmography device. A wrist cuff was inflated to a pressure of 50 mm Hg above the systolic blood pressure to exclude the hand circulation from the measurements 1 min before each measurement and throughout measurement of FBF. The upper arm–congesting cuff was inflated to 40 mm Hg for 7 s in each 15-s cycle to occlude venous outflow from the arm by using a rapid cuff inflator (EC-20, D.E. Hokanson, Inc.). The FBF output signal was transmitted to a recorder (U-228, Advance Co., Nagoya, Japan). The FBF was expressed as milliliters per minute per 100 ml of tissue of forearm volume. Then FBF was calculated by two independent observers who had no knowledge of the subjects’ profile, including the drug(s) used and results from the linear portions of the plethysmographic recordings. The intraobserver coefficient of variation was 3 ± 1.7%.
The study began at 8:30 amafter the subjects fasted for at least 12 h. The subjects were kept in the supine position in a quiet, dark, air-conditioned room (constant temperature 22°C to 25°C) throughout the study. After 30 min in the supine position, basal FBF was measured. Then, the effect of reactive hyperemia and sublingual NTG on FBF was measured. To induce reactive hyperemia, FBF was occluded by inflating the cuff on the left upper arm to a pressure of 280 mm Hg for 5 min. After release of the ischemic cuff occlusion, FBF was measured for 3 min. Nitroglycerin, 0.3 mg (Nihonkayaku Co., Tokyo, Japan) was then administered sublingually, and FBF was measured for 5 min. These studies were carried out in a randomized fashion, and each study proceeded after FBF had returned to baseline. In the preliminary study, after the release of cuff occlusion or the sublingual NTG, FBF returned to baseline within 10 min. Thus, the end of the response to reactive hyperemia or sublingual NTG was followed by a 15-min recovery period. Baseline fasting serum concentrations of total cholesterol, high density lipoprotein (HDL) cholesterol, triglycerides, creatinine, insulin, glucose and electrolytes, as well as plasma renin activity, were obtained after a 30-min rest period.
To evaluate the effects of the antihypertensive drugs on nitric oxide (NO) and prostaglandin release, we measured the FBF response to reactive hyperemia in the presence of a NO synthase inhibitor, NG-monomethyl-l-arginine (L-NMMA) (Sigma Chemical Co., St. Louis, Missouri), and a prostaglandin synthesis inhibitor, indomethacin (Banyu Pharmaceutical Co., Tokyo, Japan), in 13 hypertensive patients treated with ACE inhibitors (9 men and 4 women; mean age 52 ± 12 years), 15 hypertensive patients treated with other agents (calcium antagonists [n = 9], beta-blockers [n = 3] and diuretic agents [n = 3]; 8 men and 7 women; mean age 54 ± 13 years) and 18 untreated hypertensive patients (12 men and 6 women; mean age 50 ± 14 years). Responses of the forearm vasculature to reactive hyperemia after an intra-arterial infusion of L-NMMA or oral administration of indometacin were evaluated on separate occasions. A 23-gauge polyethylene catheter (Hakkow Co., Okayama, Japan) was inserted under local anesthesia (1% lidocaine) into the left brachial artery for infusion of L-NMMA. After maintaining the supine position for 30 min, we measured basal FBF. Then the effect of reactive hyperemia on forearm hemodynamic data was measured. After a 15-min recovery period, L-NMMA was infused intra-arterially at a dose of 8 μmol/min for 5 min before FBF measurement. We performed reactive hyperemia after initiation of a 5-min infusion of L-NMMA or 1 h after oral administration of indomethacin (50 mg).
In the preliminary study, we confirmed the reproducibility of reactive hyperemia and sublingual NTG-induced vasodilation on two separate occasions in 28 healthy male subjects (mean age 27 ± 5 years). The coefficients of variation were 4.3% and 2.8%, respectively.
Venous blood samples were obtained in tubes containing EDTA-sodium (1 mg/ml) and in polystyrene tubes. The EDTA-containing tubes were promptly chilled in an ice bath. Plasma was immediately separated by centrifugation at 3,100 rpm at 4°C for 10 min, and serum at 1,000 rpm at room temperature for 10 min. Samples were stored at −80°C until assayed. Routine chemical methods were used to determine serum concentrations of total cholesterol, HDL cholesterol, triglycerides, creatinine, glucose and electrolytes. The serum concentration of low density lipoprotein was estimated using Friedewald’s method (14).
Results are presented as the mean value ± SD. Values of p < 0.05 were considered significant. The Mann-Whitney Utest was used to evaluate differences between the hypertensive patients and normotensive subjects. Comparisons of time course curves of FBF during reactive hyperemia were analyzed by two-way analysis of variance (ANOVA) for repeated measures on one factor followed by the Bonferroni correction for multiple-paired comparisons. The repeated factor was time of reactive hyperemia and the nonrepeated factor was one group versus the other group (Fig. 1). Multigroup comparisons of variables, including maximal FBF response to NTG (Fig. 2), was carried out by one-way ANOVA followed by the Bonferroni correction. Two-way ANOVA for repeated measures on two factors was used to analyze the effect of L-NMMA or indomethacin on time course curves of FBF during reactive hyperemia. One factor was time of reactive hyperemia, and the other was before versus after L-NMMA or indomethacin administration (Fig. 3). If an interactive effect of drug reached statistical significance, the nature of this interaction was further investigated by applying two-way ANOVA for repeated measured to all six groups (normotensive subjects, untreated hypertensive subjects, those treated with calcium antagonists, ACE inhibitors, beta-blockers and diuretic agents) without consideration of the factorial design. The data were processed using StatView IV (Brainpower) and Super ANOVA (Abacus Concepts) software packages.
The baseline clinical characteristics of the 296 hypertensive patients, including 46 untreated individuals, and the 47 normotensive subjects appear in Table 1. The antihypertensive drugs used in patients with established essential hypertension are listed in Table 2. The systolic and diastolic blood pressures and forearm vascular resistance were significantly higher in the hypertensive patients than in the normotensive subjects. These variables were higher in the untreated hypertensive patients than in the treated hypertensive patients. All other variables were similar in the hypertensive patients and the normotensive subjects (Table 1).
Effect of calcium antagonists, ACE inhibitors beta-blockers and diuretic agents on endothelial function
The time course curve of FBF during reactive hyperemia, an index of endothelium-dependent vasorelaxation, was significantly less in essential hypertensive patients than in normotensive subjects (p < 0.01), regardless of whether they received antihypertensive treatment; the ACE inhibitor (p < 0.01), calcium antagonist (p < 0.01), beta-blocker (p < 0.01), diuretic agent (p < 0.01) and untreated hypertensive groups (p < 0.001) were compared with the normotensive group (Fig. 1). Reactive hyperemia was significantly greater in the ACE inhibitor–treated group than in the calcium antagonist (p < 0.01), beta-blocker (p < 0.01), diuretic agent (p < 0.05) or untreated groups (p < 0.01) but was similar among the last four groups (Fig. 1). Maximal FBF was 46.5 ± 6.2 ml/min per 100 ml tissue in normotensive subjects and 34.1 ± 5.9 ml/min per 100 ml tissue in hypertensive patients (p < 0.05). In the hypertensive subgroups, maximal FBF was 40.5 ± 5.2 ml/min per 100 ml tissue in those treated with ACE inhibitors, 32.9 ± 5.8 ml/min per 100 ml tissue in those treated with calcium antagonists, 34.0 ± 5.6 ml/min per 100 ml tissue in those treated with beta-blockers, 32.1 ± 5.9 ml/min per 100 ml tissue in those treated with diuretic agents and 31.9 ± 5.8 ml/min per 100 ml tissue in untreated hypertensive subjects (p < 0.05 vs. normotensive subjects). Maximal FBF was significantly greater in the ACE inhibitor group than in the calcium antagonist, beta-blocker or diuretic agent group or in the untreated hypertensive subjects (p < 0.05) but was similar among the last four groups. In the ACE inhibitor group, there was no significant difference in reactive hyperemia between imidapril (maximal FBF: 42.1 ± 7.1 ml/min per 100 ml tissue) and enalapril maleate (maximal FBF: 40.2 ± 6.6 ml/min per 100 ml tissue). In the calcium antagonist group, there was no significant difference in reactive hyperemia between amlodipine (maximal FBF: 31.4 ± 6.7 ml/min per 100 ml tissue), slow-release nifedipine (maximal FBF: 33.5 ± 6.4 ml/min per 100 ml tissue), nilvadipine (maximal FBF: 32.7 ± 6.8 ml/min per 100 ml tissue) and nicardipine hydrochloride (maximal FBF: 33.0 ± 6.7 ml/min per 100 ml tissue).
The increase in FBF after sublingual administration of NTG, an index of endothelium-independent vasodilation, was similar among all six groups (normotensive subjects: 5.6 ± 1.3; untreated hypertensive subjects: 5.4 ± 1.4; calcium antagonist group: 5.7 ± 1.4; ACE inhibitor group: 5.3 ± 1.4; beta-blocker group: 5.4 ± 1.5; and diuretic agent group: 5.5 ± 1.3 ml/min per 100 ml tissue, p = NS) (Fig. 2).
Effect of L-NMMA and indomethacin on reactive hyperemia
Intra-arterial infusion of L-NMMA reduced basal FBF (ACE inhibitor group: 4.6 ± 1.3 to 2.4 ± 0.4 ml/min per 100 ml tissue; other agent group: 4.5 ± 1.3 to 2.5 ± 0.5 ml/min per 100 ml tissue; untreated group: 4.4 ± 1.3 to 2.4 ± 0.4 ml/min per 100 ml tissue, p < 0.05) and abolished the augmentation in FBF response to reactive hyperemia in patients treated with ACE inhibitors (p < 0.001) (maximal FBF: 41.8 ± 7.5 to 31.5 ± 7.3 ml/min per 100 ml tissue, p < 0.05). Infusion of L-NMMA reduced reactive hyperemia in the other agent groups and untreated hypertensive groups (p < 0.001) (maximal FBF: 34.2 ± 7.1 to 27.6 ± 6.7 ml/min per 100 ml tissue in the other agent groups; 32.2 ± 6.6 to 26.0 ± 6.9 ml/min per 100 ml tissue in the untreated hypertensive groups, p < 0.05). After L-NMMA infusion, reactive hyperemia was similar among the three groups (p = NS) (Fig. 3, top). L-NMMA infusion did not alter arterial blood pressure or heart rate in any group.
Oral administration of indomethacin did not significantly change arterial blood pressure, heart rate, basal FBF or reactive hyperemia in any group (Fig. 3, bottom). After indomethacin administration, maximal FBF ranged from 40.6 ± 7.1 to 41.1 ± 7.3 ml/min per 100 ml tissue in the ACE inhibitor group; 34.0 ± 6.9 to 33.5 ± 6.7 ml/min per 100 ml tissue in other agent groups; and 31.8 ± 6.8 to 31.0 ± 6.6 ml/min per 100 ml tissue in the untreated hypertensive group (p = NS).
The present study demonstrates that reactive hyperemia, an index of endothelium-dependent vasodilation, is impaired in essential hypertensive patients as compared with normotensive control subjects. In addition, reactive hyperemia in patients treated with ACE inhibitors is significantly greater than that in patients receiving calcium antagonists, beta-blockers and diuretic agents. Furthermore, NO plays an important role in the augmentation of reactive hyperemia in hypertensive patients receiving ACE inhibitors. Although all of these agents were effective in reducing blood pressure, only ACE inhibitors improved endothelial dysfunction.
The forearm vascular response to reactive hyperemia was impaired in patients with essential hypertension, although NTG-induced vasodilation was similar in the hypertensive and normotensive subjects, indicating that in hypertensive patients, endothelial cells, but not smooth muscle cells, are selectively impaired and may be restored. These findings are consistent with previous studies that have demonstrated that endothelium-dependent vasodilation of brachial (1–3), renal (5,6), femoral (15)and small arteries is impaired in essential hypertensive patients as compared with normotensive subjects. Because reactive hyperemia increased in the ACE inhibitor group only, and no significant difference in reactive hyperemia existed between the different ACE inhibitors or calcium antagonists, the effect of antihypertensive therapy on endothelial function must not be specific for one single medication. Reactive hyperemia in hyperten-sive patients treated with ACE inhibitors was less than that in normotensive subjects, suggesting that antihypertensive therapy with ACE inhibitors may not completely restore endothelial function to the level of that in normotensive subjects.
Several experimental (12,13)and clinical studies (8–10,16,17)support our results demonstrating that ACE inhibitors can restore endothelial function, whereas other studies report that antihypertensive therapy does not alter endothelial function in hypertension (11,18). In the present study, an NO synthase inhibitor, L-NMMA, abolished the augmentation of reactive hyperemia in patients treated with ACE inhibitors, suggesting that NO plays a major role in this process. It is well known that a balance between angiotensin II and NO is important in the regulation of vascular tone (19). Angiotensin II increases vascular superoxide production through activation of membrane-associated NADH/NADPH oxidase, resulting in NO inactivation and toxic peroxynitrite production (19). Therefore, ACE inhibitors may increase NO by inhibiting angiotensin II production. Furthermore, under physiologic conditions, endogenous bradykinin is limited by the ACE. Bradykinin binds to B2receptors on the endothelial cell surface to release NO (20). The ACE inhibitors prevent bradykinin degradation, leading to increases in NO release.
In addition, the decreased bradykinin degradation induced by ACE inhibition has been shown to increase prostaglandins (21). Because the prostaglandin synthesis inhibitor indomethacin did not affect reactive hyperemia, prostaglandins do not contribute to the augmentation of reactive hyperemia by ACE inhibitors. These findings are consistent with a previous study using aspirin, a cyclooxygenase inhibitor (22).
Although various types of antihypertensive drugs in the present study were equally effective in reducing blood pressure, the forearm vascular response to reactive hyperemia was greater in the ACE inhibitor group than in the other three groups. In addition, although calcium antagonists, beta-blockers and diuretic agents all reduced blood pressure significantly, the reactive hyperemia in these groups was similar to that in the untreated hypertensive group. Our findings are consistent with the results of previous studies showing that blood pressure reduction with calcium antagonists and beta-blockers does not directly alter endothelium-dependent vasodilation in brachial, renal and small arteries in essential hypertensive patients (8–10). Thus, the reduction in blood pressure with effective antihypertensive therapy may not always lead to improved endothelial function in patients with essential hypertension.
The use of agonists to stimulate NO release, such as acetylcholine or methacholine, would allow us to draw more specific conclusions regarding the role of basal and stimulated release of NO by antihypertensive agents in the forearm circulation. Recently, Celermajer et al.(23)demonstrated that a noninvasive method of reactive hyperemia is useful in assessing endothelial function, instead of the method that uses infusion of intra-arterial vasoactive agents. Indeed, in the present study this technique was simple and reproducible and did not lead to adverse effects. We recommend that this noninvasive technique be used to assess endothelial function both in future studies and in routine clinical examinations.
Several limitations should be considered in assessing this report. First, this is not a double-blind, placebo-controlled clinical study. Second, the number of patients in each group is unmatched and relatively small, especially in the diuretic and beta-blocker groups. Therefore, the effect of diuretic agents or beta-blockers might fail to achieve statistical significance.
Among randomly recruited hypertensive patients who received monotherapy, the frequency of use of calcium antagonists was ∼70%, ACE inhibitors 20%, beta-blockers 6% and diuretic agents 4%. This frequency of antihypertensive agents used as monotherapy is consistent with a Japanese survey that shows that Japanese hypertension specialists prefer calcium antagonists and ACE inhibitors (24). In the U.S. (25)and Europe (26), beta-blockers and diuretic agents are the most widely prescribed antihypertensive drugs for first-line therapy, although recently the use of calcium antagonists and ACE inhibitors has increased.
We have demonstrated that among the most frequently prescribed classes of antihypertensive agents, only ACE inhibitors augment the forearm vascular response to reactive hyperemia through increases in NO production in patients with essential hypertension. In addition, a noninvasive method of evaluating reactive hyperemia is a useful alternative to intra-arterial infusions of vasoactive agents in the assessment of endothelial function in the forearm circulation.
The authors thank Kazuo Mizumoto, MD, PhD, Yoriaki Matsuishi, MD, PhD (Matsuishi Hospital), Hiroko Nishi, MD, Kazuto Jinushi, MD, PhD, Hideki Nomura, MD, PhD (Nomura Hospital), Yasushi Toyota, MD, Shohei Ishimaru, MD, PhD, Kazuma Shimura, MD, PhD (Shimura Hospital), Katsuhiko Ishibashi, MD, PhD (Cyugoku Electric Co. Hospital), Nobuo Sasaki, MD (Mihara Medical Association Hospital), Mitsuhiro Sanada, MD, PhD, Kozo Ohama, MD, PhD (Department of Gynecology, Hiroshima University), Chikara Goto, Mitsutoshi Kawamura, PhD, and Isao Nara, PhD (Division of Physical Theraphy, Institute of Health Sciences, Hiroshima University) for the recruitment of hypertensive patients and normotensive subjects, and Hiroaki Ikeda, MD, PhD, for the preparation of the L-NMMA, and Yuko Omura for her secretarial assistance.
☆ This study was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan (Dr. Oshima), the Japan Heart Foundation’s Grant for Research on Hypertension and Vascular Metabolism (Dr. Higashi) and a grant from the Research Foundation for Community Medicine (Dr. Higashi).
- angiotensin-converting enzyme
- analysis of variance
- forearm blood flow
- high density lipoprotein
- nitric oxide
- Received February 8, 1999.
- Revision received August 24, 1999.
- Accepted October 18, 1999.
- American College of Cardiology
- Linder L,
- Kiowski W,
- Buhler F.R,
- Lüscher T.F
- Panza J.A,
- Quyyumi A.A,
- Brush J.E Jr.
- Taddei S,
- Virdis A,
- Ghiadoni L,
- Magagna A,
- Salvetti A
- Higashi Y,
- Oshima T,
- Ozono R,
- et al.
- Lahera V,
- Navarro-Cid J,
- Cachoferio V,
- Garcia-Estan J,
- Ruilope L.M
- Higashi Y,
- Oshima T,
- Sasaki S,
- et al.
- Hirata Y,
- Hayakawa H,
- Kakoki M,
- et al.
- Friedewald W.T,
- Levy R.I,
- Fredrickson D.S
- Bode-Böger S.M,
- Böger R.H,
- Alfke H,
- et al.
- Panza J.A,
- Quyyumi A.A,
- Callahan T.S,
- Epstein S.E
- Daemen M.J.A.P,
- Lombardi D.M,
- Bosman F.T,
- Schwarts S.M
- Vanhotte P.M,
- Boulanger C.M,
- Illiano S.C,
- Nagao T,
- Yidal M,
- Mombouli J.V
- Miller V.M,
- Vanhoutte P.M