|
|
||||||||||
|
J Am Coll Cardiol, 2003; 41:1071-1077, doi:10.1016/S0735-1097(03)00088-3 © 2003 by the American College of Cardiology Foundation |
* Institute of Cardiology, Catholic University, Rome, Italy
Manuscript received August 15, 2002; revised manuscript received October 31, 2002, accepted November 19, 2002.
* Reprint requests and correspondence: Dr. Luigi M. Biasucci, Institute of Cardiology, Catholic University, Largo A. Gemelli 8, 00168 Rome, Italy.
lmbiasucci{at}virgilio.it
| Abstract |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||
Diabetes and related metabolic diseases, such as hyperinsulinemia, insulin resistance, and central obesity, are recognized as major contributors to cardiovascular morbidity and mortality. Interestingly, recent and compelling evidence has shown the significant and independent role of systemic and coronary inflammation in the initiation, progression, and precipitation of atherothrombosis superimposed on traditional risk factors (5). Given the increasing prevalence of diabetes and dysmetabolic syndrome, the diffuse atherosclerotic burden of diabetics, the possible synergistic effect of diabetes and inflammation in atherosclerosis, as well as the potential for a preventive and therapeutic benefit in the modulation of inflammation in diabetics, this review covers the basic epidemiologic and pathophysiologic aspects underlying these two disease processes (Fig. 1).
|
| Diabetes mellitus,atherothrombosis, and inflammation |
|---|
|
|
|---|
Genetic abnormalities. Several genetic abnormalities in glycemic metabolism have been associated with the development of diabetes, diabetic complications, or aspects of the dysmetabolic syndrome. All of these derangements could be also involved in atherogenesis. Transcription factors, such as peroxisome proliferator-activated receptor (PPAR)-alpha and -gamma, are now being intensively studied. Whereas PPAR-alpha primarily stimulates the beta-oxidative degradation of fatty acids, PPAR-gamma promotes lipid storage by regulating adipocyte differentiation. Peroxisome proliferator-activated receptor-alpha is the molecular target of lipid-lowering drugs such as fibrates, and PPAR-gamma interacts with insulin-sensitizer drugs such as thiazolidinediones, and they may significantly modulate atherogenicity and inflammation. In particular, mutations in the PPAR-gamma gene have recently been shown to be associated with diabetes, glucose intolerance, and hypertension (9).
The role of transcription factors in homeostasis and disease is quite complex and still unclear, owing to the heterogenicity, cell specificity, and pleiotropy of their effects. However, substantial evidence has shown that PPARs often interact intracellularly with another transcription factor, nuclear factor-kappa-B (NF-kappa-B), whose activities are mainly proinflammatory and proliferative. Peroxisome proliferator-activated receptor-alphadeficient mice show a prolonged systemic acute-phase response and a local innate and immune inflammatory reaction to common noxious stimuli, whereas PPAR-alpha activation appears to decrease vascular cell adhesion molecule-1 (VCAM-1) expression by activated endothelial cells, thereby switching off the proatherogenic-activated endothelium (10). Moreover, PPAR-alpha selectively downregulates human aortic smooth muscle cell secreting and proliferating activities (11). In particular, ligands of this transcription factor inhibit interleukin (IL)-1induced production of IL-6 and prostaglandin and expression of cyclooxygenase-2 in endothelial cells. In hyperlipidemic patients, fenofibrate-dependent PPAR-alpha activation decreases levels of IL-6, fibrinogen, and C-reactive protein (CRP) (11). On the other hand, thiazolidinedione-activated PPARs inhibit both the proliferation and migration of vascular smooth muscle cells. In endothelial cells, the PPAR agonist troglitazone markedly attenuates tumor necrosis factor-alpha (TNF-alpha)induced expression of VCAM-1 and intercellular adhesion molecule-1 (ICAM-1), whereas apolipoprotein E-deficient mice treated with troglitazone show reduced monocyte/macrophage homing to atherosclerotic plaques (12). Finally, the combined activation of PPAR-alpha and -gamma appears to decrease CRP-mediated production of monocyte chemoattractant protein-1 by human endothelial cells, an effect similar to that achieved by simvastatin. Moreover, type 2 diabetic patients treated with troglitazone have increased levels of NF-kappa-Binhibiting protein (I-kappa-B) and reduced NF-kappa-B activity, as well as systemic levels of reactive oxygen species, plasminogen activator inhibitor-1 (PAI-1), and CRP (13). Accordingly, the human promoter polymorphism 174 G/C of the IL-6 gene has been associated with an increased risk of cardiovascular disease and hypertension (14).
Dyslipidemic abnormalities. Diabetics usually show decreased high-density lipoprotein (HDL) cholesterol levels and increased low-density lipoprotein (LDL) cholesterol and triglyceride levels. Diabetes is most often associated with smaller, denser LDL that is more susceptible to oxidation and thus has greater atherogenicity (5). Even if diabetic patients with good glycemic control may have LDL levels similar to or slightly lower than those of nondiabetic individuals, fasting and postprandial levels of triglyceride-rich lipoproteins, especially very-low-density lipoprotein, are higher and those of HDL are lower than those of nondiabetics (15). In addition, glycation of LDL and other lipoproteins is quite common in diabetes, thus making the lipoproteins of diabetic patients more susceptible to oxidation and more atherogenic. Finally, several clinical studies have shown the benefit of lipid-lowering strategies in diabetics, even in cases of modestly increased or normal cholesterol levels, and these benefits are often greater than those found in nondiabetic patients (16).
Oxidative stress. Diabetics have decreased antioxidant activity and increased oxidative burden (17), and hyperglycemia increases the production of reactive oxygen species. These may shift the oxidant-antioxidant balance toward nonenzymatic oxidation of lipoproteins, thus contributing to atherogenic processes. It has also been hypothesized that reactive carbonyl groups may overwhelm the antioxidant mechanism of diabetic patients, leaving them unprotected to common oxidative stressors such as smoking.
Advanced glycation end products (AGEP). Diabetic hyperglycemia characteristically leads to nonenzymatic glycation of common macromolecules, as usually assessed through hemoglobin A1c (18). Nonenzymatic macromolecular glycation leads to the production of biomacromolecular aggregates known as AGEP. Glycated molecules as well as AGEP have been implicated in LDL modification, accumulation, and inflammatory activation. Oxidation of AGEP seems to play an important and perhaps primary role in initiating and amplifying lipid oxidation, thus contributing to the atherogenic profile of diabetics (19). Macromolecular glycated complexes increase arterial intimal noncellular growth and may involve proteins, nucleic acids, and lipid structures. The formation of complex and less degradable macromolecular aggregates and the impairment of scavenger cells cause quantitative imbalances in lipid levels and abnormal vascular clearance of proatherosclerotic lipids (20). Specific receptors for AGEP have been described, and their modulation has shown antiatherosclerotic effects in mice, whereas in normal rabbits, AGEP induces endothelial VCAM-1 and ICAM-1 expression and promotes atheroma formation (21).
Prothrombotic and antifibrinolytic abnormalities. Diabetes has several prothrombotic effects possibly involved in the initiation of atherosclerotic plaque, its progression, and complication. Diabetic rats show increased platelet adhesiveness and aggregability after common platelet-activating stimuli, such as adenosine and collagen (22). Platelet dysfunction is probably due in part to increased catabolism of arachidonic acid and increased production of thromboxanes. Diabetes is also associated with increased levels of fibrinogen, von Willebrand factor, factor VII, factor VIII, platelet factor 4, and PAI-1 levels. Peroxisome proliferator-activated receptor-alpha dysfunction or decreased activity seems to play a role in hyperfibrinogenemia (11), and fibrates may determine a reduction in fibrinogen levels through PPAR-alpha activation. Diabetic and hyperinsulinemic patients typically exhibit increased production of PAI-1 from visceral abdominal adipocytes. Finally, it is likely that elevated levels of tissue factor and PAI-1 and reduced levels of tissue-type plasminogen activator are associated with increased IL-6 production from adipocytes (according to the hypothesis of obesity as an inflammatory disease).
Dysmetabolic syndrome. The concept of dysmetabolic syndrome (metabolic syndrome X or insulin resistance syndrome) stands at the crossroads of hyperglycemia, overt type 2 diabetes mellitus, inflammation, and atherosclerosis (Fig. 2) (23). It is defined as a condition characterized at least in part by one or more of the following: glucose intolerance, relative hyperinsulinemia, peripheral insulin resistance, visceral or abdominal obesity, increased serum triglyceride concentrations, hyperuricemia, microalbuminuria, hypertension, and endothelial dysfunction. All of these entities have in common the same final effect of hyperinsulinemia, low-grade inflammation, and metabolic dysfunction. However, diabetes and the dysmetabolic syndrome seem to have both an independent detrimental effect on cardiovascular outcomes, as shown by studies that adjusted for insulin levels, diabetic status, and glycemic control (24). Hyperinsulinemia is also an independent and significant risk factor for coronary heart disease development in apparently healthy subjects. The overwhelming prevalence of this syndrome has prompted the proposal of the so-called "thrifty gene hypothesis," according to which in relatively early times, the ability to store fat may have conferred survival advantage, thus enriching ethnic groups of genes facilitating fat storage without negative feedback control (25). Several pathophysiologic mechanisms have been proposed, including abnormalities in leptin gene expression, decreased activity of PPAR-alpha, and primary inflammatory or infective processes. Nonetheless, the concept of the dysmetabolic syndrome has re-emphasized the metabolic and inflammatory role of adipocytes. In fact, among the several cellular elements involved in vascular inflammation, adipocytes merit great interest, as dysfunction in lipid metabolism is characteristic of diabetes, and as adipocytes have recently been shown able to produce IL-6 and TNF-alpha (26).
|
More recently, an abundance of clinical evidence has confirmed the pathogenetic role of inflammation in the onset of diabetes, showing that anti-inflammatory agents, such as statins (31), PPAR agonists (32), and other drugs, including angiotensin-converting enzyme inhibitors (33), may prevent or delay the onset of diabetes in high-risk subjects. Trials are currently underway to further validate this exciting hypothesis.
Role of nitric oxide (NO) in inflammation. Nitric oxide has several pleiotropic functions in the vascular milieu, beyond its well-known vasodilating capabilities. In vitro, NO selectively inhibits IL-1induced VCAM-1 expression on endothelial cells. This inhibition is paralleled by reduced monocyte adhesion to endothelial monolayers. Nitric oxide also decreases the endothelial expression of other leukocyte adhesion molecules (E-selectin and, to a lesser extent, ICAM-1) and secretable cytokines (IL-6 and -8). Molecular assays indicate that NO represses VCAM-1 gene transcription, in part by inhibiting NF-kappa-B (34). The expression of macrophage colony-stimulating factor in endothelial cells, an important regulating factor of macrophage-derived foam cells in atherosclerotic lesions, as induced by oxidized LDL or TNF-alpha, is significantly downregulated by NO through inhibition of NF-kappa-B (35). The precise mechanisms of NO-mediated NF-kappa-B inhibition are probably the induction and stabilization of I-kappa-B-alpha.
Role of cytokines and infectious agents. Experimental observations have demonstrated the central mechanistic relevance of several cytokines and chemokines networks in atherosclerotic processes (5). In this regard, diabetes, insulin resistance, and glucose intolerance all seem to have important modulating effects, with global proatherosclerotic activity. At least partly as a consequence of hyperglycemia and imbalances in insulin homeostasis, diabetes commonly induces abnormalities in the expression and activity of various cytokines and vasoactive peptides, such as TNF-alpha, IL-6, angiotensin II, and endothelin-1 (36). Serum levels of TNF-alpha, IL-6, and several adhesion molecules are significant predictors of future cardiovascular events in healthy subjects as well as diabetics, and in diabetic patients, increased VCAM-1 levels are independently associated with cardiovascular mortality (37). Interleukin-1 and TNF-alpha are produced by inflammatory cells in the atherosclerotic plaque and induce the release of IL-6 from several cell types, including smooth muscle cells. Furthermore, circulating IL-6 stimulates the hypothalamic-pituitary-adrenal axis, activation of which is associated with central obesity, hypertension, and insulin resistance (38). Thus, it is intriguing that overweight and obese individuals have increased serum levels of CRP, IL-6, TNF-alpha, and leptin. Actually, it has been recently shown that IL-6 levels decrease in obese women after significant weight loss (39).
Hyperglycemic patients demonstrate a significant production of AGEP, which are specifically recognized by inflammatory cells in atherosclerotic plaque, and induce the release of inflammatory cytokines. These mediators stimulate endothelial and smooth muscle proliferation, endothelial activation, and collagen overproduction, thus contributing to plaque growth and progression. C-reactive protein, a nonspecific marker of inflammation that also has a direct inflammatory activity in atherosclerosis, has been associated for a decade with adverse cardiovascular outcomes in patients with coronary artery disease as well as in healthy subjects (40,41). Interestingly, levels of CRP are higher in diabetics than in nondiabetics. Although elevated CRP levels in diabetics may depend on adipocyte activation or dysfunction, an additional hypothesis is that elevation of CRP might be associated with a low-grade inflammatory response driven by invading pathogens, as infections are very common in diabetics (42). The role of infectious agents in the pathogenesis of atherothrombosis might indeed be greater in diabetics than in nondiabetics, given the increased pathogen burden and decreased defense mechanisms of the former. Among the putative organisms, possible culprits include Chlamydia pneumoniae, Helicobacter pylori, cytomegalovirus, and herpes simplex virus (43). Diabetes may affect several steps of the host-infectious organism interaction. Poor glycemic control may lead to excessive AGEP production, vascular inflammation, and increased endothelial permeability, thus permitting bacterial or viral infiltration and infection of fatty streaks and, later, atheromatous plaques. Moreover, micro-organisms may contain molecules capable of eliciting host dysimmune responses through molecular mimicry (44). This may occur more often in an individual with greater global inflammatory burden and reactivity, as well as inflammatory dysfunction, such as in a diabetic.
Individuals infected with multiple pathogens such as herpes viruses, cytomegalovirus, Helicobacter pylori, and hepatitis A virus have higher CRP levels and the greatest relative risk of coronary artery disease. Thus, pathogens might contribute to the atherosclerotic process by promoting inflammatory responses, whereas infection may promote vessel wall injury initiated by oxidized lipids, smoking-derived oxidants, hypertensive shear, glycooxidized molecules, or a combination of these factors.
Among the putative cytokines involved in atherosclerosis, IL-6 appears to link local and systemic inflammation and the hepatic acute-phase response, acting as a messenger molecule, causing rises in CRP, serum amyloid A protein, and fibrinogen, as well as a decrease in albumin. Notwithstanding the role of inflammation in diabetic atherosclerosis, there is a strong and suggestive association between typical characteristics of insulin resistance and chronic low-grade inflammation. For instance, in nondiabetic subjects, levels of CRP and concentrations of the proinflammatory cytokines IL-6 and TNF-alpha are related to obesity. Moreover, CRP concentrations are related to insulin resistance, as calculated from the homeostasis assessment model, blood pressure, HDL, and triglycerides, and to markers of endothelial dysfunction. Similar conclusions have been reached, even by the assessment of noncoronary atherosclerosis, as CRP levels are partially associated with dysmetabolic syndrome markers and, together with them, predict carotid artery atherosclerosis (45). These data seem to indicate that adipose tissue is an important determinant of a low-level, chronic inflammatory state, as reflected by levels of IL-6, TNF-alpha, and CRP, thus supporting the hypothesis that such a low-level, chronic inflammatory state may induce insulin resistance and endothelial dysfunction and, therefore, link the latter phenomena with obesity and cardiovascular disease (46).
| Inflammatory markers,diabetes mellitus, and risk prediction |
|---|
|
|
|---|
In consideration of the overlap between inflammatory status and diabetic derangements, some authors have also looked at the association between these two conditions. Among 917 patients with acute coronary syndromes, diabetes and admission CRP levels were both independent predictors of long-term outcome (49). Other prospective studies considering CRP levels have confirmed the prognostic role of diabetic status (50,51). However, although the FRagmin during InStability in Coronary artery disease (FRISC) cohort has shown that diabetes plays a prognostic role independent of CRP levels (49), several other smaller studies have shown that diabetes acts as a significant predictor on univariate but not multivariate analysis. Interestingly, in some series, the presence of diabetes was associated, independent of the prognosis, with elevated CRP levels (50,52), whereas in other studies, the two parameters appeared to be independent (53,54). After percutaneous coronary interventions, the baseline CRP level was shown to be a predictor of events (including target vessel revascularization), independent of diabetic status (a condition associated with a significantly increased risk of recurrence after revascularization procedures) (5558).
Primary prevention studies among apparently healthy men (41) and women (59) have shown significant cardiac risk prediction by means of elevated CRP levels. In these data, no correlation between CRP levels and diabetes was described, and although diabetes was shown to be significantly associated with increased cardiac risk, CRP levels were still independent markers of prognosis. The MONItoring trends and determinants in CArdiovascular disease (MONICA) Augsburg cohort study was the first study to evaluate the prospective value of CRP levels in primary prevention (60). Many baseline clinical variables, including a history of diabetes, were found to be associated with higher CRP levels, and during follow-up, CRP remained an independent predictor of adverse cardiovascular events. The independent prognostic role of CRP has also been reported in patients with peripheral artery disease (61).
In conclusion, these data appear to demonstrate that CRP levels predict cardiac risk independent of diabetic status. However, the fact that diabetics tend to have higher CRP levels and that, in some clinical studies, diabetes still remained a significant independent predictor suggests that inflammation and metabolic derangements may represent two different aspects of the same process and thus both could be considered as independent predictors of cardiac risk. To the best of our knowledge, no study to date has thoroughly looked at the prognostic value of CRP between diabetics and nondiabetics separately. Therefore, these conclusions need further confirmation. In addition, the relative role of cytokines (IL-6, TNF-alpha), oxidized LDL, fibrinogen, and other markers of risk prediction require further investigation.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Feit, S. V. Manoukian, R. Ebrahimi, C. V. Pollack, E. M. Ohman, M. J. Attubato, R. Mehran, and G. W. Stone Safety and Efficacy of Bivalirudin Monotherapy in Patients With Diabetes Mellitus and Acute Coronary Syndromes: A Report From the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) Trial J. Am. Coll. Cardiol., April 29, 2008; 51(17): 1645 - 1652. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Otter, M. Winter, W. Doering, E. Standl, and O. Schnell C-Reactive Protein in Diabetic and Nondiabetic Patients With Acute Myocardial Infarction Diabetes Care, December 1, 2007; 30(12): 3080 - 3082. [Full Text] [PDF] |
||||
![]() |
M. Razmara, P. Hjemdahl, M. Yngen, C.-G. Ostenson, N. H. Wallen, and N. Li Food Intake Enhances Thromboxane Receptor-Mediated Platelet Activation in Type 2 Diabetic Patients but Not in Healthy Subjects Diabetes Care, January 1, 2007; 30(1): 138 - 140. [Full Text] [PDF] |
||||
![]() |
M. Lenzen, L. Ryden, J. Ohrvik, M. Bartnik, K. Malmberg, W. Scholte op Reimer, M. L. Simoons, and on behalf of the Euro Heart Survey Investigators Diabetes known or newly detected, but not impaired glucose regulation, has a negative influence on 1-year outcome in patients with coronary artery disease: a report from the Euro Heart Survey on diabetes and the heart Eur. Heart J., December 2, 2006; 27(24): 2969 - 2974. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. G.L. Biondi-Zoccai, M. Lotrionte, P. Agostoni, A. Abbate, M. Fusaro, F. Burzotta, L. Testa, I. Sheiban, and G. Sangiorgi A systematic review and meta-analysis on the hazards of discontinuing or not adhering to aspirin among 50 279 patients at risk for coronary artery disease Eur. Heart J., November 2, 2006; 27(22): 2667 - 2674. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. ALEXANDRAKI, C. PIPERI, C. KALOFOUTIS, J. SINGH, A. ALAVERAS, and A. KALOFOUTIS Inflammatory Process in Type 2 Diabetes: The Role of Cytokines Ann. N.Y. Acad. Sci., November 1, 2006; 1084(1): 89 - 117. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Angiolillo, A. Fernandez-Ortiz, E. Bernardo, C. Ramirez, M. Sabate, P. Jimenez-Quevedo, R. Hernandez, R. Moreno, J. Escaned, F. Alfonso, et al. Clopidogrel Withdrawal Is Associated With Proinflammatory and Prothrombotic Effects in Patients With Diabetes and Coronary Artery Disease Diabetes, March 1, 2006; 55(3): 780 - 784. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Roffi Early invasive strategy in the diabetic patient with non-ST-segment elevation acute coronary syndromes Eur. Heart J. Suppl., October 1, 2005; 7(suppl_K): K19 - K22. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Abbate and G. G.L. Biondi-Zoccai The difficult task of glycaemic control in diabetics with acute coronary syndromes: finding the way to normoglycaemia avoiding both hyper- and hypoglycaemia Eur. Heart J., July 1, 2005; 26(13): 1245 - 1248. [Full Text] [PDF] |
||||
![]() |
H. S. Gurm, I. J. Sarembock, D. J. Kereiakes, J. J. Young, R. A. Harrington, N. Kleiman, F. Feit, K. Wolski, J. A. Bittl, R. Wilcox, et al. Use of Bivalirudin During Percutaneous Coronary Intervention in Patients With Diabetes Mellitus: An Analysis From the Randomized Evaluation in Percutaneous Coronary Intervention Linking Angiomax to Reduced Clinical Events (REPLACE)-2 Trial J. Am. Coll. Cardiol., June 21, 2005; 45(12): 1932 - 1938. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Hung, B. M. McQuillan, C. M. L. Chapman, P. L. Thompson, and J. P. Beilby Elevated Interleukin-18 Levels Are Associated With the Metabolic Syndrome Independent of Obesity and Insulin Resistance Arterioscler. Thromb. Vasc. Biol., June 1, 2005; 25(6): 1268 - 1273. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Prasad, G. W. Stone, T. D. Stuckey, C. O. Costantini, P. J. Zimetbaum, M. McLaughlin, R. Mehran, E. Garcia, J. E. Tcheng, D. A. Cox, et al. Impact of diabetes mellitus on myocardial perfusion after primary angioplasty in patients with acute myocardial infarction J. Am. Coll. Cardiol., February 15, 2005; 45(4): 508 - 514. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. T. Fung, M. Schulze, J. E. Manson, W. C. Willett, and F. B. Hu Dietary Patterns, Meat Intake, and the Risk of Type 2 Diabetes in Women Arch Intern Med, November 8, 2004; 164(20): 2235 - 2240. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. M. Biasucci and G. G.L. Biondi-Zoccai Preprocedural C-Reactive Protein for Risk Prediction Before Percutaneous Coronary Intervention (PCI): A European Perspective Clin. Chem., September 1, 2004; 50(9): 1492 - 1494. [Full Text] [PDF] |
||||
![]() |
C. L. Price, P. S. Sharp, M. E. North, S. J. Rainbow, and S. C. Knight Advanced Glycation End Products Modulate the Maturation and Function of Peripheral Blood Dendritic Cells Diabetes, June 1, 2004; 53(6): 1452 - 1458. [Abstract] [Full Text] [PDF] |
||||
![]() |
P L Sanchez, J L Morinigo, P Pabon, F Martin, I Piedra, I F Palacios, and C Martin-Luengo Prognostic relations between inflammatory markers and mortality in diabetic patients with non-ST elevation acute coronary syndrome Heart, March 1, 2004; 90(3): 264 - 269. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Roffi and E. J. Topol Percutaneous coronary intervention in diabetic patients with non-ST-segment elevation acute coronary syndromes Eur. Heart J., February 1, 2004; 25(3): 190 - 198. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Luciani, G. Nasso, M. Gaudino, A. Abbate, F. Glieca, F. Alessandrini, F. Girola, F. Santarelli, and G. Possati Coronary artery bypass grafting in Type II diabetic patients: a comparison between insulin-dependent and non-insulin-dependent patients at short- and mid-term follow-up Ann. Thorac. Surg., October 1, 2003; 76(4): 1149 - 1154. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||