The endothelium has important regulatory effects on

Transcription

1 AJH 2001; 14: Endothelial Dysfunction in Resistance Arteries Is Related to High Blood Pressure and Circulating Low Density Lipoproteins in Previously Treated Hypertension Michael Hecht Olsen, Kristian Wachtell, Christian Aalkjaer, Richard B. Devereux, Harriet Dige-Petersen, and Hans Ibsen Background: Peripheral endothelial dysfunction has been demonstrated in hypertension. However, its relationship to blood pressure (BP) load, vascular structure, and metabolic disturbances in patients with long-standing, previously treated hypertension is unclear. Methods: A total of 41 patients with stage I to III essential hypertension and electrocardiographic left ventricular hypertrophy were studied. After 2 to 3 weeks of placebo treatment we measured nitroprusside-induced relaxation (NIR), acetylcholine-induced relaxation (AIR), and media:lumen ratio in isolated, subcutaneous resistance arteries by myography, as well as 24-h ambulatory BP, and serum lipids. Results: Maximal AIR correlated negatively with median 24-h diastolic BP (r 0.42, P.01), and sensitivity to AIR correlated negatively with serum low density lipoprotein (LDL) (r 0.36, P.05). In multiple regression analyses, sensitivity to AIR correlated negatively with serum LDL ( 0.33) independently of maximal NIR ( 0.41) (adjusted R , P.01). Maximal acetylcholine-induced relaxation correlated negatively with median 24-h diastolic BP ( 0.38) independently of maximal NIR ( 0.45) (adjusted R , P.001). Acetylcholine-induced relaxation was not significantly related to diabetes or to media:lumen ratio (r 0.26, NS). Conclusions: High diastolic BP and high serum LDL were associated with impaired maximal AIR and reduced sensitivity to AIR, respectively, independently of smooth muscle cell responsiveness to nitroprusside. This indicated decreasing endothelial function in small resistance arteries with increasing BP and increasing LDL in hypertension. Endothelial function was not significantly related to vascular structure of the resistance arteries or to diabetes in these patients with long-standing hypertension. Am J Hypertens 2001;14: American Journal of Hypertension, Ltd. Key Words: Hypertension, endothelial function, vascular structure, lipids. The endothelium has important regulatory effects on cardiovascular function through release of vasodilators and vasoconstrictors, playing a potential role in the regulation of peripheral resistance. 1 In patients with hypertension, tonic nitric oxide release from the endothelium may be impaired 2 and the arteriolar media:lumen ratio (MLR) is increased. 3 In patients with hypertension endothelial function, assessed by acetylcholine-induced vasodilatation in the forearm or acetylcholine-induced relaxation (AIR) in isolated subcutaneous resistance arteries has been found impaired in a number of studies, 4 but not in all. 5 As nitric oxide can inhibit smooth muscle cell growth, endothelial dysfunction may be associated with vascular remodeling and hypertrophy. 6 However, Kiowski et al 7 found that treatment with an angiotensin converting enzyme inhibitor reduced minimal forearm vascular resistance without changing endothelial function assessed as acetylcholine-induced vasodilatation in the forearm, and a Received November 15, Accepted February 13, From the Department of Clinical Physiology and Nuclear Medicine (MHO, HD-P), Internal Medicine (KW, HI), and Clinical Experimental Research (MHO, HD-P), Glostrup Hospital, University of Copenhagen, Copenhagen, Denmark; Institute of Physiology (CA), University of Århus, Århus, Denmark; Department of Medicine, Division of Cardiology, Weill Medical College of Cornell University (KW, RBD), New York, New York. Supported in part by grants from King Christian the Tenths Foundation, The Foundation of Eva and Robert Voss Hansen, and Merck & Co., Inc., Copenhagen, Denmark, as a part of the LIFE-ICARUS substudy. Address correspondence and reprint requests to Dr. Michael Hecht Olsen, Department of Clinical Physiology and Nuclear Medicine, Glostrup Hospital, University of Copenhagen, DK-2600 Glostrup, Denmark; mho@dadlnet.dk 2001 by the American Journal of Hypertension, Ltd. Downloaded from Published by Elsevier Science Inc /01/$20.00 PII S (01)

2 862 HYPERTENSION AND ENDOTHELIAL FUNCTION AJH September 2001 VOL. 14, NO. 9, PART 1 recent article by Rizzoni et al, 8 looking at patients with essential or secondary hypertension, indicates that endothelial function in isolated, subcutaneous resistance arteries is more closely related to blood pressure (BP) load than to vascular structure of the arteries. These relationships might be disturbed during long-standing hypertension, possibly due to antihypertensive treatment and metabolic abnormalities known to impair endothelial function. 9 Therefore, this study was undertaken to investigate the relative strength of association between endothelial function of subcutaneous resistance arteries assessed in vitro as AIR and BP load, vascular structure, diabetes, and serum lipids in older patients with long-standing hypertension and electrocardiographic left ventricular (LV) hypertrophy. Materials and Methods Subjects We studied 41 patients (34 men and seven women, aged years [mean 65 6 years]) with essential hypertension and left ventricular hypertrophy as determined by screening electrocardiography before study entry, after 2 to 3 weeks of placebo treatment. All patients were participants in the Insulin Carotids US Scandinavia (ICARUS) substudy 10 of the Losartan Intervention For Endpoint- Reduction in Hypertension (LIFE) trial and met the study s inclusion and exclusion criteria. 11 They had all given informed consent and the study was accepted by the Danish Committee of Ethical Science. There were nine active smokers, 18 ex-smokers, and 14 nonsmokers. The mean duration of known hypertension was 7 years and 7 months, ranging from 0 to 35 years. Nine patients had newly diagnosed hypertension and three of these patients had received antihypertensive treatment for a few months. Before the placebo washout period, five patients had received -receptor blockers or diuretics only. The remaining 30 had received vasodilators such as calcium antagonists, angiotensin converting enzyme inhibitors, or angiotensin II receptor antagonists. Glucose intolerance was found in 29 patients, 15 of whom had diabetes according to criteria of the World Health Organization. 12 Additional patient characteristics are listed in Table 1. Table 1. Patient characteristics Women/men 7/34 Age (y) 65 (64 67) Height (cm) 172 ( ) Weight (kg) 84 (81 88) Body surface area (m 2 ) 1.98 ( ) Body mass index (kg/m 2 ) 28.4 ( ) Plasma glucose (mmol/l) 6.5 ( ) Plasma insulin (pmol/l) 55 (44 66) Serum triglycerides (mmol/l) 1.68 ( ) Total serum cholesterol (mmol/l) 6.0 ( ) Serum HDL (mmol/l) 1.26 ( ) Serum LDL (mmol/l) 4.0 ( ) Serum VLDL (mmol/l) 0.74 ( ) Office systolic blood pressure (mm Hg) 174 ( ) Office diastolic blood pressure (mm Hg) 95 (92 98) Daytime ambulatory systolic BP (mm Hg) 157 ( ) Daytime ambulatory diastolic BP (mm Hg) 88 (85 92) HDL high density lipoprotein; LDL low density lipoprotein; VLDL very low density lipoprotein; BP blood pressure. Data are expressed as mean values with 95% confidence intervals. Methods Echocardiography was performed according to a previously used echocardiographic protocol. 13 Standardized examinations included two-dimensional guided M-mode echocardiography and selected two-dimensional recordings. Studies were performed using a high-quality commercially available echocardiograph (VingMed CFM-800, General Electric, Norway) equipped with probes 3.0 to 3.5 MHz and 2.0 to 2.5 MHz along with a super-vhs video recorder. Measurements were made blindly at the Echocardiography Reading Center at The New York Presbyterian Hospital-Weill Medical College of Cornell University, New York, by experienced physician readers using computerized review stations (Digisonics, Inc., Houston, TX) equipped with digitizing tablet and monitor screen overlay for calibration and performance of each needed measurement. The LV dimensions were measured according to recommendations of the American Society of Echocardiography. When optimal orientation of the LV or atrial imaging views could not be obtained, correctly oriented two-dimensional linear dimension measurements were made by the leading-edge convention. End-diastolic LV dimensions were used to calculate LV mass by an anatomically validated formula that gives values closely related to autopsy LV weight (r 0.90), which showed excellent reproducibility. A subcutaneous biopsy specimen measuring approximately 1 cm 3 was taken from the gluteal region under local anesthesia. The biopsy specimen was placed in a cold physiologic saline solution. The small resistance arteries (100 to 500 m) were isolated under a microscope and mounted on a wire myograph (Automated Dual Wire Myograph System, model 500A, J. P. Trading I/S, Århus, Denmark). The thickness of the different layers and lumen were measured, and the vessels were normalized to 100 mm Hg distension pressure based on the passive length tension curve 14 for calculation of MLR. In most patients, MLR represented the mean of two resistance arteries defined as vessels with a normalized internal diameter 500 m. All vessels were stretched to 90% of the distension at 100 mm Hg and exposed to cumulatively increasing con-

3 AJH September 2001 VOL. 14, NO. 9, PART 1 HYPERTENSION AND ENDOTHELIAL FUNCTION 863 centrations of norepinephrine (0.01, 0.03, 0.1, 0.3, 0.5, 1.0, 3.0, 5.0, 10, 30, and 50 mol/l) until maximal constriction; the concentrations giving between 80% and 100% of maximal constriction were chosen as preconstrictors. For preconstriction we used 3 mol/l norepinephrine in 14 patients, 1 mol/l in 24 patients, and 0.5 mol/l in three patients. After submaximal preconstriction with norepinephrine, acetylcholine was added in cumulatively increasing amounts from to 100 mol/l in steps of half-log units until maximal relaxation. Based on the results of others, 4 a maximal relaxation 90% was considered as normal. After a washout of acetylcholine and submaximal preconstriction with norepinephrine, nitroprusside from to 100 mol/l in steps of half-log units and nitroprusside 1 and 10 mmol/l were added in cumulatively increasing amounts until maximal relaxation. The maximal relative dilatation and sensitivity, assessed as the negative logarithm of the concentration leading to 50% of maximal effect (pd 2 ), were calculated using GraphPad Prism. Although the wire myograph is thought to be less reliable when studying endothelial function as compared with the pressure myograph, we chose the wire myograph because we wanted to measure vascular structure as well as endothelial function. In addition, Schiffrin et al have demonstrated that both methods provide closely correlated functional results. 15 Ambulatory BP was measured with an automatic BP device using the cuff-oscillometric method (Takeda TM- 2421, A&D Co, Ltd, Tokyo, Japan). 16 It was applied after 1 h of resting to the left arm and worn for 24 h. Oscillometric BP were measured every half hour during daytime and every hour at night (11 PM to 6 AM). In 35 patients we had 20 readable measurements during 24 h, which was considered sufficient to calculate median values for 24-h BP for each patient. 17 For assays, plasma glucose concentrations were measured using a Beckman glucose analyzer 2 (Beckman Instruments Inc., Fullerton, CA) and a glucose oxidase method. Plasma insulin concentrations were determined by enzyme immunoassay, as described by Andersen et al. 18 Statistical Analysis For data management and statistical analyses, Statistica, version 5.1 (StatSoft, Inc, Tulsa, OK) was used. Parametric statistics were used calculating mean values and 95% confidence intervals. We performed simple linear regression analyses calculating the regression quotient (r) and stepwise, backward, multiple linear regression analyses calculating the standardized regression quotient for each parameter ( ) and the common adjusted regression quotient for the model (adjusted R 2 ). Two-tailed values of P.05 were considered statistically significant. Results Table 2. Cardiovascular characteristics LV mass (g) 240 ( ) LV mass index (g/m 2 ) 122 ( ) LV PWTD (mm) 1.09 ( ) Measurements of resistance arteries in vitro Lumen diameter ( m) 280 ( ) Media thickness ( m) 31 (29 33) MLR (myography) (%) 11.4 ( ) MLR (histology) (%) 10.4 ( ) pd 2 NE (logmol/l) 6.8 ( ) pd 2 NP (logmol/l) 6.2 ( ) Maximal NIR (%) 85 (81 89) pd 2 Ach (logmol/l) 7.4 ( ) Maximal AIR (%) 82 (75 90) LV left ventricular; PWTD posterior wall thickness in enddiastole; MLR media:lumen ratio; pd 2 minus logarithm to the concentration at 50% of maximal effect; NE norepinephrine; NP nitroprusside; Ach acetylcholine; NIR nitroprusside-induced relaxation; AIR acetylcholine-induced relaxation. Data are expressed as mean values with 95% confidence intervals. Cardiovascular characteristics are listed in Table 2. In simple regression analyses, maximal AIR correlated negatively with median value of 24-h diastolic BP (Fig. 1), negatively with pd 2 for norepinephrine, and positively with maximal nitroprusside-induced relaxation (NIR) (Table 3). However, maximal AIR did not relate to diabetes, serum lipids, LV mass, or MLR. Maximal NIR was not related to median 24-h diastolic BP (r 0.14, NS). In multiple regression analyses maximal AIR correlated negatively with median 24-h diastolic BP ( 0.38) and positively with maximal NIR ( 0.45) (adjusted R , P.001). AIR was not significantly related to median 24-h systolic BP or to office BP. In simple regression analyses, the pd 2 for acetylcholine was not significantly correlated with any of the BP measurements. It did correlate negatively with serum low density lipoprotein (LDL) (Fig. 2) and positively with the maximal nitroprusside-induced relaxation (NIR) and pd 2 for nitroprusside. It was not related to MLR, LV mass, or diabetes (Table 3). In two different models using multiple regression analyses, pd 2 for acetylcholine correlated negatively with serum LDL ( 0.33, Fig. 1) and positively with maximal NIR ( 0.41) (adjusted R , P.01); or, alternatively, it correlated negatively with serum LDL ( 0.33) and positively with pd 2 for nitroprusside ( 0.33) (adjusted R , P.01). In simple regression analyses, MLR measured by myography correlated positively to end-diastolic posterior LV wall thickness (r 0.35, P.05). The MLR measured by myography was not significantly related to 24-h systolic BP (r 0.22, NS) or 24-h diastolic BP (r 0.10, NS). The structural and functional parameters of the isolated resistance arteries were not related to previous antihypertensive treatment. Although median 24-h systolic BP tended to be higher in active smokers (161 [155 to 166] mm Hg v 151 [144 to 159] mm Hg, NS), there were no smoking-related differences in median 24-h diastolic BP or in structural and

4 864 HYPERTENSION AND ENDOTHELIAL FUNCTION AJH September 2001 VOL. 14, NO. 9, PART 1 FIG. 1. The relationship between median 24-h diastolic blood pressure and maximal acetylcholine-induced relaxation in isolated, subcutaneous resistance arteries in vitro. Filled circles represent active smokers and open circles represent nonactive smokers. functional parameters of the isolated resistance arteries. Furthermore, none of the regression quotients from the correlations between maximal AIR and median 24-h diastolic BP and between sensitivity to AIR and serum LDL were markedly affected by disregarding active smokers (r 0.37, P.07 and r 0.32, P.08). Discussion Despite long-standing hypertension in this group of older patients (among which the majority were glucose intolerant and were previously treated with different antihypertensive agents), we demonstrated a significant impairment Table 3. function Acetylcholine-induced relaxation in relation to age, metabolic factors, vascular structure, and Maximal AIR Sensitivity to AIR N R P Value R P Value Age (y) NS Insulin sensitivity (M) NS 0.02 NS Insulin sensitivity (M/IG) NS 0.15 NS Glucose tolerance NS 0.02 NS Total plasma cholesterol NS Plasma high density lipoprotein NS 0.11 NS Plasma low density lipoprotein NS Median 24-h systolic BP NS 0.10 NS Median 24-h diastolic BP NS Left ventricular mass NS 0.20 NS Left ventricular mass index NS 0.22 NS Measurements of resistance arteries in vitro Maximal NIR pd 2 NP NS pd 2 NE NS Media:lumen ratio (myography) NS 0.09 NS M whole body glucose uptake; NS not significant; M/IG whole body glucose uptake index; other abbreviations as in Tables 1 and 2.

5 AJH September 2001 VOL. 14, NO. 9, PART 1 HYPERTENSION AND ENDOTHELIAL FUNCTION 865 FIG. 2. The relationship between serum low density lipoproteins and sensitivity to acetylcholine-induced relaxation (negative logarithm of the concentration leading to 50% of maximal effect) in isolated, subcutaneous resistance arteries in vitro. Filled circles represent active smokers and open circles represent nonactive smokers. of endothelial function as indicated by a nitroprussideindependent decrease in maximal AIR and sensitivity to AIR with increasing BP load and increasing levels of LDL, respectively. Endothelial dysfunction was not significantly related to diabetes or to vascular remodeling. The relationship between endothelial dysfunction and high diastolic BP confirms findings by Rizzoni et al, 8 who demonstrated that AIR was negatively related to 24-h ambulatory BP. Based on previously published results, endothelial function could be related to both systolic BP due to increased sheer stress 1 and to diastolic BP due to its closer relation to peripheral resistance artery remodeling. 19 In our study, endothelial function was more closely related to diastolic than to systolic BP. The fact that we were able to demonstrate a significant correlation between maximal AIR and 24-h ambulatory diastolic BP after 2 to 3 weeks of placebo treatment in previously treated hypertensive patients may suggest that endothelial function changes quite rapidly with changing BP within a period of a few weeks. This is supported by animal experiments as well as in human studies in which endothelial function improved during either 4 weeks 20 or 3 months 21 of antihypertensive treatment, and decreased progressively with increasing BP after cessation of treatment. 22 Our results also indicate that the relationship between actual BP burden and endothelial function is so robust that it still exists despite the presence of a number of confounders such as older age, long-standing hypertension with LV hypertrophy, previous antihypertensive treatment, and abnormal glucose metabolism. The observation that endothelial dysfunction in hypertension is not significantly related to vascular remodeling is supported by observations in human subjects as well as in animal experiments by Rizzoni et al 8,23,24 that demonstrated parallel changes in AIR and BP independently of changes in MLR. Because vascular structure changes slowly 25 and endothelial function has been shown to respond quickly to changes such as those in serum lipids, 26 it is possible that the lack of relationship between endothelial function and vascular structure in subcutaneous resistance arteries in our study reflects, as indicated by animal studies, 23 a different time course of responsiveness of the endothelial layer and the vascular wall to short-term changes in BP load. At the level of the medium-sized conduit arteries in our hypertensive patients we found a relationship between impaired flow-mediated dilation in the brachial artery and high intima-media thickness of the common carotid artery 27 and high LV mass. 28 However, as nitroglycerin-induced dilation in the brachial artery was compromised as well as flow-mediated dilation in patients with vascular hypertrophy, 27 the reduced flow-mediated dilatation may have been due to reduced vasodilatory capacity as well as to endothelial dysfunction. In this in vitro study, maximal AIR was unrelated to serum lipids, diabetes, and, as previously reported, to insulin sensitivity, 29 suggesting that the relative impor-

6 866 HYPERTENSION AND ENDOTHELIAL FUNCTION AJH September 2001 VOL. 14, NO. 9, PART 1 tance of metabolic disturbances for vascular function is less than that of the BP load in these older patients with long-standing previously treated hypertension. This is in agreement with our in vivo findings that flow-mediated dilation in the brachial artery was inversely related to 24-h ambulatory BP but was unrelated to serum lipids and plasma glucose. 27 The lack of association between cardiovascular risk factors such as diabetes, smoking, and hypertension (which are known normally to coexist) is probably due to selection bias, and may also explain the lack of relationship between endothelial function, smoking, and metabolic disturbances in this group of patients. The patients in this study were high-risk patients with long-standing hypertension who were selected for the LIFE study. The majority of patients had received various kinds of antihypertensive treatment, but an untreated period of more than 2 to 3 weeks was considered to be unacceptable. We were not able to detect any influence of previous antihypertensive treatment on either vascular structure or function. We do not think that the lack of correlation between endothelial function on one hand and vascular structure and diabetes on the other is due to difficulties in measuring MLR, as MLR is correlated to LV wall thickness; instead, in this rather small sample size with many confounders, the lack of correlation may be due to type two errors. Despite our finding of a negative correlation between maximal AIR and 24-h diastolic BP, the sensitivity to AIR did not correlate significantly with the ambulatory BP. The sensitivity to AIR decreased with increasing serum concentrations of LDL independently of the smooth muscle cell response to nitroprusside, but did not relate significantly to diabetes or vascular structure. Serum LDL has been associated with increased production of oxygen-derived free radicals 30 known to be able to inactivate nitric oxide. 31 Therefore, increased inactivation of nitric oxide released from the endothelium may explain the negative correlation between the sensitivity to AIR and the serum concentration of LDL. Goode et al 9 found reduced sensitivity to AIR as well as impaired maximal AIR in isolated resistance arteries from patients with hypercholesterolemia, and Schiffrin et al 15 recently found a correlation between total serum cholesterol and maximal AIR in isolated resistance arteries from patients with hypertension. In our study, serum LDL was not significantly associated with maximal AIR, perhaps because the increased inactivation of nitric oxide was overcome by increased nitric oxide release after the high concentrations of acetylcholine, or perhaps because the hyperpolarizing factor (which is not inactivated by free radicals) was sufficient to induce relaxation in patients with high serum LDL. 32 In conclusion, high BP and high serum LDL were (independently of smooth muscle cell responsiveness to nitroprusside) associated with impaired maximal AIR and reduced sensitivity to AIR, respectively, indicating endothelial dysfunction in isolated, resistance arteries in patients with high BP or high circulating LDL. Endothelial function was not significantly related to vascular structure of the resistance arteries or to diabetes in patients with long-standing, previously treated hypertension, which may reflect a dissociation in the course of development and/or regression of functional and structural vascular changes. Our results also indicate that the relationship between actual BP burden and endothelial function must be very robust as it exists in spite of the presence of a number of confounders such as high age, long-standing hypertension with LV hypertrophy, previous antihypertensive treatment, and abnormal glucose metabolism. Acknowledgments We acknowledge the technical assistance and scientific contributions by Michael J. Mulvany, Institute of Pharmacology in Århus, Denmark. We also acknowledge the work by Ruth Peterson. References 1. Boegehold MA: Heterogeneity of endothelial function within the circulation. Curr Opin Nephrol Hypertens 1998;7: Panza JA, Casino PR, Kilcoyne CM, Quyyumi AA: Role of endothelium-derived nitric oxide in the abnormal endothelium-dependent vascular relaxation of patients with essential hypertension. Circulation 1993;87: Heagerty AM, Aalkjaer C, Bund SJ, Korsgaard N, Mulvany MJ: Small artery structure in hypertension. Dual processes of remodeling and growth. Hypertension 1993;21: Schiffrin EL, Deng LY: Comparison of effects of angiotensin I-converting enzyme inhibition and beta-blockade for 2 years on function of small arteries from hypertensive patients. Hypertension 1995;25: Thybo NK, Mulvany MJ, Jastrup B, Nielsen H, Aalkjaer C: Some pharmacological and elastic characteristics of isolated subcutaneous small arteries from patients with essential hypertension. J Hypertens 1996;14: Intengan HD, Schiffrin EL: Role of endothelium in modulation of structural changes of small arteries in hypertension: effects of therapeutic intervention. J Hypertens 1998;16(Suppl 8):S97 S Kiowski W, Linder L, Nuesch R, Martina B: Effects of cilazapril on vascular structure and function in essential hypertension. Hypertension 1996;27: Rizzoni D, Porteri E, Castellano M, Bettoni G, Muiesan ML, Tiberio G, Giulini SM, Rossi GP, Bernini GP, Agabiti Rosei E: Endothelial dysfunction in hypertension is independent from the etiology and from vascular structure. Hypertension 1998;31: Goode GK, Heagerty AM: In vitro responses of human peripheral small arteries in hypercholesterolemia and effects of therapy. Circulation 1995;91: Olsen MH, Fossum E, Hjerkinn E, Wachtell K, Høieggen A, Nesbitt SD, Andersen UB, Phillips RA, Gaboury CL, Ibsen H, Kjeldsen SE, Julius S: Relative influence of insulin resistance versus blood pressure on vascular changes in longstanding hypertension. ICARUS, a LIFE substudy. J Hypertens 2000;18: Dahlof B, Devereux RB, Julius S, Kjeldsen SE, Beevers G, de Faire U, Fyhrquist F, Hedner T, Ibsen H, Kristianson K, Lederballe- Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H: Characteristics of 9194 patients with left ventricular hypertrophy: the LIFE study. Hypertension 1998;32: World Health Organization: WHO Technical Reports Series 727. Geneva,1985; p 9.

7 AJH September 2001 VOL. 14, NO. 9, PART 1 HYPERTENSION AND ENDOTHELIAL FUNCTION Devereux RB, Roman MJ, de Simone G, O Grady MJ, Paranicas M, Yeh JL, Fabsitz RR, Howard BV: Relations of left ventricular mass to demographic and hemodynamic variables in American Indians: the Strong Heart Study. Circulation 1997;96: Mulvany MJ, Halpern W: Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res 1977;41: Schiffrin EL, Park JB, Intengan HD, Touyz RM: Correction of arterial structure and endothelial dysfunction in human essential hypertension by the angiotensin receptor antagonist losartan. Circulation 2000;101: Imai Y, Sasaki S, Minami N, Munakata M, Hashimoto J, Sakuma H, Sakuma M, Watanabe N, Imai K, Sekino H, Abe K: The accuracy and performance of the A&D TM 2421, a new ambulatory blood pressure monitoring device based on the cuff-oscillometric method and the Korotkoff sound technique. Am J Hypertens 1992;5: Conway J, Coats A: Value of ambulatory blood pressure monitoring in clinical pharmacology. J Hypertens 1989;7(Suppl):S29 S Andersen L, Dinesen B, Jorgensen PN, Poulsen F, Roder ME: Enzyme immunoassay for intact human insulin in serum or plasma. Clin Chem 1993;39: Schiffrin EL, Deng LY: Relationship between small-artery structure and systolic, diastolic and pulse pressure in essential hypertension. J Hypertens 1999;17: Li JS, Schiffrin EL: Effect of short-term treatment of SHR with the novel calcium channel antagonist mibefradil on function of small arteries. Am J Hypertens 1997;10: Mühlen Bvz, Kahan T, Millgård J, Hägg A, Lind L: Both irbesartan and atenolol improved endothelium-dependent vasodilatation in the human forearm in hypertensive patients. A randomised, doubleblind trial. J Hypertens 2000;18(Suppl 2):S Rizzoni D, Castellano M, Porteri E, Bettoni G, Muiesan ML, Agabiti Rosei E: Delayed development of hypertension after short-term nitrendipine treatment. Hypertension 1994;24: Rizzoni D, Castellano M, Porteri E, Bettoni G, Muiesan ML, Agabiti-Rosei E: Vascular structural and functional alterations before and after the development of hypertension in SHR. Am J Hypertens 1994;7: Rizzoni D, Castellano M, Porteri E, Bettoni G, Muiesan ML, Cinelli A, Agabiti Rosei E: Effects of low and high doses of fosinopril on the structure and function of resistance arteries. Hypertension 1995; 26: Schiffrin EL, Deng LY, Larochelle P: Progressive improvement in the structure of resistance arteries of hypertensive patients after 2 years of treatment with an angiotensin I-converting enzyme inhibitor. Comparison with effects of a beta-blocker. Am J Hypertens 1995;8: Vogel RA, Corretti MC, Plotnick GD: Effect of a single high-fat meal on endothelial function in healthy subjects. Am J Cardiol 1997;79: Olsen MH, Wachtell K, Dige-Petersen H, Rokkedal J, Ibsen H: Severe hypertension is associated with reduced flow-mediated vasodilatation in the brachial arteries and thickening of the carotid arteries. A LIFE substudy (abst). Am J Hypertens 1999;12:177A. 28. Olsen MH, Hermann KL, Dige-Petersen H, Rokkedal J, Ibsen H: Parallel changes in left ventricular mass and vasodilatory response of the brachial artery wall in patients with longstanding hypertension. A LIFE substudy (abst). Circulation 1999;100:I Olsen MH, Andersen UB, Wachtell K, Ibsen H, Dige-Petersen H: A possible link between endothelial dysfunction and insulin resistance in hypertension. A LIFE substudy. Blood Press 2000;9: Ohara Y, Peterson TE, Harrison DG: Hypercholesterolemia increases endothelial superoxide anion production. J Clin Invest 1993; 91: Tschudi MR, Mesaros S, Lüscher TF, Malinski T: Direct in situ measurement of nitric oxide in mesenteric resistance arteries. Increased decomposition by superoxide in hypertension. Hypertension 1996;27: Prieto D, Simonsen U, Hernandez M, Garcia-Sacristan A: Contribution of K channels and ouabain-sensitive mechanisms to the endothelium-dependent relaxations of horse penile small arteries. Br J Pharmacol 1998;123: