the Cardiovascular Research Institute at the Washington Hospital Center, Washington, DC.
Abstract
Background— Patients with cardiovascular risk factors have endothelial dysfunction, a key element in the pathogenesis of atherosclerosis. The thiazolidinediones have been shown to exert multiple antiatherosclerotic actions in diabetic patients. This study tested the hypothesis that pioglitazone improves endothelial function in nondiabetic patients with major risk factors.
Methods and Results— The study had a randomized, double-blind, placebo-controlled, crossover design. Eighty patients with either hypertension or hypercholesterolemia were enrolled. Insulin sensitivity was assessed by the Quantitative Insulin Sensitivity Check Index (QUICKI), and patients were further classified as insulin sensitive or insulin resistant. In each treatment phase, patients received either pioglitazone 45 mg daily or placebo for 8 weeks. Endothelial function and laboratory tests were performed at the end of each 8-week period. Treatment with pioglitazone significantly lowered plasma insulin (–22.9%; P<0.001), improved QUICKI insulin sensitivity index (3.7%; P<0.001), increased HDL cholesterol (8.2%; P<0.001), and reduced triglycerides (–15.1%; P=0.003), free fatty acids (–14%; P=0.005), and C-reactive protein (–28.6%; P=0.001). Pioglitazone treatment significantly improved endothelium-dependent dilation to bradykinin (P=0.01) without affecting the response to sodium nitroprusside (P=0.31). In multivariable analysis, only changes in total cholesterol were predictors of improved endothelial reactivity with pioglitazone.
Conclusions— In nondiabetic patients with cardiovascular risk factors, pioglitazone treatment enhances insulin sensitivity, decreases C-reactive protein, and improves endothelial vasodilator function. These effects do not appear to be closely related, suggesting that pioglitazone may have beneficial vascular properties independent of its effect on insulin sensitivity and inflammation.
Key Words: atherosclerosis endothelium-derived factors inflammation risk factors vasodilation
Introduction
The endothelium regulates vascular wall homeostasis by maintaining a relaxed vascular tone and low levels of oxidative stress, as well as modulating platelet aggregation and leukocyte adhesion.1 In response to a variety of noxious stimuli, the endothelium undergoes a phenotypic modulation to a nonadaptive state, characterized by enhanced expression of adhesion molecules, synthesis of proinflammatory and prothrombotic factors, increased oxidative stress, and abnormal modulation of vascular tone. This state of endothelial dysfunction is central in the development of atherosclerosis2 and may contribute to the clinical expression of the disease. Indeed, patients with cardiovascular risk factors but no clinical evidence of atherosclerosis have endothelial dysfunction, measured as blunted response to endothelial vasodilator agents such as bradykinin and acetylcholine.3,4 Moreover, impaired endothelium-dependent vasodilation may be a marker of adverse prognosis in patients with risk factors without atherosclerosis,5,6 in stable ischemic heart disease,6 and in acute coronary syndromes.7
Clinical Perspective p 875
Insulin resistance, defined as impaired insulin-mediated glucose disposal, may also play a major role in the development of atherosclerosis.8 Decreased insulin sensitivity is often associated with 1 or more of the components of the metabolic syndrome, a cluster of independent cardiovascular risk factors comprising abdominal obesity, atherogenic dyslipidemia, increased blood pressure, and a proinflammatory, prothrombotic state.9 Insulin resistance also exerts significant effects on vascular homeostasis and is associated with blunted endothelium-dependent vasodilation.10
The peroxisome proliferator-activated receptor- (PPAR) is a nuclear receptor that has recently emerged as a pivotal intracellular controller of systemic and vascular processes, including inflammation and atherosclerosis.11 PPAR is expressed in all major cell types involved in the initiation and evolution of the atherosclerotic plaque, including endothelial cells, vascular smooth muscle cells, macrophages, and T lymphocytes, where it may exert antiinflammatory and potentially antiatherogenic effects.11 PPAR is also expressed in adipose, liver, and skeletal muscle tissues, and its stimulation results in improved insulin-mediated glucose disposal.12 In fact, the thiazolidinediones enhance insulin sensitivity by activating PPAR receptors13 and are currently in use for the treatment of type 2 diabetes mellitus. In agreement with the positive vascular effects of PPAR stimulation, thiazolidinediones improve endothelial function, independent of glucose control, in patients with type 2 diabetes.14 However, whether this beneficial effect extends to nondiabetic patients without insulin resistance is unknown.
Therefore, the present study was undertaken to test the hypothesis that PPAR activation with pioglitazone improves endothelial function in nondiabetic patients with cardiovascular risk factors. We also sought to determine whether any changes in endothelium-dependent vascular relaxation induced by pioglitazone are related to changes in insulin sensitivity and in C-reactive protein (CRP) levels, a marker of systemic inflammation.
Methods
Study Sample
Patients aged between 18 and 70 years with a diagnosis of essential hypertension (systolic blood pressure 140 mm Hg or diastolic blood pressure 90 mm Hg) or hypercholesterolemia (total cholesterol 6.18 mmol/L [240 mg/dL]) were considered candidates for the study. Each subject underwent screening including a detailed medical history and physical examination, ECG, complete blood count, and chemistry panel including liver function tests, total cholesterol, and insulin levels. Subjects on statins were asked to stop taking them at least 1 month before the screening visit, and these drugs were not allowed throughout the study. Women on hormonal contraception or on hormone replacement therapy were not included in the study. In fertile women, pregnancy was excluded by STAT pregnancy test at the time of screening and before the performance of each vascular function study. Antihypertensive medications were withdrawn under monitoring for 2 weeks before the screening visit and 2 weeks before each of the vascular function studies. Patients were excluded if they had both hypertension and hypercholesterolemia, a diagnosis or clinical evidence of diabetes mellitus, or any disease that could affect participation in the study or its results.
Study Design
The study had a double-blind, placebo-controlled, crossover design. The main end point was the comparison of forearm vascular responses to local infusion of bradykinin, acetylcholine, and sodium nitroprusside after each treatment period. Secondary end points were comparison of laboratory parameters including, but not limited to, plasma levels of glucose, insulin, lipids, and CRP after each treatment period.
After determination of eligibility at screening, participants were classified by the research coordinator (L.A.M.) into 1 of 4 subgroups on the basis of diagnosis (ie, hypertension or hypercholesterolemia) and insulin sensitivity status (ie, insulin sensitive or insulin resistant). To ensure equal numbers in each of the 4 subgroups, classification of patients into a subgroup after screening was stopped once 20 evaluable patients previously classified in that subgroup had completed the study. The randomization table for this study was developed by the Research Pharmacy of the Washington Hospital Center using random number patterns from the Web site www.randomizer.com. The sequences of random numbers were blocked in groups of 10 patients with runs of no more than 4 consecutive occurrences of the same treatment sequence. The randomization schedule was kept in the Research Pharmacy area and was available only to unblinded pharmacy personnel until after the database was locked. At that time, the unblinded patient treatment information was made available to the investigators.
Insulin sensitivity was determined at the time of screening with the use of the previously validated Quantitative Insulin Sensitivity Check Index (QUICKI) method.15 On the basis of interpolation of available data, subjects were considered insulin sensitive if their QUICKI index was >0.3400 and insulin resistant if their QUICKI index was 0.3400.15
Eligible patients underwent two 8-week treatment phases of either pioglitazone 45 mg daily or equivalent placebo. Endothelial function studies and blood sample drawing for hematologic and biochemical assays were performed at the end of each 8-week treatment period. In fertile women, an effort was made to perform the vascular studies on the same day of the menstrual cycle. The study protocol was approved by the MedStar Research Institute Investigational Review Board (protocol No. 2001-318), and all patients gave written informed consent.
Endothelial Function Studies
Endothelial function was tested as previously described in detail.3,4 Briefly, forearm blood flow was measured by strain-gauge plethysmography after intra-arterial infusion of increasing doses of the endothelium-dependent vasodilators bradykinin (infusion rates: 100, 200, and 400 ng/min) and acetylcholine (infusion rates: 7.5, 15, and 30 μg/min)16,17 and of the endothelium-independent vasodilator sodium nitroprusside (infusion rates: 0.8, 1.6, and 3.2 μg/min).18 Drugs sequence was randomized to avoid bias related to the order of infusion.
Statistical Analyses
Sample size calculation was based on differences between the values of maximum forearm blood flow induced by each of 3 drugs (bradykinin, acetylcholine, and sodium nitroprusside) in each of the 4 groups of participants (insulin-sensitive and insulin-resistant hypercholesterolemics and insulin-sensitive and insulin-resistant hypertensives) measured after 8 weeks on a placebo and after 8 weeks on pioglitazone. Mean values and their SEs of maximum forearm blood flow induced by acetylcholine in 12 normal subjects and 12 hypercholesterolemic patients were previously recorded in our laboratory as 17.5±7.7 and 8.0±5.1, respectively19; these translated into SD estimates of 26.7 and 17.7, respectively. Corresponding values for 14 hypertensive patients were 8.9±5,20 producing a SD estimate of 18.7. On the basis of these data, several combinations of sample size, power, and detectable differences in maximum forearm blood flow were derived at an overall type I error level of 0.05 for differences on each of 3 stimulating drugs for hypercholesterolemic and hypertensive patients with the use of 2-sided paired t tests. A sample of 20 subjects in each of the 4 prespecified subgroups was calculated to be necessary to detect a 22% difference in forearm blood flow values with 90% power and =0.05. With anticipation of a 33% dropout rate, samples of up to 1.5 times those indicated (ie, 30 subjects) were enrolled to yield 20 evaluable participants in each subgroup.
All group data are expressed as mean±SEM. Within-group analyses were performed by paired t test and 1-way and 2-way ANOVA for repeated measures. Group comparisons were performed by unpaired t test and 2-way ANOVA. Correlation analyses were performed with use of the Pearson correlation coefficient. To identify predictors of changes in the vascular response in a multivariable setting, multiple linear regression analysis was used. All calculated probability values are 2 tailed, and a probability value <0.05 was considered to indicate statistical significance.
Results
Study Sample
Data on recruitment, eligibility, randomization, and withdrawals are detailed in Figure 1. In brief, patient recruitment was terminated after a total of 101 patients had been enrolled, at which point evaluable data were available for 80 patients (20 in each of the 4 subgroups; dropout rate, 21%). The first patient was screened on February 28, 2002, and the last patient completed the study on June 29, 2004. Of the 21 patients who did not complete the study, 1 was withdrawn for a serious adverse event. This occurred in a 52-year-old hypertensive woman who suffered a brief episode of acute decompensated heart failure while on pioglitazone. In 1 of the 80 evaluable patients, the brachial artery could not be cannulated at the end of the second treatment period, and therefore endothelial function studies could not be performed. Thus, the results from 79 patients are included in this report. The baseline characteristics of study participants are reported in Table 1. Of the 32 women, 23 (72%) were postmenopausal, and 9 (28%) reported regular menstrual cycles. Among the hypertensive patients, 7 (18%) were not on chronic antihypertensive treatment, 21 were on monotherapy (7 on diuretics, 6 on ACE inhibitors, 5 on dihydropyridine calcium channel blockers, 1 on angiotensin receptor blockers, 1 on -blockers, and 1 on -blockers), 8 on dual therapy, 2 on therapy with 3 agents, and 1 on therapy with 4 drugs.
Effects of Pioglitazone on Body Weight and on Laboratory Parameters
As shown in Table 2, pioglitazone treatment did not modify body weight or body mass index (BMI) compared with placebo. Compared with placebo, pioglitazone induced an increase in HDL cholesterol (8.2%) and a decrease in triglycerides (–15.1%) and free fatty acids (–14%). Small changes in total cholesterol (–3.1%) and LDL cholesterol levels (–5.5%) were not statistically significant. Pioglitazone treatment caused a slight but statistically significant decline in hematocrit and hemoglobin values (–3.8% and –3.3%, respectively) and in Na+ concentrations (–0.7%). However, blood urea nitrogen and creatinine levels were not significantly affected (Table 2). Throughout the study, no significant elevations of liver function tests were observed in any of the participants either during active treatment or during placebo treatment.
Treatment with pioglitazone caused a significant decrease of plasma insulin (–22.9%) and a significant improvement in insulin sensitivity, as assessed by QUICKI (3.7%). However, fasting plasma glucose was not affected (–2.1%) (Table 2). The effect of pioglitazone on insulin and QUICKI was significant in both insulin-sensitive and insulin-resistant subjects (Table 3). To determine whether the effects of pioglitazone on these parameters differ depending on baseline insulin sensitivity, a test of interaction for differential treatment effect between the insulin sensitivity subgroups was performed. The results of the models did not show any significant interaction (P=0.75 for glucose, P=0.14 for insulin, and P=0.99 for QUICKI by 2-way ANOVA for repeated measures), indicating that the effect of treatment on these variables is independent of baseline insulin sensitivity status.
Effects of Pioglitazone on Vascular Responses
Forearm blood flow increases during bradykinin were significantly higher after pioglitazone treatment than after placebo (Figure 2, left). Similarly, there was a trend toward a higher dose-dependent vasodilator effect of acetylcholine after pioglitazone treatment than after placebo, although this difference did not reach statistical significance (Figure 2, center). In contrast, the vasodilator response to sodium nitroprusside after pioglitazone was not significantly different compared with that obtained after placebo (Figure 2, right).
Interaction of Effects of Pioglitazone and Insulin Sensitivity on the Vascular Responses to Bradykinin, Acetylcholine, and Sodium Nitroprusside
We sought to determine whether the effect of pioglitazone on vascular responses was related to the presence of insulin resistance at baseline or to the effect of the drug on insulin sensitivity. When the responses to bradykinin, acetylcholine, and sodium nitroprusside were compared between insulin-sensitive and insulin-resistant patients, no significant differences were observed at the end of placebo (P=0.29, P=0.135, and P=0.120 for bradykinin, acetylcholine, and sodium nitroprusside, respectively, by 2-way ANOVA for repeated measures) or pioglitazone treatment (P=0.80, P=0.646, and P=0.161 for bradykinin, acetylcholine, and sodium nitroprusside, respectively, by 2-way ANOVA for repeated measures) (Table 4). In the overall study sample, no relationship was observed between the changes in insulin sensitivity and the effects on the vascular responses to bradykinin and acetylcholine induced by pioglitazone (r=–0.004, P=0.97, and r=–0.01, P=0.91, respectively, by Pearson correlation coefficient).
Interaction of Effects of Pioglitazone and Diagnosis on Vascular Responses to Bradykinin, Acetylcholine, and Sodium Nitroprusside
To investigate whether pioglitazone exerts differential effects on vascular reactivity depending on the underlying diagnosis, we performed separate subgroup analyses in patients with hypertension and in patients with hypercholesterolemia (Table 5). A significant improvement in the response to bradykinin after pioglitazone was observed in hypercholesterolemic (P=0.012 by 2-way ANOVA for repeated measures) but not in hypertensive patients (P=0.41 by 2-way ANOVA for repeated measures). No significant changes were observed in the response to acetylcholine and sodium nitroprusside after pioglitazone treatment in either subgroup when analyzed separately.
Interaction of Effects of Pioglitazone and BMI on Vascular Responses to Bradykinin, Acetylcholine, and Sodium Nitroprusside
In our sample, the average BMI was 31 kg/m2, indicating a high prevalence of obesity among the participants. Of the 79 patients, 36 (46%) had a BMI 30 kg/m2, and 43 (54%) had a BMI <30 kg/m2. To determine whether the changes in vascular reactivity secondary to treatment with pioglitazone differed between the obese and the nonobese patients, we compared the differences in blood flow responses to bradykinin, acetylcholine, and sodium nitroprusside induced by pioglitazone treatment between participants with a BMI 30 kg/m2 and those with a BMI <30 kg/m2. When compared by 2-way ANOVA for repeated measures, no significant differences in the changes of the response to bradykinin (P=0.11), acetylcholine (P=0.27), and sodium nitroprusside (P=0.76) were observed between the 2 groups.
Baseline Insulin Sensitivity, Effects of Pioglitazone, and Their Interaction on CRP Levels
In 2 patients, CRP measurements were >20 mg/L. These results were likely due to systemic inflammation not ascribable to a cardiovascular source and were therefore excluded from the analyses. In the overall sample, treatment with pioglitazone caused a significant 28.6% reduction of CRP levels compared with placebo (from 3.57±0.45 to 2.55±0.33 mg/L; P=0.0008 by 2-way ANOVA for repeated measures). To determine whether pioglitazone affected CRP levels depending on insulin sensitivity, we compared CRP values between insulin-sensitive and insulin-resistant subjects after placebo and after active treatment. With placebo, CRP levels were significantly higher in the insulin-resistant than in the insulin-sensitive group (4.60±0.73 versus 2.58±0.48 mg/L, respectively; P=0.02 by 2-way ANOVA for repeated measures). Compared with placebo, pioglitazone induced a significant decrease of CRP levels only in insulin-resistant subjects (from 4.60±0.73 to 2.89±0.52 mg/L; P=0.0001 by 2-way ANOVA for repeated measures). Consequently, CRP levels were similar between the 2 groups at the end of active treatment with pioglitazone (2.89±0.52 versus 2.22±0.41 mg/L, respectively; P=0.38 by 2-way ANOVA for repeated measures) (Figure 3).
Predictors of Improved Response to Bradykinin After Active Treatment
To determine the variables associated with improved vascular reactivity to bradykinin after pioglitazone, we performed correlation and multivariable analyses between the changes in the response to the highest dose of bradykinin and the variations in glucose, insulin, QUICKI, lipid profile, and CRP. In univariate analysis, the changes in total cholesterol and in free fatty acids were the only significant predictors of improved endothelial reactivity with pioglitazone (Table 6). However, in a multivariable model including variations in total cholesterol, free fatty acids, QUICKI, and CRP, only total cholesterol maintained its significant relationship with the changes in bradykinin response (r=0.376, r2=0.142, P=0.032).
Discussion
The results of this investigation indicate that PPAR stimulation with pioglitazone in nondiabetic patients with cardiovascular risk factors leads to enhancement of insulin sensitivity, decrease in markers of systemic inflammation, and improvement in endothelial vasodilator function. However, these effects do not appear to be closely related, suggesting that pioglitazone may have beneficial properties independent of its effect on insulin sensitivity.
Several lines of evidence indicate that PPAR stimulation with thiazolidinediones may exert beneficial vascular effects independent of glycemic control. For example, in vitro studies have demonstrated that PPAR is expressed in human endothelial cells,21 in which it may enhance nitric oxide release22 and modulate chemokine synthesis,23 and in vascular smooth muscle cells, where it may downregulate the expression of angiotensin II type 1 receptors,24 inhibit cell migration,25 and reduce the release of matrix-degrading enzymes.25 In accord with this evidence, in vivo studies in patients with type 2 diabetes have shown that treatment with troglitazone may improve brachial artery flow-mediated dilation without affecting the response to nitroglycerin26 and that rosiglitazone, but not neglitinide, increases endothelium-dependent response to acetylcholine,14 indicating that the impact of thiazolidinediones on endothelial function is at least partly independent of their blood glucose–lowering effects.
Insulin resistance is an emerging metabolic risk factor that is associated with cardiovascular disease risk.9 Furthermore, insulin resistance is strongly associated with the presence of other metabolic risk factors, such as atherogenic dyslipidemia, glucose intolerance, and proinflammatory state, as epitomized by the characteristic cluster of the metabolic syndrome.9 Our and other laboratories have shown that reduced insulin sensitivity in the absence of other risk factors is associated with endothelial dysfunction.10 In addition, we have recently demonstrated that the accompanying hyperinsulinemia may impair endothelium-dependent vasodilation independent of insulin sensitivity and lipid profile.10 Because thiazolidinediones increase insulin sensitivity,12 an improvement in endothelial function could potentially be secondary to the amelioration of insulin resistance. Our results show that treatment with pioglitazone significantly enhanced insulin sensitivity in both insulin-sensitive and insulin-resistant subjects. However, the improvement in endothelial function was similar between these 2 groups and did not correlate with the change in the calculated insulin sensitivity index.
These observations have important implications with regard to the mechanism of action and the potential beneficial effects of pioglitazone. First, the lack of relation between the effects on endothelial function and on insulin sensitivity suggests that PPAR agonism may lead to activation of different signaling pathways, leading to multiple and apparently unrelated metabolic and vascular actions. Second, these findings expand the interest in further investigating the potential beneficial effects of PPAR stimulation with pioglitazone to patients at risk of cardiovascular disease who do not have insulin resistance. In recent years, the role of inflammation in the pathogenesis of endothelial dysfunction has been increasingly recognized.27 Because PPAR activation inhibits the expression of inflammatory genes,28 the effect of thiazolidinediones on endothelial function could potentially be mediated by an improvement in inflammation. In our study, treatment with pioglitazone significantly decreased CRP levels. However, the improved endothelial reactivity did not correlate with the variations in circulating CRP, suggesting that the effects of pioglitazone on endothelium-dependent vasodilation are independent of the reduction of this inflammatory marker.
PPAR is a key controller of adipocyte differentiation and fatty acid uptake and storage12 and modulates the expression of genes involved in fatty acids and triglycerides synthesis.29 In our study, treatment with pioglitazone exerted multiple effects on lipid metabolism, with small, non–statistically significant decreases in total cholesterol and LDL cholesterol, significant reduction of free fatty acids and triglycerides, and significant increase of HDL cholesterol. Given the impact of lipoprotein on endothelial function,4 changes in lipid profile could underlie the improvement in endothelium-dependent dilation observed with pioglitazone. Interestingly, in our study the variation in total cholesterol was the only significant predictor of improved endothelial reactivity in both univariate and multivariable analyses. Because the changes in total cholesterol concentration did not reach statistical significance, it is possible that this correlation reflects a sum of the effects of pioglitazone on the lipoprotein profile. However, this association was relatively weak, and other factors may contribute to the effects of pioglitazone on endothelial function.
To further understand whether the improvement in endothelial function secondary to treatment with pioglitazone was related to the underlying dyslipidemia, we performed separate analyses in the hypercholesterolemic and in the hypertensive subgroups. A significant improvement in the response to bradykinin after pioglitazone was observed only in patients with hypercholesterolemia, suggesting that the effects of pioglitazone on endothelial function may be mechanistically linked to the presence of dyslipidemia. However, our study was not specifically designed to compare the effect between patients with different diagnoses; therefore, we cannot conclude that our observations are truly limited to patients with hypercholesterolemia.
Given the prevalence of overweight and obese insulin-resistant patients in our sample, it is possible that some of the enrolled participants may have had a subclinical form of diabetes mellitus (ie, without fasting hyperglycemia) that could have been detected by the use of additional laboratory tests. However, our data indicate that the improvement in endothelial function was not related to the variations in glucose, insulin, or the QUICKI assessment of insulin sensitivity. Thus, we do not believe that the potential presence of subclinical diabetes in our sample may have influenced the results of our study or their interpretation. Moreover, when we compared the effects of pioglitazone between obese (ie, with a BMI 30) and nonobese (ie, BMI <30) participants, no significant differences were noted in the effects of active treatment on vasodilator responses between the 2 groups, further indicating that pioglitazone may exert vascular effects independent of body weight and glucose metabolism.
Pioglitazone treatment was well tolerated, and most of the adverse events were mild and considered unrelated to participation in the study. However, 1 hypertensive patient with mild left ventricular hypertrophy and diastolic dysfunction developed an episode of heart failure while on active treatment. We believe that this occurrence may have been related to the known potential for sodium and water retention associated with thiazolidinedione treatment. In fact, compared with placebo, pioglitazone induced a mild but statistically significant decrease in Na+ concentration and in hematocrit and hemoglobin levels, suggesting a hemodiluting effect. Hence, as reported in a recent consensus statement from the American Heart Association and the American Diabetes Association, caution should be exerted in the use of thiazolidinediones in patients with cardiovascular conditions potentially sensitive to fluid overload.30
Taken together, the findings of our study indicate that PPAR activation with pioglitazone may exert multiple beneficial vascular effects in normoglycemic patients with major atherosclerotic risk factors. Pioglitazone treatment improved endothelial function and exerted significant positive actions on insulin sensitivity, plasma insulin, markers of inflammation, and lipid profile, which may independently contribute to the overall cardiovascular risk.9,10 Several investigations have confirmed the prognostic value of endothelial function,5–7 suggesting that therapies that reverse endothelial dysfunction may result in a decreased cardiovascular risk. Our finding may serve as the basis for the development of large-scale clinical trials designed to determine the effects of thiazolidinediones or other PPAR agonists on cardiovascular outcomes in nondiabetic patients with major risk factors.
Acknowledgments
Disclosures
This study was funded by an investigator-initiated grant from Takeda Pharmaceuticals North America, Inc. The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
References
Vane JR, nggrd EE, Botting RM. Regulatory functions of the vascular endothelium. N Engl J Med. 1990; 323: 27–36.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Panza JA, Garcia C, Kilcoyne CM, Quyyumi AA, Cannon RO. Impaired endothelium-dependent vasodilation in patients with essential hypertension: evidence that nitric oxide abnormality is not localized to a single signal transduction pathway. Circulation. 1995; 91: 1732–1738.
Casino PR, Kilcoyne CM, Quyyumi AA, Hoeg JM, Panza JA. The role of nitric oxide in endothelium-dependent vasodilation of hypercholesterolemic patients. Circulation. 1993; 88: 2541–2547.
Perticone F, Ceravolo R, Pujia A, Ventura G, Iacopino S, Scozzafava A, Ferraro A, Chello M, Mastroroberto P, Verdecchia P, Schillaci G. Prognostic significance of endothelial dysfunction in hypertensive patients. Circulation. 2001; 104: 191–196.
Halcox JPJ, Schenke WH, Zalos G, Mincemoyer R, Prasad A, Waclawiw MA, Nour KR, Quyyumi AA. Prognostic value of coronary vascular endothelial dysfunction. Circulation. 2002; 106: 653–658.
Fichtlscherer S, Breuer S, Zeiher AM. Prognostic value of systemic endothelial dysfunction in patients with acute coronary syndromes. Further evidence for the existence of the "vulnerable" patient. Circulation. 2004; 110: 1926–1932.
Haffner SM. Epidemiology of insulin resistance and its relation to coronary artery disease. Am J Cardiol. 1999; 84: 11J–14J.
Grundy SM, Brewer HB, Cleeman JI, Smith SC, Lenfant C. Definition of metabolic syndrome: report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation. 2004; 109: 433–438.
Campia U, Sullivan G, Bryant MB, Waclawiw MA, Quon MJ, Panza JA. Insulin impairs endothelium-dependent vasodilation independent of insulin sensitivity or lipid profile. Am J Physiol. 2004; 286: H76–H82.
Marx N, Duez H, Fruchart JC, Staels B. Peroxisome proliferator activated receptor and atherogenesis: regulator of gene expression in vascular cells. Circ Res. 2004; 94: 1168–1178.
Yki-Jrvinen H. Thiazolidinediones. N Engl J Med. 2004; 351: 1106–1118.
Lehman JM, Moore LB, Smith-Oliver TA, Wilkison WO, Willson TM, Kliewer SA. An antidiabetic thiazolidinedione is a high affinity ligand for peroxisome proliferator-activated receptor (PPAR). J Biol Chem. 1995; 270: 12953–12956.
Pistrosch F, Passauer J, Fischer S, Fuecker K, Hanefeld M, Gross P. In type 2 diabetes, rosiglitazone therapy for insulin resistance ameliorates endothelial dysfunction independent of glucose control. Diabetes Care. 2004; 27: 484–490.
Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, Quon MJ. Quantitative Insulin-sensitivity Check Index (QUICKI): a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab. 2000; 85: 2402–2410.
Kiowski W, Linder L, Kleinbloesem C, van Brummelen P, Buhler FR. Blood pressure control by the renin-angiotensin system in normotensive subjects: assessment by angiotensin converting enzyme and renin inhibition. Circulation. 1992; 85: 1–8.
Furchgott RF, Zawadski JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle to acetylcholine. Nature. 1980; 288: 373–376.
Bohme E, Graf H, Schultz G. Effects of sodium nitroprusside and other smooth muscle relaxants on cyclic-GMP formation in smooth muscles and platelets. Adv Cycl Nucl Res. 1978; 9: 131–143.
Gilligan DM, Guetta V, Panza JA, Garcia CE, Quyyumi AA, Cannon RO III. Selective loss of microvascular endothelial function in human hypercholesterolemia. Circulation. 1994; 90: 35–41.
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: 1468–1474.
Delerive P, Martin-Nizard F, Chinetti G, Trottein F, Fruchart JC, Najib J, Duriez P, Staels B. Peroxisome proliferator-activated receptor activators inhibit thrombin-induced endothelin-1 production in human vascular endothelial cells by inhibiting the activator protein-1 signaling pathway. Circ Res. 1999; 85: 394–402.
Calnek DS, Mazzella L, Roser S, Roman J, Hart MC. Peroxisome proliferator-activated receptor ligands increase release of nitric oxide from endothelial cells. Arterioscler Thromb Vasc Biol. 2003; 23: 52–57.
Marx N, Mach F, Sauty A, Sarafi M, Libby P, Plutzky J, Luster AD. PPAR activators inhibit interferon--induced expression of the T cell active CXC chemokines IP-10, Mig, and I-TAC in human endothelial cells. J Immunol. 2000; 164: 6503–6508.
Takeda K, Ichiki T, Tokunou T, Funakoshi Y, Iino N, Hirano K, Kanaide H, Takeshita A. Peroxisome proliferator-activated receptor activators downregulate angiotensin II type 1 receptor in vascular smooth muscle cells. Circulation. 2000; 102: 1834–1839.
Marx N, Schnbeck U, Lazar MA, Libby P, Plutzky J. Peroxisome proliferator activated receptor activators inhibit gene expression and migration in human vascular smooth muscle cells. Circ Res. 1998; 83: 1097–1103.
Murakami T, Mizuno S, Ohsato K, Moriuchi I, Arai Y, Nio Y, Kaku B, Takahashi Y, Ohnaka M. Effects of troglitazone on frequency of coronary vasospastic-induced angina pectoris in patients with diabetes mellitus. Am J Cardiol. 1999; 84: 92–94.
Fichtlscherer S, Rosenberger G, Walter DH, Breuer S, Dimmeler S, Zeiher AM. Elevated C-reactive protein levels and impaired endothelial vasoreactivity in patients with coronary artery disease. Circulation. 2000; 102: 1000–1006.
Barbier O, Torra IP, Duguay Y, Blanquart C, Fruchart JC, Glineur C, Staels B. Pleiotropic actions of peroxisome proliferator-activated receptors in lipid metabolism and atherosclerosis. Arterioscler Thromb Vasc Biol. 2002; 22: 717–726.
Chinetti G, Fruchard J-C, Staels B. Peroxisome proliferator-activated receptors (PPARs): nuclear receptor at the crossroads between lipid metabolism and inflammation. Inflammation Res. 2000; 49: 497–505.
Nesto RW, Bell D, Bonow RO, Fonseca V, Grundy SM, Horton ES, Le Winter M, Porte D, Semenkovich CF, Smith S, Young LH, Kahn R. Thiazolidinedione use, fluid retention, and congestive heart failure: a consensus statement from the American Heart Association and American Diabetes Association. Circulation. 2003; 108: 2941–2948.
Dyslipidemia Prevalence, Treatment, and Control in the Multi-Ethnic Study of Atherosclerosis (MESA)
Mitral and Aortic Annular Calcification Are Highly Associated With Systemic Calcified Atherosclerosis
Relationship Between C-Reactive Protein and Subclinical Atherosclerosis
Disruption of the Cathepsin K Gene Reduces Atherosclerosis Progression and Induces Plaque Fibrosis but Accelerates Macrophage Foam Cell Formation
Cyclooxygenase-1 Deficiency in Bone Marrow Cells Increases Early Atherosclerosis in Apolipoprotein E– and Low-Density Lipoprotein Receptor–Null Mice
Effect of Intensive Versus Standard Lipid-Lowering Treatment With Atorvastatin on the Progression of Calcified Coronary Atherosclerosis Over 12 Months
Sevelamer Prevents Uremia-Enhanced Atherosclerosis Progression in Apolipoprotein E–Deficient Mice
Deletion of Angiotensin II Type 2 Receptor Exaggerated Atherosclerosis in Apolipoprotein E–Null Mice
Differential Influence of Chemokine Receptors CCR2 and CXCR3 in Development of Atherosclerosis In Vivo
Compromised LCAT Function Is Associated With Increased Atherosclerosis