Coronary Circulatory Dysfunction in Insulin Resistance, Impaired Glucose Tolerance, and Type 2 Diabetes Mellitus
http://www.100md.com
《循环学杂志》
the Departments of Molecular and Medical Pharmacology and Medicine (J.O.P., M.H.-P., A.D.F., T.H.S., J.W.S., H.R.S.)
Division of Endocrinology, Diabetes and Hypertension (M.J.Q., W.A.H.)
David Geffen School of Medicine at UCLA, Los Angeles, Calif.
Abstract
Background— Abnormal coronary endothelial reactivity has been demonstrated in diabetes and is associated with an increased rate of cardiovascular events. Our objectives were to investigate the presence of functional coronary circulatory abnormalities over the full spectrum of insulin resistance and to determine whether these would differ in severity with more advanced states of insulin resistance.
Methods and Results— Myocardial blood flow (MBF) was measured with positron emission tomography and 13N-ammonia to characterize coronary circulatory function in states of insulin resistance without carbohydrate intolerance (IR), impaired glucose tolerance (IGT), and normotensive and hypertensive type 2 diabetes mellitus (DM) compared with insulin-sensitive (IS) individuals. Indices of coronary function were total vasodilator capacity (mostly vascular smooth muscle–mediated) during pharmacological vasodilation and the nitric oxide–mediated, endothelium-dependent vasomotion in response to cold pressor testing. Total vasodilator capacity was similar in normoglycemic individuals (IS, IR, and IGT), whereas it was significantly decreased in normotensive (–17%) and hypertensive (–34%) DM patients. Compared with IS, endothelium-dependent coronary vasomotion was significantly diminished in IR (–56%), as well as in IGT and normotensive and hypertensive diabetic patients (–85%, –91%, and –120%, respectively).
Conclusions— Progressively worsening functional coronary circulatory abnormalities of nitric oxide–mediated, endothelium-dependent vasomotion occur with increasing severity of insulin-resistance and carbohydrate intolerance. Attenuated total vasodilator capacity accompanies the more clinically evident metabolic abnormalities in diabetes.
Key Words: blood flow ; diabetes mellitus ; endothelium ; hypertension
Introduction
Abnormal coronary endothelial reactivity represents one of the earliest manifestations of vascular disease and is a key element in the development of atherosclerosis.1 It exists not only in diabetes and in groups at high risk for type 2 diabetes mellitus, such as impaired glucose tolerance (IGT), but also in insulin-resistant individuals who do not have significant metabolic abnormalities.2–5 The cardiovascular mortality rate in IGT is twice that of normal glucose tolerance, increases 2- to 3-fold in type 2 diabetes mellitus,6 and rises further when diabetes is associated with hypertension.7 Because the high mortality cannot be fully explained by conventional cardiovascular risk factors, it has been attributed to factors related to insulin resistance and its potentially adverse effects on endothelial function. A key issue is whether and how endothelial dysfunction and vascular injury progress as insulin resistance progresses to type 2 diabetes mellitus. Positron emission tomography (PET) affords noninvasive measurements of myocardial blood flow (MBF).8,9 MBF responses to vascular smooth muscle cell (VSMC) relaxing agents like dipyridamole or adenosine yield information on total coronary vasodilator function. MBF responses to sympathetic stimulation with cold pressor testing (CPT) provide information on endothelium-related coronary vasomotion. We hypothesized that abnormalities in coronary circulatory function would parallel insulin resistance severity and carbohydrate intolerance in the progression to type 2 diabetes mellitus. To address this issue, we used PET to compare coronary circulatory function in Mexican-American, nonsmoking individuals who were insulin-sensitive (IS) or insulin-resistant (IR), or who had IGT or type 2 diabetes mellitus (DM) with and without hypertension.
Methods
Study Population
One-hundred seventy-four Mexican Americans underwent screening that included medical history, physical examination, blood pressure measurements, and an ECG; fasting blood collection for routine testing (blood cell count, chemistry, and lipid panel) and routine urinalysis were performed by the Medical Center Clinical Pathology Laboratory. All individuals without a history of diabetes underwent a 2-hour 75-g oral glucose tolerance test with measurement of glucose (glucose oxidase technique; YSI, Inc) and insulin levels (Human Insulin-Specific RIA Kit; Linco Research, Inc). Glucose tolerance was categorized according to American Diabetes Association criteria4 (IGT, fasting glucose 6.11 to 7.00 mmol/L and 2-hour plasma glucose 7.77 to 11.10 mmol/L; DM, fasting plasma glucose 7.00 mmol/L or 2-hour glucose 11.10 mmol/L). Nondiabetic individuals underwent peripheral insulin sensitivity measurement by a modified hyperinsulinemic-euglycemic clamp.5 Briefly, a priming dose of human insulin (Novolin; Novo Nordisk) was followed by insulin infusion (60 mU · m–2 · min–1) for 120 minutes to achieve plasma insulin levels 700 pmol/L. Blood was sampled every 5 minutes, and 20% glucose infusion rate (GINF) was adjusted to maintain euglycemia (5.00 to 5.55 mmol/L). The GINF (mg/kg of body weight per minute) over the last 30 minutes of the clamp reflected insulin action and defined IS (GINF 7.5 mg · kg–1 · min–1) and IR (GINF 4.0 mg · kg–1 · min–1) individuals.5 Of the 174 individuals, 14 declined to participate in the study. Forty individuals were excluded from the study (hypertension [n=3], hypercholesterolemia [n=9], or triglyceride levels 3.40 mmol/L [n=15] in the absence of diabetes; smoking [n=2]; females taking oral contraceptives [n=6]; diabetic patients taking hypoglycemic agents [n=4]; and hypertensive diabetic patients taking ;-blockers [n=1]). The remaining 120 individuals were assigned to 1 of 5 study groups: (1) IS group (n=19) without coronary risk factors and normal insulin sensitivity; (2) IR group (n=47), without coronary risk factors and normal carbohydrate tolerance; (3) IGT group (n=25); (4) DM group without hypertension (n=21); and (5) hypertensive group of diabetic individuals with hypertension (HTN group; n=8). Diabetic patients were subgrouped on the basis of the presence (blood pressure >135/80 mm Hg) or absence of hypertension, which is known to adversely affect coronary circulation.10 Comparisons across study groups were performed with the homeostasis model assessment (HOMA) index of insulin resistance11 (fasting plasma glucose [mmol/L]xfasting plasma insulin [pmol/L]/162), which has been found to correlate with glucose clamp measurement in nondiabetic, diabetic, or hypertensive populations.12 Of the 76 women in the study, 6 were postmenopausal (cessation of menses 1year) and were also diabetic. The time since diagnosis of diabetes averaged 2.1±3.2 years (range 0 to 11 years). None of the participants were taking antihypertensive, antidiabetic, or lipid-lowering medications. The study was approved by the UCLA Institutional Review Board, and written informed consent was obtained from all participants.
Evaluation and Measurement of MBF
MBF was measured with 13N-ammonia and PET (ECAT EXACT HR/HR+, CTI/Siemens). Beginning with the intravenous injection of 13N-ammonia (555 to 740 MBq), serial transaxial images were acquired for 19 minutes, as described previously.13 Reoriented short- and long-axis myocardial slices of the last 15-minute transaxial image and the corresponding polar maps were submitted to visual analysis of the relative MBF distribution. Regions of interest assigned on the last 15-minute image to coronary artery territories and the left ventricular blood pool were copied to the serial images acquired during the first 2 minutes to generate time-activity curves and to estimate MBF as described previously.13 Because no significant MBF differences existed between coronary territories, regional MBFs were averaged.
Participants refrained from drinking caffeine-containing substances for 24 hours before study and had been fasting for 6 hours. MBF was measured at rest, then during CPT, and finally during pharmacological vasodilation with adenosine or dipyridamole, separated by 35 to 45 minutes. CPT was performed by hand immersion in ice water for 2 minutes, with administration of 13N-ammonia and imaging at 45 seconds after onset of the test. Image acquisition and 13N-ammonia administration began 3 minutes after completion of the 4-minute infusion of dipyridamole (0.54 mg/kg) or 3 minutes after the 6-minute infusion of adenosine (140 μg · kg–1 · min–1) was begun. Studies in nondiabetic individuals used dipyridamole, whereas adenosine was used in 11 of 21 normotensive diabetic participants and in all hypertensive diabetic participants. Heart rate, blood pressure, and a 12-lead ECG were recorded at 1-minute intervals during each procedure. Heart rates and blood pressures were averaged for the first 2 minutes of data acquisition, and the rate-pressure product (RPP=heart ratexsystolic blood pressure) was derived as an index of cardiac work. Image artifacts due to subject motion, side effects to adenosine or dipyridamole, and poor image quality precluded image acquisition or analysis of 1 rest, 6 CPT, and 11 hyperemic studies.
Statistical Analysis
Comparisons across groups were performed with 1-way ANOVA with post hoc Scheffe tests; results are presented as mean±SD, unless otherwise indicated. Skewed variables (P>0.05 on skewness-kurtosis test of normality; fasting plasma glucose and insulin, GINF, HOMA, and triglycerides) were analyzed after logarithmic transformation. Because significant baseline intergroup differences existed, ANCOVA14 was used to compare mean MBF response after adjustment for mean differences in potentially confounding variables (covariates: age, gender, body mass index [BMI], and resting flow). Interaction terms (groupxcovariate) were negligible (P0.22); pairwise group differences were assessed with least significant difference tests. The association of MBF responses to CPT and vasodilation with metabolic and clinical variables was assessed with univariate (Spearman rank) correlation and multivariable linear modeling (stepwise forward regression of variables with P0.1 significance in univariate analysis). Significance was considered for 2-sided P0.05, and Stata software (version 8.2, Stata Corporation) was used for analyses.
Results
Study Population and Laboratory Findings
Hemodynamic and MBF Findings
Blood pressure at baseline was similar in the normotensive study groups but was elevated in the HTN group (Table 1). Both CPT and adenosine or dipyridamole infusion significantly raised blood pressures in all study groups with the exception of a decline during pharmacological vasodilation in the DM group. Nevertheless, systolic blood pressures in the HTN group remained higher than in the other groups during CPT and during pharmacological vasodilation. The increase in RPP from rest to CPT (RPP; defined in absolute units) was similar in all 5 study groups, whereas the average RPP during CPT was higher in the HTN group than in the remaining groups (Table 2).
At rest, MBF was higher in the HTN group than in the other study groups (P<0.01; Table 2), and a trend existed for higher values in the IR group than in the IS group (P=0.11). Multivariate analysis identified the RPP as the only independent determinant of MBF at rest. Accordingly, normalization of MBFs by myocardial work (MBF · RPP–1) abolished differences in flow estimates of the HTN group compared those in the other study groups and between the IR and IS groups, so that the observed intergroup differences in resting MBF were related to differences in cardiac work.
Pharmacological vasodilation significantly raised MBF in all study groups. The flow increase was similar across the groups, although hyperemic MBF in the 2 diabetes groups tended to be lower than in the other groups (Table 2). When hyperemic MBFs were combined for all diabetes patients (DM and hypertensive groups), they were significantly lower than in the normoglycemic IR groups (IR and IGT; P>0.001). Accordingly, the myocardial flow reserve as the ratio of hyperemic to rest MBF in the diabetic patients was reduced significantly compared with that of the IS group (Table 2). Because patients in both diabetes groups were older than individuals in the other study groups, hyperemic MBFs were further adjusted for age by ANCOVA. Age-adjusted hyperemic MBFs were highest in the IS group and tended to decline progressively from the normoglycemic to the hyperglycemic IR groups (Table 4). The adjusted mean hyperemic MBFs in the normotensive diabetes group were significantly lower than in the IGT group (P<0.05; Table 4) but higher than in the hypertensive diabetes group (P<0.01). Furthermore, on nonparametric testing for trends across ordered groups, the trend of a progressive decline of hyperemic MBFs from the IS to the hypertensive groups achieved statistical significance (P<0.002). Univariate analysis of possible associations between hyperemic MBFs and other clinical findings revealed significant inverse correlations with fasting plasma glucose and MBF at rest and a positive correlation with HDL cholesterol (Table 3). There was a trend to inversely correlate with age, LDL cholesterol, and systolic blood pressure. On multivariate analysis, however, only fasting plasma glucose and HDL cholesterol remained independent predictors of hyperemic MBFs.
CPT consistently raised MBF in IS individuals; by contrast, individual flow responses in the other groups varied and ranged from increases to no change or to decreases. Average values of MBF during CPT varied between groups, although there were no significant intergroup differences with the exception of a significant difference between the hypertensive diabetes and the IGT groups (Table 2). As observed previously in normal volunteers,1 the cold-induced increase in RPP was accompanied in IS individuals by a proportionate increase in MBF (r=0.52; P=0.028). As Table 2 indicates, RPP increased by 41±23%, whereas MBF increased by 42±13%. In hypertensive diabetic patients, however, the percent increase in RPP was only 21±24%, whereas MBFs remained unchanged (–2±12%).
Already increased values of RPP and MBF account at least in part for the attenuated responses when expressed in percentages of the rest RPP or MBF. To reduce a baseline-dependent bias, responses to CPT were estimated in absolute units as the difference in MBF or in RPP during CPT and at rest. In the IS group, RPP increased with CPT by RPP=2.7±1.4x103 mm Hg/min and was associated with a MBF=0.25±0.08 mL · min–1 · g–1 flow increase. In the IR group, RPP similarly increased by 2.6±1.8x103 mm Hg/min (P=0.99 versus IS), but MBF increased by only 0.09±0.17x103 mL · min–1 · g–1 (P<0.001 versus IS). Furthermore, in the hypertensive diabetic group, RPP rose by 2.1±2.7x103 mm Hg/min, which was not significantly different from the increase in the IS group, whereas MBF remained virtually unchanged (–0.02±0.11 mL · min–1 · g–1). Thus, despite comparable increases in RPP (and thus, increases in cardiac work), responses in MBF to cold were significantly diminished or absent in the 4 IR groups and significantly differed from those in the IS control group (Table 2).
On univariate analysis, MBF was found to depend on BMI (Table 3), and significant intergroup differences existed for age and gender. Accordingly, the flow difference values were adjusted for BMI, age, gender, and MBF at rest. They are listed in Table 4 and demonstrate significant intergroup differences (Figure 2). The adjusted mean MBF responses in the IR group were significantly lower than in the IS group (P<0.001) but higher than in the 2 diabetic groups (P<0.05). Furthermore, nonparametric tests for trend across ordered groups indicated that the progressive decline of the MBF response across the IR states was statistically significant (P=0.02). Possible associations of the flow response (MBF) to CPT with clinical and laboratory findings were explored with univariate and multivariate analysis. In addition to the correlations with BMI and MBF at rest as mentioned above, univariate analysis revealed statistically significant inverse correlations between MBF and fasting plasma glucose, insulin levels, and HOMA, as well as with triglyceride concentrations. Furthermore, MBF correlated positively with HDL cholesterol levels. On multivariate analysis, only HDL cholesterol and triglyceride levels were found to be independent predictors of MBF.
Discussion
In this study, coronary vasomotor function was compared in a single ethnic group, Mexican-Americans, who were IS or who had IR, IGT, DM, or DM with hypertension. Mexican-Americans have a high prevalence of insulin resistance, metabolic syndrome, and DM.15 The main findings of the study are as follows: (1) the presence of a functional abnormality of the coronary circulation across the spectrum of insulin resistance, and (2) progression of this functional alteration with carbohydrate intolerance. In states of insulin resistance that lead to diabetes, this alteration primarily involved the flow response to CPT, which is largely endothelium dependent. In the presence of overt type 2 DM, there was also an attenuation of the total vasodilator capacity that tended to worsen with hypertension. When adjusted for age, gender, BMI, and MBF at rest, hyperemic MBFs and MBF responses to CPT differed between IR groups, and the trend of progressively declining MBFs was found to be statistically significant. These data suggest that vascular injury occurs when insulin resistance is the only abnormality, even in the absence of traditional coronary risk factors, and progresses as carbohydrate intolerance appears. These results are important in view of the worldwide epidemic of insulin resistance and are of concern because of the relatively early stage of appearance of vascular injury in the cascade of insulin resistance.
Flow responses to CPT largely reflect endothelium-related coronary vasomotor function and correlate with invasively and noninvasively measured indices of nitric oxide (NO)–mediated changes in coronary flow.1,16–18 Sympathetic stimulation with CPT raises both heart rate and systolic blood pressure and, thus, the RRP, an index of myocardial work. In the normal coronary circulation, the increase in myocardial work is associated with a proportionate metabolically mediated increase in coronary flow as previously reported and confirmed in the IS individuals in the present study.1,18 In the IR individuals, however, despite comparable increases in RRP, MBF failed to increase or even declined with CPT. This attenuated flow response has been reported with PET in patients with type 2 DM.19–21 Similarly, intracoronary acetylcholine stimulation in type 2 DM patients produced decreases rather than increases in conduit vessel diameter, whereas CPT induced a decrease in epicardial coronary diameter compared with an increase seen in normal controls.18 Thus, diabetes is associated with impaired function of both the conduit and resistance vessels.
An important finding of the present investigation was that the alteration of coronary vasomotor function was not confined to type 2 DM but was also present in individuals with normoglycemic IR. Previous investigations in peripheral arteries provided some evidence for the presence of endothelial dysfunction at different stages of development of diabetes. Acetylcholine-stimulated peripheral artery dilation was shown to be diminished in obese individuals with and without IGT, and both skin blood flow and brachial artery responses were diminished in relatives of type 2 DM patients.22–25 These studies suggested an impairment of the endothelium-related function of both the conduit and resistance vessels of the peripheral vasculature in IR that was similar to that described for the coronary vasculature in DM. However, flow abnormalities in the spectrum of IR that lead to type 2 DM were not reported previously in a composite study. We recently reported attenuation of cold-stimulated coronary flow in IR in the absence of glucose intolerance.5 The present study demonstrates a progressive impairment of coronary circulatory function with progressive carbohydrate intolerance. In the early stage of normoglycemic resistance, the impairment affects the endothelium-dependent vasomotor function that tends to worsen progressively with more severe states of IR and ultimately reduces the total vasodilator capacity, including the endothelium and the VSMC-dependent coronary circulatory function in hyperglycemic states of DM. Interestingly, the greatest loss in endothelium-dependent flow occurred with IR alone, tended to worsen with IGT and DM, and tended to further decrease in the presence of hypertension.
Mechanisms underlying the progressive impairment of coronary circulatory function from normoglycemic to hyperglycemic states of IR remain to be elucidated. The most attractive concept, however, refers to a decrease of NO bioavailability either due to diminished production, increased inactivation of NO, or both and may result from several mechanisms.3 Diminished IS, elevation of plasma free fatty acids, and hypertriglyceridemia have been implicated to alter intracellular signaling of NO synthase activity and to reduce NO production.3 In type 2 DM, hyperglycemia has been associated with increased production of reactive oxygen species, which results in inactivation of NO.26 An increase in endothelin activity may have further opposed the flow response to cold stimulation.27 Multivariate analysis in the present study identified HDL cholesterol and triglyceride plasma levels as independent predictors of the flow response to CPT. Indeed, low HDL cholesterol and high triglyceride levels appear early in IR and define the metabolic system.15 In one previous study, HDL levels were directly correlated, and in another study, triglyceride levels were inversely correlated with endothelium-dependent brachial artery function.28,29 Elevated plasma triglyceride levels were therefore likely to have further contributed to impairment of the endothelium-dependent coronary circulatory function.
Responses of MBF to adenosine or dipyridamole reflect the total coronary vasodilator capacity. Across the spectrum of IR, total vasodilator capacity was attenuated in the presence of DM. Although patients with diabetes in the present study were older than the other groups, previous studies from our laboratory failed to observe an age-dependent decline in hyperemic MBF,8 and when adjusted for age, hyperemic flows remained lower in patients with DM. The average 16% attenuation in the present investigation is comparable to the 20% reductions in hyperemic flows in previous studies.18–20,30,31 Adenosine- or dipyridamole-induced increases in MBF are mediated largely through VSMC relaxation, although endothelium-dependent mechanisms contribute to the hyperemic response, because inhibition of NO synthase decreases hyperemic flow in both the peripheral and the coronary circulation.32,33 Attenuation of total vasodilator capacity in type 2 DM may, therefore, reflect functional alterations of the endothelium, VSMC, or both. Previous studies demonstrating diminished vasodilator responsiveness to exogenous NO donors in patients with type 2 DM2 suggest that VSMC function was affected. This possibility would not be unexpected in DM, in which longstanding IR and hyperinsulinemia and, in hypertension, mechanical stress may alter VSMC function and structure. Alternatively, it is conceivable that additive effects of various mechanisms of reduced NO bioavailability may have diminished endothelial function progressively to an extent to which its contribution to total vasodilator capacity became significant.16
An important question is whether the coronary vasomotor abnormalities are reversible. Use of thiazolidinedione insulin sensitizers normalized responses to CPT in IR individuals without carbohydrate intolerance.5 This normalization was associated with improvement, but not normalization, of IS. The angiotensin II type 1 receptor blocker valsartan improved endothelium-dependent coronary vasomotor function measured by PET but did not affect IS.34 In addition, improvement in glucose control also improved CPT responses in DM.35 Thus, improvement in endothelium-related flow does not necessarily parallel IR. It is likely that therapies that decrease oxidative stress improve coronary vasomotor function, whatever the cause of the dysfunction. Because low HDL cholesterol correlated with abnormalities in both CPT responses and total vasodilator capacity, it would also be useful to determine whether increasing HDL cholesterol may improve coronary vasomotor dysfunction in IR.
Conclusions
Coronary vasomotor responses to CPT, which are largely endothelium related, are abnormal in IR in the absence of traditional coronary risk factors, worsen in the presence of carbohydrate intolerance, and tend to deteriorate further with diabetes and hypertension. HDL cholesterol and triglycerides are predictors of this response. Total coronary vasodilator capacity remains normal until the onset of DM. Fasting plasma glucose and HDL cholesterol are determinants of this response. Longitudinal studies are needed to understand the natural sequence of events that lead to abnormal coronary circulatory function in individuals at risk of developing type 2 DM. Nevertheless, these results underscore the progressive vascular injury that occurs in IR and carbohydrate intolerance and reinforces the need to prevent diabetes and the metabolic syndrome.
Acknowledgments
This study was supported in part by research grants No. HL 33177 (Dr Schelbert) and No. HL60030 (Dr Hsueh) from the National Heart, Lung, and Blood Institute, Bethesda, Md, and GCRC grant No. M01RR00865 (Dr Hsueh). The authors thank Roxana De La Rosa and Drs Isabel Bulnes-Enriquez, Tamara Modilevsky, and Katherine Yu for patient recruitment, Larry Pang and staff for assisting in the PET studies, Nagichettiar Satyamurthy and staff for 13N-ammonia, and Ashley Denault and Alfonso Orozco for administrative assistance.
References
Zeiher AM, Drexler H, Wollschlager H, Just H. Endothelial dysfunction of the coronary microvasculature is associated with coronary blood flow regulation in patients with early atherosclerosis. Circulation. 1991; 84: 1984–1992.
Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager MA. Impaired nitric oxide-mediated vasodilation in patients with non–insulin-dependent diabetes mellitus. J Am Coll Cardiol. 1996; 27: 567–574.
Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Circulation. 2003; 108: 1527–1532.
Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 2003; 26 (suppl 1): S5–S20.
Quinones MJ, Hernandez-Pampaloni M, Schelbert H, Bulnes-Enriquez I, Jimenez X, Hernandez G, De La Rosa R, Chon Y, Yang H, Nicholas SB, Modilevsky T, Yu K, Van Herle K, Castellani LW, Elashoff R, Hsueh WA. Coronary vasomotor abnormalities in insulin-resistant individuals. Ann Intern Med. 2004; 140: 700–708.
Haffner SM. Coronary heart disease in patients with diabetes. N Engl J Med. 2000; 342: 1040–1042.
Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR, Cull CA, Wright AD, Turner RC, Holman RR. Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): prospective observational study. BMJ. 2000; 321: 412–419.
Czernin J, Muller P, Chan S, Brunken RC, Porenta G, Krivokapich J, Chen K, Chan A, Phelps ME, Schelbert HR. Influence of age and hemodynamics on myocardial blood flow and flow reserve. Circulation. 1993; 88: 62–69.
Campisi R, Czernin J, Schoder H, Sayre JW, Schelbert HR. L-Arginine normalizes coronary vasomotion in long-term smokers. Circulation. 1999; 99: 491–497.
Preik M, Kelm M, Rosen P, Tschope D, Strauer BE. Additive effect of coexistent type 2 diabetes and arterial hypertension on endothelial dysfunction in resistance arteries of human forearm vasculature. Angiology. 2000; 51: 545–554.
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985; 28: 412–419.
Hermans MP, Levy JC, Morris RJ, Turner RC. Comparison of insulin sensitivity tests across a range of glucose tolerance from normal to diabetes. Diabetologia. 1999; 42: 678–687.
Campisi R, Czernin J, Schoder H, Sayre JW, Marengo FD, Phelps ME, Schelbert HR. Effects of long-term smoking on myocardial blood flow, coronary vasomotion, and vasodilator capacity. Circulation. 1998; 98: 119–125.
Huitema B. The Analysis of Covariance and Alternatives. New York, NY: Wiley; 1980.
Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults: findings from the third National Health and Nutrition Examination Survey. JAMA. 2002; 287: 356–359.
Zeiher AM, Drexler H, Saurbier B, Just H. Endothelium-mediated coronary blood flow modulation in humans: effects of age, atherosclerosis, hypercholesterolemia, and hypertension. J Clin Invest. 1993; 92: 652–662.
Schindler TH, Nitzsche EU, Olschewski M, Brink I, Mix M, Prior J, Facta A, Inubushi M, Just H, Schelbert HR. PET-measured responses of MBF to cold pressor testing correlate with indices of coronary vasomotion on quantitative coronary angiography. J Nucl Med. 2004; 45: 419–428.
Nitenberg A, Valensi P, Sachs R, Dali M, Aptecar E, Attali JR. Impairment of coronary vascular reserve and ACh-induced coronary vasodilation in diabetic patients with angiographically normal coronary arteries and normal left ventricular systolic function. Diabetes. 1993; 42: 1017–1025.
Di Carli MF, Bianco-Batlles D, Landa ME, Kazmers A, Groehn H, Muzik O, Grunberger G. Effects of autonomic neuropathy on coronary blood flow in patients with diabetes mellitus. Circulation. 1999; 100: 813–819.
Kjaer A, Meyer C, Nielsen FS, Parving HH, Hesse B. Dipyridamole, cold pressor test, and demonstration of endothelial dysfunction: a PET study of myocardial perfusion in diabetes. J Nucl Med. 2003; 44: 19–23.
Momose M, Abletshauser C, Neverve J, Nekolla SG, Schnell O, Standl E, Schwaiger M, Bengel FM. Dysregulation of coronary microvascular reactivity in asymptomatic patients with type 2 diabetes mellitus. Eur J Nucl Med Mol Imaging. 2002; 29: 1675–1679.
Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G, Baron AD. Obesity/insulin resistance is associated with endothelial dysfunction: implications for the syndrome of insulin resistance. J Clin Invest. 1996; 97: 2601–2610.
Caballero AE, Arora S, Saouaf R, Lim SC, Smakowski P, Park JY, King GL, LoGerfo FW, Horton ES, Veves A. Microvascular and macrovascular reactivity is reduced in subjects at risk for type 2 diabetes. Diabetes. 1999; 48: 1856–1862.
Balletshofer BM, Rittig K, Enderle MD, Volk A, Maerker E, Jacob S, Matthaei S, Rett K, Haring HU. Endothelial dysfunction is detectable in young normotensive first-degree relatives of subjects with type 2 diabetes in association with insulin resistance. Circulation. 2000; 101: 1780–1784.
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 Heart Circ Physiol. 2004; 286: H76–H82.
Lin KY, Ito A, Asagami T, Tsao PS, Adimoolam S, Kimoto M, Tsuji H, Reaven GM, Cooke JP. Impaired nitric oxide synthase pathway in diabetes mellitus: role of asymmetric dimethylarginine and dimethylarginine dimethylaminohydrolase. Circulation. 2002; 106: 987–992.
Cardillo C, Campia U, Bryant MB, Panza JA. Increased activity of endogenous endothelin in patients with type II diabetes mellitus. Circulation. 2002; 106: 1783–1787.
Kuhn FE, Mohler ER, Satler LF, Reagan K, Lu DY, Rackley CE. Effects of high-density lipoprotein on acetylcholine-induced coronary vasoreactivity. Am J Cardiol. 1991; 68: 1425–1430.
Lupattelli G, Lombardini R, Schillaci G, Ciuffetti G, Marchesi S, Siepi D, Mannarino E. Flow-mediated vasoactivity and circulating adhesion molecules in hypertriglyceridemia: association with small, dense LDL cholesterol particles. Am Heart J. 2000; 140: 521–526.
Nahser PJ Jr, Brown RE, Oskarsson H, Winniford MD, Rossen JD. Maximal coronary flow reserve and metabolic coronary vasodilation in patients with diabetes mellitus. Circulation. 1995; 91: 635–640.
Yokoyama I, Yonekura K, Ohtake T, Yang W, Shin WS, Yamada N, Ohtomo K, Nagai R. Coronary microangiopathy in type 2 diabetic patients: relation to glycemic control, sex, and microvascular angina rather than to coronary artery disease. J Nucl Med. 2000; 41: 978–985.
Smits P, Williams SB, Lipson DE, Banitt P, Rongen GA, Creager MA. Endothelial release of nitric oxide contributes to the vasodilator effect of adenosine in humans. Circulation. 1995; 92: 2135–2141.
Buus NH, Bottcher M, Hermansen F, Sander M, Nielsen TT, Mulvany MJ. Influence of nitric oxide synthase and adrenergic inhibition on adenosine-induced myocardial hyperemia. Circulation. 2001; 104: 2305–2310.
Qui;ones MJ, Schindler TH, Bulnes-Enriquez I, Facta A, Moore L, De La Rosa R, Lyon C, Aurea G, Garcia M, Alvarado A, Ramirez A, Modilevsky T, Yu K, Daley W, Baron M, Schelbert H, Hsueh WA. Angiotensin receptor blockade with valsartan normalizes coronary endothelial function in prediabetic subjects. Circulation. 2004; 110 (suppl 3): III-80. Abstract.
Facta A, Prior J, Schindler TH, Cadenas J, Quinones MJ, Hsueh WA, Schelbert H. Effect of glucose lowering on coronary circulatory dysfunction in type 2 diabetes mellitus. J Am Coll Cardiol. 2004; 43: 339A–340A.Abstract.(John O. Prior, MD, PhD; M)
Division of Endocrinology, Diabetes and Hypertension (M.J.Q., W.A.H.)
David Geffen School of Medicine at UCLA, Los Angeles, Calif.
Abstract
Background— Abnormal coronary endothelial reactivity has been demonstrated in diabetes and is associated with an increased rate of cardiovascular events. Our objectives were to investigate the presence of functional coronary circulatory abnormalities over the full spectrum of insulin resistance and to determine whether these would differ in severity with more advanced states of insulin resistance.
Methods and Results— Myocardial blood flow (MBF) was measured with positron emission tomography and 13N-ammonia to characterize coronary circulatory function in states of insulin resistance without carbohydrate intolerance (IR), impaired glucose tolerance (IGT), and normotensive and hypertensive type 2 diabetes mellitus (DM) compared with insulin-sensitive (IS) individuals. Indices of coronary function were total vasodilator capacity (mostly vascular smooth muscle–mediated) during pharmacological vasodilation and the nitric oxide–mediated, endothelium-dependent vasomotion in response to cold pressor testing. Total vasodilator capacity was similar in normoglycemic individuals (IS, IR, and IGT), whereas it was significantly decreased in normotensive (–17%) and hypertensive (–34%) DM patients. Compared with IS, endothelium-dependent coronary vasomotion was significantly diminished in IR (–56%), as well as in IGT and normotensive and hypertensive diabetic patients (–85%, –91%, and –120%, respectively).
Conclusions— Progressively worsening functional coronary circulatory abnormalities of nitric oxide–mediated, endothelium-dependent vasomotion occur with increasing severity of insulin-resistance and carbohydrate intolerance. Attenuated total vasodilator capacity accompanies the more clinically evident metabolic abnormalities in diabetes.
Key Words: blood flow ; diabetes mellitus ; endothelium ; hypertension
Introduction
Abnormal coronary endothelial reactivity represents one of the earliest manifestations of vascular disease and is a key element in the development of atherosclerosis.1 It exists not only in diabetes and in groups at high risk for type 2 diabetes mellitus, such as impaired glucose tolerance (IGT), but also in insulin-resistant individuals who do not have significant metabolic abnormalities.2–5 The cardiovascular mortality rate in IGT is twice that of normal glucose tolerance, increases 2- to 3-fold in type 2 diabetes mellitus,6 and rises further when diabetes is associated with hypertension.7 Because the high mortality cannot be fully explained by conventional cardiovascular risk factors, it has been attributed to factors related to insulin resistance and its potentially adverse effects on endothelial function. A key issue is whether and how endothelial dysfunction and vascular injury progress as insulin resistance progresses to type 2 diabetes mellitus. Positron emission tomography (PET) affords noninvasive measurements of myocardial blood flow (MBF).8,9 MBF responses to vascular smooth muscle cell (VSMC) relaxing agents like dipyridamole or adenosine yield information on total coronary vasodilator function. MBF responses to sympathetic stimulation with cold pressor testing (CPT) provide information on endothelium-related coronary vasomotion. We hypothesized that abnormalities in coronary circulatory function would parallel insulin resistance severity and carbohydrate intolerance in the progression to type 2 diabetes mellitus. To address this issue, we used PET to compare coronary circulatory function in Mexican-American, nonsmoking individuals who were insulin-sensitive (IS) or insulin-resistant (IR), or who had IGT or type 2 diabetes mellitus (DM) with and without hypertension.
Methods
Study Population
One-hundred seventy-four Mexican Americans underwent screening that included medical history, physical examination, blood pressure measurements, and an ECG; fasting blood collection for routine testing (blood cell count, chemistry, and lipid panel) and routine urinalysis were performed by the Medical Center Clinical Pathology Laboratory. All individuals without a history of diabetes underwent a 2-hour 75-g oral glucose tolerance test with measurement of glucose (glucose oxidase technique; YSI, Inc) and insulin levels (Human Insulin-Specific RIA Kit; Linco Research, Inc). Glucose tolerance was categorized according to American Diabetes Association criteria4 (IGT, fasting glucose 6.11 to 7.00 mmol/L and 2-hour plasma glucose 7.77 to 11.10 mmol/L; DM, fasting plasma glucose 7.00 mmol/L or 2-hour glucose 11.10 mmol/L). Nondiabetic individuals underwent peripheral insulin sensitivity measurement by a modified hyperinsulinemic-euglycemic clamp.5 Briefly, a priming dose of human insulin (Novolin; Novo Nordisk) was followed by insulin infusion (60 mU · m–2 · min–1) for 120 minutes to achieve plasma insulin levels 700 pmol/L. Blood was sampled every 5 minutes, and 20% glucose infusion rate (GINF) was adjusted to maintain euglycemia (5.00 to 5.55 mmol/L). The GINF (mg/kg of body weight per minute) over the last 30 minutes of the clamp reflected insulin action and defined IS (GINF 7.5 mg · kg–1 · min–1) and IR (GINF 4.0 mg · kg–1 · min–1) individuals.5 Of the 174 individuals, 14 declined to participate in the study. Forty individuals were excluded from the study (hypertension [n=3], hypercholesterolemia [n=9], or triglyceride levels 3.40 mmol/L [n=15] in the absence of diabetes; smoking [n=2]; females taking oral contraceptives [n=6]; diabetic patients taking hypoglycemic agents [n=4]; and hypertensive diabetic patients taking ;-blockers [n=1]). The remaining 120 individuals were assigned to 1 of 5 study groups: (1) IS group (n=19) without coronary risk factors and normal insulin sensitivity; (2) IR group (n=47), without coronary risk factors and normal carbohydrate tolerance; (3) IGT group (n=25); (4) DM group without hypertension (n=21); and (5) hypertensive group of diabetic individuals with hypertension (HTN group; n=8). Diabetic patients were subgrouped on the basis of the presence (blood pressure >135/80 mm Hg) or absence of hypertension, which is known to adversely affect coronary circulation.10 Comparisons across study groups were performed with the homeostasis model assessment (HOMA) index of insulin resistance11 (fasting plasma glucose [mmol/L]xfasting plasma insulin [pmol/L]/162), which has been found to correlate with glucose clamp measurement in nondiabetic, diabetic, or hypertensive populations.12 Of the 76 women in the study, 6 were postmenopausal (cessation of menses 1year) and were also diabetic. The time since diagnosis of diabetes averaged 2.1±3.2 years (range 0 to 11 years). None of the participants were taking antihypertensive, antidiabetic, or lipid-lowering medications. The study was approved by the UCLA Institutional Review Board, and written informed consent was obtained from all participants.
Evaluation and Measurement of MBF
MBF was measured with 13N-ammonia and PET (ECAT EXACT HR/HR+, CTI/Siemens). Beginning with the intravenous injection of 13N-ammonia (555 to 740 MBq), serial transaxial images were acquired for 19 minutes, as described previously.13 Reoriented short- and long-axis myocardial slices of the last 15-minute transaxial image and the corresponding polar maps were submitted to visual analysis of the relative MBF distribution. Regions of interest assigned on the last 15-minute image to coronary artery territories and the left ventricular blood pool were copied to the serial images acquired during the first 2 minutes to generate time-activity curves and to estimate MBF as described previously.13 Because no significant MBF differences existed between coronary territories, regional MBFs were averaged.
Participants refrained from drinking caffeine-containing substances for 24 hours before study and had been fasting for 6 hours. MBF was measured at rest, then during CPT, and finally during pharmacological vasodilation with adenosine or dipyridamole, separated by 35 to 45 minutes. CPT was performed by hand immersion in ice water for 2 minutes, with administration of 13N-ammonia and imaging at 45 seconds after onset of the test. Image acquisition and 13N-ammonia administration began 3 minutes after completion of the 4-minute infusion of dipyridamole (0.54 mg/kg) or 3 minutes after the 6-minute infusion of adenosine (140 μg · kg–1 · min–1) was begun. Studies in nondiabetic individuals used dipyridamole, whereas adenosine was used in 11 of 21 normotensive diabetic participants and in all hypertensive diabetic participants. Heart rate, blood pressure, and a 12-lead ECG were recorded at 1-minute intervals during each procedure. Heart rates and blood pressures were averaged for the first 2 minutes of data acquisition, and the rate-pressure product (RPP=heart ratexsystolic blood pressure) was derived as an index of cardiac work. Image artifacts due to subject motion, side effects to adenosine or dipyridamole, and poor image quality precluded image acquisition or analysis of 1 rest, 6 CPT, and 11 hyperemic studies.
Statistical Analysis
Comparisons across groups were performed with 1-way ANOVA with post hoc Scheffe tests; results are presented as mean±SD, unless otherwise indicated. Skewed variables (P>0.05 on skewness-kurtosis test of normality; fasting plasma glucose and insulin, GINF, HOMA, and triglycerides) were analyzed after logarithmic transformation. Because significant baseline intergroup differences existed, ANCOVA14 was used to compare mean MBF response after adjustment for mean differences in potentially confounding variables (covariates: age, gender, body mass index [BMI], and resting flow). Interaction terms (groupxcovariate) were negligible (P0.22); pairwise group differences were assessed with least significant difference tests. The association of MBF responses to CPT and vasodilation with metabolic and clinical variables was assessed with univariate (Spearman rank) correlation and multivariable linear modeling (stepwise forward regression of variables with P0.1 significance in univariate analysis). Significance was considered for 2-sided P0.05, and Stata software (version 8.2, Stata Corporation) was used for analyses.
Results
Study Population and Laboratory Findings
Hemodynamic and MBF Findings
Blood pressure at baseline was similar in the normotensive study groups but was elevated in the HTN group (Table 1). Both CPT and adenosine or dipyridamole infusion significantly raised blood pressures in all study groups with the exception of a decline during pharmacological vasodilation in the DM group. Nevertheless, systolic blood pressures in the HTN group remained higher than in the other groups during CPT and during pharmacological vasodilation. The increase in RPP from rest to CPT (RPP; defined in absolute units) was similar in all 5 study groups, whereas the average RPP during CPT was higher in the HTN group than in the remaining groups (Table 2).
At rest, MBF was higher in the HTN group than in the other study groups (P<0.01; Table 2), and a trend existed for higher values in the IR group than in the IS group (P=0.11). Multivariate analysis identified the RPP as the only independent determinant of MBF at rest. Accordingly, normalization of MBFs by myocardial work (MBF · RPP–1) abolished differences in flow estimates of the HTN group compared those in the other study groups and between the IR and IS groups, so that the observed intergroup differences in resting MBF were related to differences in cardiac work.
Pharmacological vasodilation significantly raised MBF in all study groups. The flow increase was similar across the groups, although hyperemic MBF in the 2 diabetes groups tended to be lower than in the other groups (Table 2). When hyperemic MBFs were combined for all diabetes patients (DM and hypertensive groups), they were significantly lower than in the normoglycemic IR groups (IR and IGT; P>0.001). Accordingly, the myocardial flow reserve as the ratio of hyperemic to rest MBF in the diabetic patients was reduced significantly compared with that of the IS group (Table 2). Because patients in both diabetes groups were older than individuals in the other study groups, hyperemic MBFs were further adjusted for age by ANCOVA. Age-adjusted hyperemic MBFs were highest in the IS group and tended to decline progressively from the normoglycemic to the hyperglycemic IR groups (Table 4). The adjusted mean hyperemic MBFs in the normotensive diabetes group were significantly lower than in the IGT group (P<0.05; Table 4) but higher than in the hypertensive diabetes group (P<0.01). Furthermore, on nonparametric testing for trends across ordered groups, the trend of a progressive decline of hyperemic MBFs from the IS to the hypertensive groups achieved statistical significance (P<0.002). Univariate analysis of possible associations between hyperemic MBFs and other clinical findings revealed significant inverse correlations with fasting plasma glucose and MBF at rest and a positive correlation with HDL cholesterol (Table 3). There was a trend to inversely correlate with age, LDL cholesterol, and systolic blood pressure. On multivariate analysis, however, only fasting plasma glucose and HDL cholesterol remained independent predictors of hyperemic MBFs.
CPT consistently raised MBF in IS individuals; by contrast, individual flow responses in the other groups varied and ranged from increases to no change or to decreases. Average values of MBF during CPT varied between groups, although there were no significant intergroup differences with the exception of a significant difference between the hypertensive diabetes and the IGT groups (Table 2). As observed previously in normal volunteers,1 the cold-induced increase in RPP was accompanied in IS individuals by a proportionate increase in MBF (r=0.52; P=0.028). As Table 2 indicates, RPP increased by 41±23%, whereas MBF increased by 42±13%. In hypertensive diabetic patients, however, the percent increase in RPP was only 21±24%, whereas MBFs remained unchanged (–2±12%).
Already increased values of RPP and MBF account at least in part for the attenuated responses when expressed in percentages of the rest RPP or MBF. To reduce a baseline-dependent bias, responses to CPT were estimated in absolute units as the difference in MBF or in RPP during CPT and at rest. In the IS group, RPP increased with CPT by RPP=2.7±1.4x103 mm Hg/min and was associated with a MBF=0.25±0.08 mL · min–1 · g–1 flow increase. In the IR group, RPP similarly increased by 2.6±1.8x103 mm Hg/min (P=0.99 versus IS), but MBF increased by only 0.09±0.17x103 mL · min–1 · g–1 (P<0.001 versus IS). Furthermore, in the hypertensive diabetic group, RPP rose by 2.1±2.7x103 mm Hg/min, which was not significantly different from the increase in the IS group, whereas MBF remained virtually unchanged (–0.02±0.11 mL · min–1 · g–1). Thus, despite comparable increases in RPP (and thus, increases in cardiac work), responses in MBF to cold were significantly diminished or absent in the 4 IR groups and significantly differed from those in the IS control group (Table 2).
On univariate analysis, MBF was found to depend on BMI (Table 3), and significant intergroup differences existed for age and gender. Accordingly, the flow difference values were adjusted for BMI, age, gender, and MBF at rest. They are listed in Table 4 and demonstrate significant intergroup differences (Figure 2). The adjusted mean MBF responses in the IR group were significantly lower than in the IS group (P<0.001) but higher than in the 2 diabetic groups (P<0.05). Furthermore, nonparametric tests for trend across ordered groups indicated that the progressive decline of the MBF response across the IR states was statistically significant (P=0.02). Possible associations of the flow response (MBF) to CPT with clinical and laboratory findings were explored with univariate and multivariate analysis. In addition to the correlations with BMI and MBF at rest as mentioned above, univariate analysis revealed statistically significant inverse correlations between MBF and fasting plasma glucose, insulin levels, and HOMA, as well as with triglyceride concentrations. Furthermore, MBF correlated positively with HDL cholesterol levels. On multivariate analysis, only HDL cholesterol and triglyceride levels were found to be independent predictors of MBF.
Discussion
In this study, coronary vasomotor function was compared in a single ethnic group, Mexican-Americans, who were IS or who had IR, IGT, DM, or DM with hypertension. Mexican-Americans have a high prevalence of insulin resistance, metabolic syndrome, and DM.15 The main findings of the study are as follows: (1) the presence of a functional abnormality of the coronary circulation across the spectrum of insulin resistance, and (2) progression of this functional alteration with carbohydrate intolerance. In states of insulin resistance that lead to diabetes, this alteration primarily involved the flow response to CPT, which is largely endothelium dependent. In the presence of overt type 2 DM, there was also an attenuation of the total vasodilator capacity that tended to worsen with hypertension. When adjusted for age, gender, BMI, and MBF at rest, hyperemic MBFs and MBF responses to CPT differed between IR groups, and the trend of progressively declining MBFs was found to be statistically significant. These data suggest that vascular injury occurs when insulin resistance is the only abnormality, even in the absence of traditional coronary risk factors, and progresses as carbohydrate intolerance appears. These results are important in view of the worldwide epidemic of insulin resistance and are of concern because of the relatively early stage of appearance of vascular injury in the cascade of insulin resistance.
Flow responses to CPT largely reflect endothelium-related coronary vasomotor function and correlate with invasively and noninvasively measured indices of nitric oxide (NO)–mediated changes in coronary flow.1,16–18 Sympathetic stimulation with CPT raises both heart rate and systolic blood pressure and, thus, the RRP, an index of myocardial work. In the normal coronary circulation, the increase in myocardial work is associated with a proportionate metabolically mediated increase in coronary flow as previously reported and confirmed in the IS individuals in the present study.1,18 In the IR individuals, however, despite comparable increases in RRP, MBF failed to increase or even declined with CPT. This attenuated flow response has been reported with PET in patients with type 2 DM.19–21 Similarly, intracoronary acetylcholine stimulation in type 2 DM patients produced decreases rather than increases in conduit vessel diameter, whereas CPT induced a decrease in epicardial coronary diameter compared with an increase seen in normal controls.18 Thus, diabetes is associated with impaired function of both the conduit and resistance vessels.
An important finding of the present investigation was that the alteration of coronary vasomotor function was not confined to type 2 DM but was also present in individuals with normoglycemic IR. Previous investigations in peripheral arteries provided some evidence for the presence of endothelial dysfunction at different stages of development of diabetes. Acetylcholine-stimulated peripheral artery dilation was shown to be diminished in obese individuals with and without IGT, and both skin blood flow and brachial artery responses were diminished in relatives of type 2 DM patients.22–25 These studies suggested an impairment of the endothelium-related function of both the conduit and resistance vessels of the peripheral vasculature in IR that was similar to that described for the coronary vasculature in DM. However, flow abnormalities in the spectrum of IR that lead to type 2 DM were not reported previously in a composite study. We recently reported attenuation of cold-stimulated coronary flow in IR in the absence of glucose intolerance.5 The present study demonstrates a progressive impairment of coronary circulatory function with progressive carbohydrate intolerance. In the early stage of normoglycemic resistance, the impairment affects the endothelium-dependent vasomotor function that tends to worsen progressively with more severe states of IR and ultimately reduces the total vasodilator capacity, including the endothelium and the VSMC-dependent coronary circulatory function in hyperglycemic states of DM. Interestingly, the greatest loss in endothelium-dependent flow occurred with IR alone, tended to worsen with IGT and DM, and tended to further decrease in the presence of hypertension.
Mechanisms underlying the progressive impairment of coronary circulatory function from normoglycemic to hyperglycemic states of IR remain to be elucidated. The most attractive concept, however, refers to a decrease of NO bioavailability either due to diminished production, increased inactivation of NO, or both and may result from several mechanisms.3 Diminished IS, elevation of plasma free fatty acids, and hypertriglyceridemia have been implicated to alter intracellular signaling of NO synthase activity and to reduce NO production.3 In type 2 DM, hyperglycemia has been associated with increased production of reactive oxygen species, which results in inactivation of NO.26 An increase in endothelin activity may have further opposed the flow response to cold stimulation.27 Multivariate analysis in the present study identified HDL cholesterol and triglyceride plasma levels as independent predictors of the flow response to CPT. Indeed, low HDL cholesterol and high triglyceride levels appear early in IR and define the metabolic system.15 In one previous study, HDL levels were directly correlated, and in another study, triglyceride levels were inversely correlated with endothelium-dependent brachial artery function.28,29 Elevated plasma triglyceride levels were therefore likely to have further contributed to impairment of the endothelium-dependent coronary circulatory function.
Responses of MBF to adenosine or dipyridamole reflect the total coronary vasodilator capacity. Across the spectrum of IR, total vasodilator capacity was attenuated in the presence of DM. Although patients with diabetes in the present study were older than the other groups, previous studies from our laboratory failed to observe an age-dependent decline in hyperemic MBF,8 and when adjusted for age, hyperemic flows remained lower in patients with DM. The average 16% attenuation in the present investigation is comparable to the 20% reductions in hyperemic flows in previous studies.18–20,30,31 Adenosine- or dipyridamole-induced increases in MBF are mediated largely through VSMC relaxation, although endothelium-dependent mechanisms contribute to the hyperemic response, because inhibition of NO synthase decreases hyperemic flow in both the peripheral and the coronary circulation.32,33 Attenuation of total vasodilator capacity in type 2 DM may, therefore, reflect functional alterations of the endothelium, VSMC, or both. Previous studies demonstrating diminished vasodilator responsiveness to exogenous NO donors in patients with type 2 DM2 suggest that VSMC function was affected. This possibility would not be unexpected in DM, in which longstanding IR and hyperinsulinemia and, in hypertension, mechanical stress may alter VSMC function and structure. Alternatively, it is conceivable that additive effects of various mechanisms of reduced NO bioavailability may have diminished endothelial function progressively to an extent to which its contribution to total vasodilator capacity became significant.16
An important question is whether the coronary vasomotor abnormalities are reversible. Use of thiazolidinedione insulin sensitizers normalized responses to CPT in IR individuals without carbohydrate intolerance.5 This normalization was associated with improvement, but not normalization, of IS. The angiotensin II type 1 receptor blocker valsartan improved endothelium-dependent coronary vasomotor function measured by PET but did not affect IS.34 In addition, improvement in glucose control also improved CPT responses in DM.35 Thus, improvement in endothelium-related flow does not necessarily parallel IR. It is likely that therapies that decrease oxidative stress improve coronary vasomotor function, whatever the cause of the dysfunction. Because low HDL cholesterol correlated with abnormalities in both CPT responses and total vasodilator capacity, it would also be useful to determine whether increasing HDL cholesterol may improve coronary vasomotor dysfunction in IR.
Conclusions
Coronary vasomotor responses to CPT, which are largely endothelium related, are abnormal in IR in the absence of traditional coronary risk factors, worsen in the presence of carbohydrate intolerance, and tend to deteriorate further with diabetes and hypertension. HDL cholesterol and triglycerides are predictors of this response. Total coronary vasodilator capacity remains normal until the onset of DM. Fasting plasma glucose and HDL cholesterol are determinants of this response. Longitudinal studies are needed to understand the natural sequence of events that lead to abnormal coronary circulatory function in individuals at risk of developing type 2 DM. Nevertheless, these results underscore the progressive vascular injury that occurs in IR and carbohydrate intolerance and reinforces the need to prevent diabetes and the metabolic syndrome.
Acknowledgments
This study was supported in part by research grants No. HL 33177 (Dr Schelbert) and No. HL60030 (Dr Hsueh) from the National Heart, Lung, and Blood Institute, Bethesda, Md, and GCRC grant No. M01RR00865 (Dr Hsueh). The authors thank Roxana De La Rosa and Drs Isabel Bulnes-Enriquez, Tamara Modilevsky, and Katherine Yu for patient recruitment, Larry Pang and staff for assisting in the PET studies, Nagichettiar Satyamurthy and staff for 13N-ammonia, and Ashley Denault and Alfonso Orozco for administrative assistance.
References
Zeiher AM, Drexler H, Wollschlager H, Just H. Endothelial dysfunction of the coronary microvasculature is associated with coronary blood flow regulation in patients with early atherosclerosis. Circulation. 1991; 84: 1984–1992.
Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager MA. Impaired nitric oxide-mediated vasodilation in patients with non–insulin-dependent diabetes mellitus. J Am Coll Cardiol. 1996; 27: 567–574.
Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Circulation. 2003; 108: 1527–1532.
Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 2003; 26 (suppl 1): S5–S20.
Quinones MJ, Hernandez-Pampaloni M, Schelbert H, Bulnes-Enriquez I, Jimenez X, Hernandez G, De La Rosa R, Chon Y, Yang H, Nicholas SB, Modilevsky T, Yu K, Van Herle K, Castellani LW, Elashoff R, Hsueh WA. Coronary vasomotor abnormalities in insulin-resistant individuals. Ann Intern Med. 2004; 140: 700–708.
Haffner SM. Coronary heart disease in patients with diabetes. N Engl J Med. 2000; 342: 1040–1042.
Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR, Cull CA, Wright AD, Turner RC, Holman RR. Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): prospective observational study. BMJ. 2000; 321: 412–419.
Czernin J, Muller P, Chan S, Brunken RC, Porenta G, Krivokapich J, Chen K, Chan A, Phelps ME, Schelbert HR. Influence of age and hemodynamics on myocardial blood flow and flow reserve. Circulation. 1993; 88: 62–69.
Campisi R, Czernin J, Schoder H, Sayre JW, Schelbert HR. L-Arginine normalizes coronary vasomotion in long-term smokers. Circulation. 1999; 99: 491–497.
Preik M, Kelm M, Rosen P, Tschope D, Strauer BE. Additive effect of coexistent type 2 diabetes and arterial hypertension on endothelial dysfunction in resistance arteries of human forearm vasculature. Angiology. 2000; 51: 545–554.
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985; 28: 412–419.
Hermans MP, Levy JC, Morris RJ, Turner RC. Comparison of insulin sensitivity tests across a range of glucose tolerance from normal to diabetes. Diabetologia. 1999; 42: 678–687.
Campisi R, Czernin J, Schoder H, Sayre JW, Marengo FD, Phelps ME, Schelbert HR. Effects of long-term smoking on myocardial blood flow, coronary vasomotion, and vasodilator capacity. Circulation. 1998; 98: 119–125.
Huitema B. The Analysis of Covariance and Alternatives. New York, NY: Wiley; 1980.
Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults: findings from the third National Health and Nutrition Examination Survey. JAMA. 2002; 287: 356–359.
Zeiher AM, Drexler H, Saurbier B, Just H. Endothelium-mediated coronary blood flow modulation in humans: effects of age, atherosclerosis, hypercholesterolemia, and hypertension. J Clin Invest. 1993; 92: 652–662.
Schindler TH, Nitzsche EU, Olschewski M, Brink I, Mix M, Prior J, Facta A, Inubushi M, Just H, Schelbert HR. PET-measured responses of MBF to cold pressor testing correlate with indices of coronary vasomotion on quantitative coronary angiography. J Nucl Med. 2004; 45: 419–428.
Nitenberg A, Valensi P, Sachs R, Dali M, Aptecar E, Attali JR. Impairment of coronary vascular reserve and ACh-induced coronary vasodilation in diabetic patients with angiographically normal coronary arteries and normal left ventricular systolic function. Diabetes. 1993; 42: 1017–1025.
Di Carli MF, Bianco-Batlles D, Landa ME, Kazmers A, Groehn H, Muzik O, Grunberger G. Effects of autonomic neuropathy on coronary blood flow in patients with diabetes mellitus. Circulation. 1999; 100: 813–819.
Kjaer A, Meyer C, Nielsen FS, Parving HH, Hesse B. Dipyridamole, cold pressor test, and demonstration of endothelial dysfunction: a PET study of myocardial perfusion in diabetes. J Nucl Med. 2003; 44: 19–23.
Momose M, Abletshauser C, Neverve J, Nekolla SG, Schnell O, Standl E, Schwaiger M, Bengel FM. Dysregulation of coronary microvascular reactivity in asymptomatic patients with type 2 diabetes mellitus. Eur J Nucl Med Mol Imaging. 2002; 29: 1675–1679.
Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G, Baron AD. Obesity/insulin resistance is associated with endothelial dysfunction: implications for the syndrome of insulin resistance. J Clin Invest. 1996; 97: 2601–2610.
Caballero AE, Arora S, Saouaf R, Lim SC, Smakowski P, Park JY, King GL, LoGerfo FW, Horton ES, Veves A. Microvascular and macrovascular reactivity is reduced in subjects at risk for type 2 diabetes. Diabetes. 1999; 48: 1856–1862.
Balletshofer BM, Rittig K, Enderle MD, Volk A, Maerker E, Jacob S, Matthaei S, Rett K, Haring HU. Endothelial dysfunction is detectable in young normotensive first-degree relatives of subjects with type 2 diabetes in association with insulin resistance. Circulation. 2000; 101: 1780–1784.
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 Heart Circ Physiol. 2004; 286: H76–H82.
Lin KY, Ito A, Asagami T, Tsao PS, Adimoolam S, Kimoto M, Tsuji H, Reaven GM, Cooke JP. Impaired nitric oxide synthase pathway in diabetes mellitus: role of asymmetric dimethylarginine and dimethylarginine dimethylaminohydrolase. Circulation. 2002; 106: 987–992.
Cardillo C, Campia U, Bryant MB, Panza JA. Increased activity of endogenous endothelin in patients with type II diabetes mellitus. Circulation. 2002; 106: 1783–1787.
Kuhn FE, Mohler ER, Satler LF, Reagan K, Lu DY, Rackley CE. Effects of high-density lipoprotein on acetylcholine-induced coronary vasoreactivity. Am J Cardiol. 1991; 68: 1425–1430.
Lupattelli G, Lombardini R, Schillaci G, Ciuffetti G, Marchesi S, Siepi D, Mannarino E. Flow-mediated vasoactivity and circulating adhesion molecules in hypertriglyceridemia: association with small, dense LDL cholesterol particles. Am Heart J. 2000; 140: 521–526.
Nahser PJ Jr, Brown RE, Oskarsson H, Winniford MD, Rossen JD. Maximal coronary flow reserve and metabolic coronary vasodilation in patients with diabetes mellitus. Circulation. 1995; 91: 635–640.
Yokoyama I, Yonekura K, Ohtake T, Yang W, Shin WS, Yamada N, Ohtomo K, Nagai R. Coronary microangiopathy in type 2 diabetic patients: relation to glycemic control, sex, and microvascular angina rather than to coronary artery disease. J Nucl Med. 2000; 41: 978–985.
Smits P, Williams SB, Lipson DE, Banitt P, Rongen GA, Creager MA. Endothelial release of nitric oxide contributes to the vasodilator effect of adenosine in humans. Circulation. 1995; 92: 2135–2141.
Buus NH, Bottcher M, Hermansen F, Sander M, Nielsen TT, Mulvany MJ. Influence of nitric oxide synthase and adrenergic inhibition on adenosine-induced myocardial hyperemia. Circulation. 2001; 104: 2305–2310.
Qui;ones MJ, Schindler TH, Bulnes-Enriquez I, Facta A, Moore L, De La Rosa R, Lyon C, Aurea G, Garcia M, Alvarado A, Ramirez A, Modilevsky T, Yu K, Daley W, Baron M, Schelbert H, Hsueh WA. Angiotensin receptor blockade with valsartan normalizes coronary endothelial function in prediabetic subjects. Circulation. 2004; 110 (suppl 3): III-80. Abstract.
Facta A, Prior J, Schindler TH, Cadenas J, Quinones MJ, Hsueh WA, Schelbert H. Effect of glucose lowering on coronary circulatory dysfunction in type 2 diabetes mellitus. J Am Coll Cardiol. 2004; 43: 339A–340A.Abstract.(John O. Prior, MD, PhD; M)